European Competition Law Annual 2010: Merger Control in European and Global Perspective 9781472561299, 9781849462006

“Every year, top-level market regulators, academics and legal and economic practitioners contribute to the Annual Compet

230 67 10MB

English Pages [448] Year 2013

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

European Competition Law Annual 2010: Merger Control in European and Global Perspective
 9781472561299, 9781849462006

Citation preview

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

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

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

vii  

15th ANNUAL EU C OMP ET IT ION L AW AND POLIC Y WOR KS HOP : MERGER CONTROL IN EUROPEAN AND GLOBAL PERSPECTIVE

Organizers Philip Lowe and Mel Marquis Chairs John Fingleton Philip Lowe Giorgio Monti

Office of Fair Trading, London European Commission, Brussels European University Institute, Florence Project participants

Rafael Allendesalazar Andreas Bardong Simon Bishop Josep Borrell Fontelles John Boyce Rachel Brandenburger Jochen Burrichter Cristina Caffarra Theofanis Christoforou Maher Dabbah Thomas Deisenhofer Götz Drauz Kirsten Edwards Adam Fanaki Amelia Fletcher Ian Forrester Nicholas Forwood Cal Goldman F. Enrique González-Díaz Klaus Gugler

Martínez Lage, Allendesalazar & Brokelmann, Madrid Bundeskartellamt, Bonn RBB Economics, Brussels European University Institute, Florence Slaughter and May, Brussels U.S. Department of Justice, Washington, D.C. Hengeler Müller, Düsseldorf CRA International, London European Commission, Brussels Queen Mary School of Law European Commission, Brussels Shearman and Sterling, Brussels LECG, London (now Office of Fair Trading, London) Davies Ward Phillips & Vineberg, Toronto Office of Fair Trading, London White and Case, Brussels General Court, Luxembourg Blake, Cassels and Graydon, Toronto Cleary Gottlieb Steen Hamilton, Brussels WU Vienna University of Economics and Business, Vienna

xii  List of Participants Barry Hawk Scott Hemphill Seonghoon Jeon Etsuko Kameoka William Kovacic Nicholas Levy Johannes Lübking Mel Marquis Abel Mateus Andreas Mundt Alison Oldale Lars-Hendrik Röller Howard Shelanski Tadashi Shiraishi Mario Siragusa Irwin Stelzer Jim Venit Sven Völcker Vanessa Yanhua Zhang Xinzhu Zhang

Fordham University School of Law, New York Columbia University School of Law, New York (now Office of the New York State Attorney General) Korean Fair Trade Commission and Sogang University, Seoul Van Bael & Bellis, Brussels U.S. Federal Trade Commission, Washington, D.C. (now George Washington University School of Law, Washington, D.C.) Cleary Gottlieb Steen Hamilton, Brussels and London European Commission, Brussels European University Institute, Florence and University of Verona New University of Lisbon and University College London Bundeskartellamt, Bonn Competition Commission, London European School of Management and Technology, Berlin (on leave) U.S. Federal Trade Commission, Washington, D.C. University of Tokyo Graduate Schools for Law and Politics, Tokyo Cleary Gottlieb Steen Hamilton, Rome Hudson Institute, Washington & Nuffield College, Oxford Skadden, Brussels WilmerHale, Brussels Renmin University of China, Beijing and Global Economics Group, Beijing and New York Jiangxi University of Finance and Economics, Nanchang and Chinese Academy of Social Sciences, Beijing

Mel Marquis*

Regulating Mergers: Substantive and Procedural Issues, Judicial Review, International Convergence and Best Practices

This volume contains a collection of contributions prepared for the 15th edition of the Annual EU Competition Law and Policy Workshop, held on 12–13 November 2010 at the European University Institute in Florence. The 2010 Workshop was devoted to a variety of issues related to merger control. The subject is not dealt with comprehensively but a broad range of issues of relevance to competition experts in Europe and elsewhere have been covered. The Workshop was both the inaugural event for my co-organizer and co-editor, Philip Lowe, and the final Workshop organized under the auspices of the Robert Schuman Center for Advanced Studies. Organizationally the Workshop was transferred in 2011, along with your correspondent, to the EUI’s Law Department, where our activities continue. Since 2011 the Workshop has considered issues related to the integration of public and private enforcement of competition law and to the interplay between competition, public policies and regulation.

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. 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 roundtable debates that explore specific competition-related issues in an informal and non-commercial environment. One of our primary objectives is to stimulate critical reflection on the part of both the Workshop participants and the broader public. * Part-time Professor of Law, EUI, Florence; Professore a contratto, Università degli studi di Verona; and Visiting Professor, LUMSA, Rome. On specific points made in this introductory chapter I benefited from the kind feedback of Massimo Motta. In addition to my esteemed collaborator Philip Lowe, many friends and colleagues have supported the Workshop, including among others (in particular our sponsors) Gerald Barling, Joachim Bornkamm, John Cooke, Claus Ehlermann, John Fingleton, Mattia Guidi, Barry Hawk, Valentine Korah, Bill Kovacic, Rosanna Lewis, Petros Mavroidis, Giorgio Monti, Kanako Nagao, Aindrias Ó Caoimh, Savvas Papasavvas, Tom Rosch, Daniela Romagnoli and Diane Wood. Collaborating with me on the preparation of the transcripts included in this volume was Emma Linklater, a PhD researcher in the Law Department of the EUI.

xiv  Merger Control in European and Global Perspective The 15th edition of the Workshop2 was entitled: Merger Control in European and Global Perspective. The objective was to examine and better understand a panorama of topical issues in the field of merger control across multiple jurisdictions. As seen in the title of this introduction, the themes encompassed substantive and procedural issues, judicial review, international convergence and best practices. To some extent the title is both overinclusive and underinclusive. For example, few contributions devoted close attention to non-horizontal mergers,3 or to selection, design and implementation of remedies in different merger scenarios, although other aspects of remedies such as cross-border coordination and cooperation were addressed. In terms of procedure, more discussion could have concentrated on due process or procedural rights of third parties. And while certain younger and/or developing jurisdictions are represented here – with chapters covering, e.g., China, Brazil and the Middle East – more could have been said about the specific challenges faced by developing countries in terms of, among other things, capacity/ resource constraints and regime effectiveness.4 The title is underinclusive in the sense that “international convergence” does not necessarily convey the ideas of inter-agency cooperation and comity, which figured prominently in the discussions and contributions. Another issue which was discussed but which does not appear in the title is the obviously important question of how to develop a robust framework to evaluate how a given merger control system measures up. However, on the whole the title provides a rough impression of the range of subjects addressed in this volume. A more schematic breakdown of the proceedings appears below (see “Structure of the Workshop”). As usual, a rather belabored chapter-by-chapter review is provided below. I won’t begin with any elaborate attempt to neatly encapsulate all of the content of these chapters. The following slanted preface will have to do. Since the adoption of the EC Merger Regulation in 1989, global and multijurisdictional merger control progressed rapidly through its “infancy”, developing in the 1990s increasingly sophisticated and data-intensive methods 2 For information concerning previous editions of the Workshop and the corresponding European Competition Law Annual series, see the website of the Workshop: http://www.eui.eu/ DepartmentsAndCentres/Law/ResearchAndTeaching/ResearchThemes/CompetitionWorkshop/ Index.aspx. The various books in the series are advertised on Hart’s website: see http://www. hartpublishingusa.com/books/series.asp?sc=European+Competition+Law+Annual&st=European+Co mpetition+Law+Annual. 3 A few concrete non-horizontal merger cases are discussed by Xinzhu Zhang and Vanessa Yanhua Zhang, and a few other papers touch on the subject (see, e.g., the papers by Irwin Stelzer and Seonghoon Jeon). Each of these papers is reviewed in this chapter. See also the oral remarks of Johannes Lübking, p. 5 (referring to a few non-horizontal cases handled by the European Commission once it had adopted its Non-Horizontal Merger Guidelines). 4 For recent discussion of these challenges (which, despite many positive developments in fora such as the ICN, still in some ways presents a gloomy picture), see OECD Policy Roundtable, “Cross-Border Merger Control: Challenges for Developing and Emerging Economies” (13 February 2012), DAF/ COMP/GF(2011)13, http://www.oecd.org/competition/mergers/50114086.pdf; UNCTAD Secretariat, “Cross-border anticompetitive practices: The challenges for developing countries and economies in transition”, TD/B/C.I/CLP/16 (19 April 2012), http://unctad.org/meetings/en/SessionalDocuments/ ciclpd16_en.pdf, pp. 13-20.

Introduction 

xv  

drawing on the theory of industrial organization.5 The prospect of devoting substantial resources to the complex task of vetting mergers (often with substantially heterogeneous characteristics) within strict time frames was not so intimidating as to staunch a global merger control “movement” that has produced a lush thicket of regulatory regimes, a movement that may be linked to, inter alia, processes of economic globalism and to some limited extent institutional isomorphism. The global turn toward multipolarity, with the increasing economic and political significance of emerging economies, has reacted to and reinforced these processes. Today’s vast matrix of merger control systems ideally protects consumers from the point of view of price, choice, quality and innovation (and protects sellers against exploitation of buyer power) but it also presents significant coordination and timing problems and tends to raise regulatory costs. On the whole, it is manageable but messy.6 Taking for granted the impossibility of a global one-stop shop for large transactions affecting multiple jurisdictions, public authorities and private actors navigating the “thicket” find themselves in a constant search for other solutions (with varying results) along various dimensions: substantive and procedural, internal and external, bilateral/regional and global, formal and informal, and so on. Against this background, a likely impression of the proceedings contained here is one of juxtaposition: on the one hand, there are trends toward consolidation, consensus and convergence, with strong cooperative efforts paying off; yet at the same time global merger control is in a period of transition and increasing variability in some respects, as leading jurisdictions press forward with increasingly demanding investigative techniques. It is doubtful that the most refined techniques can be and/or will be implemented across jurisdictions, and to that extent (and for other institutional and cultural reasons) we will surely continue to see global diversity and multi-speed (antitrust and) merger control.7 But this may be an endemic feature of regulatory convergence; 5 Oliver Budzinski pairs the IO/economic effects turn in merger control (and antitrust more generally) with the constant neglect of modern institutional economics, an imbalance that has resulted in underenforcement and has moreover made it difficult if not impossible to capitalize on what industrial economics has to offer. See Budzinski, “An Institutional Analysis of the Enforcement Problems in Merger Control” (June 2010), http://static.sdu.dk/mediafiles/F/8/A/%7BF8A24E9259AE-4457-8A67-AC2A805592A1%7Dbudzinski101.pdf. 6 Despite a variety of efforts to smooth out the creases, potentially costly differences across jurisdictions remain. See, e.g., Larry Fullerton and Megan Alvarez, “Convergence in International Merger Control”, Antitrust 20 (Spring 2012) (noting points of divergence attributable in their view to the 2010 revision of the US Horizontal Merger Guidelines); Jonathan Galloway, “Convergence in International Merger Control”, 5(2) Competition Law Review 179 (2009). Cf. also Dane Holbrook, “International Merger Control Convergence: Resolving Multijurisdictional Review Problems”, 7 UCLA Journal of International Law and Foreign Affairs 345 (2002). 7 It seems redundant to say, but references to convergence should not be construed as a normative call for the relentless pursuit of uniformity (“blind” convergence). One-size-fits-all is rarely accurate from a descriptive point of view, and presumptively it is apt to be risky. What can be commended, in brief, is “informed convergence” toward better common solutions in the common interest, taking account of their relative costs and of any counter-indicative benefits of experimentation and creative tension. (Cf. the oral remarks of Kirsten Edwards, p. 197, and Bill Kovacic, p. 207.) Informed convergence in this perspective leaves adequate space for the informed divergence (a term used by David Lewis, John Fingleton and others) to which our past Workshop editions have referred – divergence based on affirmative choices such as protecting consumers in local markets, developmental differences, cultural

xvi  Merger Control in European and Global Perspective the diversity within the EU itself is testimony to the fact that convergence is not an outcome, nor an inevitable, unidirectional engrenage but a multifaceted process with no fixed destination.

Structure of the Workshop In the present volume, the debates held at the 2010 Workshop are broken down into introductory remarks by Irwin Stelzer, followed by five main sessions, or “Panels”. The following broad topics were covered: (1) Merger enforcement across jurisdictions: Substantive issues (market definition, unilateral/coordinated effects, innovation, efficiencies); (2) Merger policy assessment and judicial review; (3) Merger enforcement across jurisdictions: Procedural issues; (4) International convergence: Substantive and procedural issues and the scope for comity; and (5) Merger control and best practices. As indicated above, the remainder of this chapter provides an overview of the Workshop’s written contributions. As per my usual caveat, this is no substitute for the contents of the other chapters, let alone for the book as a whole. It is just a selective discussion of some of the issues raised by the Workshop participants, and a guide to some of the connections between the contributions. As a rare concession to brevity I do not summarize in this chapter the proceedings reported in the transcripts of the oral discussions. differences and the like; and allows for a dynamic process involving pioneers, a bit of “peer review” and opt-in or abstention. Different observers will regard global convergence/divergence in merger control in a variety of ways. I would just note here the observations of Thomas Cheng, who writes: “Given the possibility of an outright rejection of a proposed transaction, merger review should be considered as full entity regulation, which means that there is a strong case for convergence from the perspective of regulatory accommodation. The need to avoid conflicting merger review decisions, however, should not be overstated. Here, it is again important to distinguish between genuinely globalized markets and local markets. If a merger involves global markets, and if the competitive effects of the transaction are felt the same way in different jurisdictions, there is a strong argument for a harmonized approach to merger analysis and perhaps a single decision on the transaction. Assuming that these jurisdictions pursue similar goals under their competition law [an assumption which, as Cheng points out in his article, requires significant qualification] and their competition authorities are analyzing the same transaction in the same market(s) involving the same parties, conflicting decisions by different authorities are indefensible. This argument loses its force when the markets are not globalized and the authorities are analyzing the effects of the merger on their respective national markets. […] While it may be frustrating for the proponents of a transaction to abort a merger simply because one jurisdiction objects to it, if an informed and well-supported economic analysis of the transaction demonstrates irremediable anticompetitive effects in a local market, the desire for uniformity cannot deny the local authority’s prerogative to protect its own market and consumers. Even for a full entity regulation, the argument for harmonization is not absolute.” Cheng, “Convergence and Its Discontents: A Reconsideration of the Merits of Convergence of Global Competition Law”, 12 Chicago Journal of International Law 433, 463-464 (2012).

Introduction 

xvii  

Irwin Stelzer kicks things off with his paper, “Merger Policy and Schumpeter’s Creative and Destructive Gale”. Joseph Schumpeter was a thinker whose role has never quite been fully debated or defined in the specific context of EU competition law8 – even if his name has been invoked with some regularity.9 In general, the deregulatory policy prescriptions that seem to follow from Schumpeter’s faith in long-term productivity and wealth creation may be less likely than ever to gain traction in the current climate of skepticism toward free markets. Competition enforcers have long been resistant to hands-off approaches to high-tech industries on the basis of Schumpeterian arguments, as one has seen in particular in cases involving the alleged abuse of dominance. But what of merger control in particular? Although he doesn’t formulate the issue quite in the following terms, Stelzer suggests that, in the field of non-horizontal mergers, doubt should be resolved against intervention, at least in markets with Schumpeterian competition. Such an approach boils down to an ‘inverse precautionary principle’ for certain types of mergers. It is easy to understand why Stelzer is concerned with achieving a proper threshold of intervention – if Schumpeter’s assumptions were right, entrepreneurs will do all they can to generate dynamic efficiencies (fixed cost savings, fieldsweeping new products, etc.) of a magnitude likely to dwarf the static efficiencies a merger may produce, but the analytical framework established by law and policy may be ill-equipped for the consideration of dynamic efficiencies. That is to say, once it is determined, that a merger would significantly impede “effective competition”, it may be difficult to overcome that finding with difficultto-quantify and/or long-delay gains in dynamic efficiency.10 Stelzer therefore suggests that the competitive dynamics and possible efficiencies should be considered at an earlier stage, which might entail a determination that a merger would not impede competition in the first place, in particular because post-merger market “dominance” might prove to be transitory. From a dynamic perspective, it follows that market definition and market shares should be regarded with great caution and should be corroborated as far as possible, in particular where there is product differentiation. (Stelzer does not go so far as to suggest abandoning those traditional tools altogether; nor, for that matter, do many of the proponents of newer techniques.) Of course, some jurisdictions already follow his advice, and 8 Some discussions point in this direction, however. See, e.g., Jonathan Galloway, “Driving Innovation: A Case for Targeted Competition Policy in Dynamic Markets”, 34 World Competition 73 (2011) (counseling against a Schumpeterian approach for antitrust treatment (in the EU) of innovative industries; arguably, this reflects the views of many if not most Europeans – all the more so in industries characterized by network effects and market tipping). 9 Of course, there is literature aplenty on Schumpeter and innovation in a more general sense, with contributions typically pitting Schumpeter against Ken Arrow. For a recent discussion of the affinities between their approaches, see Carl Shapiro, “Competition and Innovation: Did Arrow Hit the Bull’s Eye?”, in Josh Lerner and Scott Stern, eds., The Rate and Direction of Inventive Activity Revisited, University of Chicago Press (2012), pp. 361 et seq., with references. 10 In more detail, see Lars-Hendrik Röller, “Efficiencies in EU Merger Control: Do They Matter?”, this volume, pp. 61 et seq. (discussed below). See also OECD, Dynamic Efficiencies in Merger Analysis, DAF/COMP(2007)41 (15 May 2008), http://www.oecd.org/competition/mergers/40623561. pdf.

xviii  Merger Control in European and Global Perspective regard data such as large market shares as indicators which need to be checked holistically against other factors, particularly if, say, the market of reference exhibits rapid growth or high volatility, or possibly if it is so new that its future direction of evolution is anybody’s guess.11 Other (younger) jurisdictions may not yet have the same self-assurance.

1. Merger enforcement across jurisdictions: Substantive issues (market definition, unilateral/coordinated effects, innovation, efficiencies)12 Contributions by Edwards, Hemphill, Hawk and Röller Despite the fanfare that accompanied the 2010 revision to the U.S. Merger Guidelines13 (and the ideas that direct methods of measuring a merger’s market power consequences can sometimes be employed, that SSNIP is often misused and that market definition and concentration may not be the initial focus of the assessment), most of the traditional tools remain an important part of the competitive analysis in merger control, both for the agencies (which may find market definition very useful or may be faced with evidence that limits options14) and in court (where, for example, the US statute refers to a “line of commerce”15). A notable example of the lasting importance of “markets” is the ultimately abandoned AT&T/T-Mobile 4-to-3 merger, where the DOJ pleaded before the 11 On the latter point, see Howard Shelanski’s oral remarks, this volume, pp. 19-20. In the hardest cases, agencies may not have to face a binary choice: time-limited and proportionate behavioral remedies can in some cases provide supplementary options. 12 Papers contained in other sections, especially section 4, cover substantive analysis as well. This is the case for example, in the papers by Seonghoon Jeon and by Xinzhu Zhang and Vanessa Yanhua Zhang (discussed later in this chapter – see session 4). 13 On the background to the 2010 incarnation of the Guidelines, see among others D. Daniel Sokol, “Antitrust, Institutions, and Merger Control”, 17 George Mason Law Review 1055, 1105-1109 (2010); and DOJ/FTC, Commentary on the Horizontal Merger Guidelines (March 2006), http://www.justice. gov/atr/public/guidelines/215247.htm. The revised US Guidelines were of course not the only ones issued in 2010. For example, the UK’s OFT and CC also adopted merger guidelines that year. (For a visual tour of “illustrative price rise” (IPR) analysis in the UK (first used in 2005), see Amelia Fletcher’s slides at http://www.competitioneconomics.org/dyn/files/basic_items/364-file/FletcherWalters-UPP%20up%20and%20away.pdf.) Indeed, since 2009, other (revised or debut) merger guidelines – covering but not limited to horizontal effects – have been adopted in Canada, China, France, Germany, India, Ireland and Japan, among other jurisdictions. The ICN’s Recommended Practices for Merger Analysis, updated most recently in April 2010 (i.e., about eight months before the US Guidelines were published), endorse the traditional tools (market shares, market definition with product differentiation, etc.) but caution against too much reliance on them and also stress the need to consider market-specific, effects-oriented evidence. See http://www.internationalcompetitionnetwork. org/uploads/library/doc316.pdf. 14 See and compare the oral remarks of Alison Oldale (p. 10), Simon Bishop (p. 14) and Howard Shelanski (pp. 16-19). 15 See Irwin Stelzer, oral remarks, p.2.

Introduction 

xix  

federal district court both relevant markets and concentration indicators.16 But whether the initial focus in a given case is market definition or upward price pressure (the latter tool being designed to measure incentives to raise price), data that give indications as to competitive constraints, including substitutability where potential substitutes are differentiated, are likely to be important.17 Kirsten Edwards addresses the subject of (hypothetical) customer switching in her paper, “Estimating Diversion Ratios: Some Thoughts on Customer Survey Design”. Customer surveys are not the only method of collecting data to derive diversion ratios (since one might refer, for example, to exogenous shocks and actual switching, win/loss reports, etc.), but given the frequency with which surveys are used and their potential probative value, Edwards conducts her own survey of surveys with a view to identifying methodological errors and sound design principles. The general message of the paper is that surveys are a useful but complex tool whose results can potentially be misleading unless an adequate investment is made in ex ante design and ex post checks. Reviewing a number of merger investigations by the European Commission and the UK’s Competition Commission, Edwards explains how certain types of errors (sampling error, coverage error, non-response error, measurement error) can creep into customer surveys if proper care is not taken. Along the way she offers suggestions to design and execute surveys that minimize the risk of such errors. Further research in this area could be fruitful considering that, as Edwards points out, no consensus has yet developed as to the best approaches to design techniques and error control. Scott Hemphill comments on the 2010 U.S. Guidelines in “Higher Profits as a Merger Defense: Innovation, Appropriability and the Horizontal Merger Guidelines”. It is a very specific critique, as Hemphill focuses particularly a single passage of the Guidelines: When evaluating the effects of a merger on innovation, the Agencies consider the ability of the merged firm to conduct research or development more effectively. Such efficiencies may spur innovation but not affect short-term pricing. The Agencies also consider the ability of the merged firm to appropriate a greater fraction of the benefits resulting from its innovations. Licensing and intellectual property conditions may be important to this enquiry, as they affect the ability of a firm to appropriate the benefits of its innovation. Research and development cost savings may be substantial and yet not be cognizable efficiencies because they are difficult to verify or result from anticompetitive reductions in innovative activities.18 16 See the DOJ’s Complaint at http://www.justice.gov/atr/casesf274600/274613.htm. The allegedly affected markets were defined according to distinctions between residential and business customers, and between national and local markets. Regarding concentration factors, the DOJ applied the more relaxed presumption of adverse effects described in Section 5.3 of the 2010 Merger Guidelines, which increased the relevant threshold from a post-merger HHI of 1,800 to a post-merger HHI of 2,500, assuming a delta of at least 200. The enhanced threshold was easily surpassed in this case since, for example, the HHI score for the national business user market would have exceeded 3,400 if the transaction had proceeded as proposed. 17 Cf. Carl Shapiro, “The 2010 Horizontal Merger Guidelines: From Hedgehog to Fox in Forty Years”, 77 Antitrust Law Journal 701, 708 (2010) (“same piece of evidence may be relevant to competitive effects and to market definition”). 18 US DOJ and FTC, Horizontal Merger Guidelines (19 August 2010), http://www.justice.gov/atr/ public/guidelines/hmg-2010.pdf, § 10 (“Efficiencies”), p. 31 (emphasis added).

xx  Merger Control in European and Global Perspective From this inelegant paragraph Hemphill deduces that the US Agencies purport to adopt a new “defense” that may be claimed if it has been found under Section 7 of the Clayton Act that a given merger substantially lessens competition. He calls this an “increased appropriation defense”, i.e., a defense based on actual or likely higher margins. To put flesh on the discussion Hemphill proceeds to introduce the US case of FTC v Lundbeck,19 where a US District Court denied the FTC’s application for an order to unwind an alleged “acquisition to monopoly” by Lundbeck (then Ovation) of NeoProfen, a therapeutic drug due for FDA approval which, like Lundbeck’s own drug Indocin IV, was designed to treat a rare heart condition afflicting infants.20 After Hemphill wrote his paper, the Eighth Circuit confirmed the District Court’s judgment.21 Although Chairman Leibowitz and Commissioners Brill and Ramirez found the outcome to be “profoundly wrong”, they declined to apply for certiorari.22 According to the District Court, the FTC’s allegation that the acquisition was contrary to Section 7 depended on whether the target drug NeoProfen belonged to the same market as Indocin IV. This required the court to arbitrate between, among other evidence, the testimony of specialized physicians – neonatologists – who for evident reasons tended to be price-insensitive (which the judge took to mean that there would be no effective price competition between the products even if owned by different firms) and, on the other hand, an eye-popping price increase of 1,300% for Indocin IV – two days after Lundbeck obtained the rights to NeoProfen.23 Hemphill shows the defects of the court’s analysis in order to set up a discussion of how, if hypothetically the decision had gone the other way, the parties might still have saved their merger under a (second) hypothetical FTC v. Lundbeck, Inc., Nos. 08-6379, 08-6381, 2010 WL 3810015 (D. Minn., 31 August 2010). In addition to divestiture and rescission, the FTC’s complaint, brought under Section 7 of the Clayton Act (asset acquisition) and Sections 5(a) and 13(b)(2) of the Federal Trade Commission Act (application for injunction against unfair method of competition in the form of willful maintenance of monopoly power) also sought disgorgement of Lundbeck’s allegedly ill-gotten gains. Although postconsummation challenges to mergers and acquisitions are rather infrequent (but not terribly so), it is well established that the US Agencies can bring such actions. See, e.g., U.S. v. E.I. DuPont de Nemours & Co., 353 U.S. 586 (1957) (DOJ challenge of DuPont’s acquisition of an interest in General Motors around 30 years later). For recent discussion of the FTC’s practice, see J. Thomas Rosch, “Consummated Merger Challenges – The Past is Never Dead”, speech, Washington, DC, 29 March 2012, http://www. ftc.gov/speeches/rosch/120329springmeetingspeech.pdf (referring to Lundbeck at pp. 4-5). 21 FTC v. Lundbeck, Inc., 650 F.3d 1236 (8th Cir. 2011). Since the determination of the relevant market was essentially a matter for the trier or fact, i.e., the District Court, the FTC was obliged, and failed, to demonstrate to the Circuit Court that the District Court’s finding on market definition was clearly erroneous. The standard of review may have been a crucial factor in the Eighth Circuit’s judgment, as there was ample room to doubt the District Court’s counterintuitive findings. For a critical discussion of the appellate judgment, see Herbert Hovenkamp, “Mergers, Market Dominance and the Lundbeck Case” (December 2011), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1968151. 22 See their Statement of 20 January 2012, http://www.ftc.gov/os/closings/ publicltrs/120120lundbeck-jdl-brill-ramirez.pdf. Disagreeing with his colleagues, Commissioner Rosch listed several reasons why seeking review by the Supreme Court would have been appropriate in this case. See Separate Statement of 20 January 2012, http://www.ftc.gov/os/closings/ publicltrs/120120lundbeck-rosch.pdf. 23 Merck, the previous owner of Indocin IV, had charged about $78 per treatment. Following the contested acquisition, Lundbeck raised the price of Indocin IV to $1614. It set prices for NeoProfen in a range of $1450 to $1522 per treatment. Prior to regulatory approval, it was anticipated that the price of NeoProfen would be around $450 to $500 per treatment, but such forecasts had been made when the price of Indocin IV was still relatively very low. 19 20

Introduction 

xxi  

assumption that the 2010 Guidelines could have (retroactively) provided the basis for claiming an “increased appropriation defense”. Hemphill does not argue that the transaction could have been justified under the Guidelines. His aim, rather, is to show how, given the ambiguity of the paragraph quoted above concerning appropriability, it is difficult to know for sure whether the defense could have applied. The interpretive difficulty is that, from a practical point of view, increased appropriation will in most cases imply higher post-merger prices. That would normally spell trouble for a proposed merger. Having incorporated the language of appropriation, the Guidelines may have failed to follow through, leaving only a mirage of a defense.24 This is one of Hemphill’s “puzzles”.25 In most cases, if the defense is to bring added value, it must be available to justify static consumer harm on the basis of enhanced incentives to innovate. But the Guidelines continue, unsurprisingly, to stress that merger analysis is largely concerned with price and output effects. Indeed, in an earlier passage of the same section, the Guidelines cite merger effects that “prevent” price increases as one form of cognizable efficiency.26 The apparent cognitive dissonance arising from these different passages will have to be sorted out in practice until a more coherent synthesis can be incorporated in a future version of the Guidelines.27 One can’t help but note that in certain cases the US courts and the FTC have not been reading from the same page, as Lundbeck vividly suggests. Older habits of mind (such as the idea that even when products are differentiated market boundaries determine where competitive pressures switch on and off) may make the courts reluctant to let go of their familiar orthodoxies, even in a case where an 24 The de jure existence and de facto unavailability of efficiency claims in antitrust law constitute an enduring paradox. The invariable problem relates to what Dan Crane calls an “asymmetry” principle: evidence suggesting post-merger price increased will be examined with piqued interest; evidence of efficiencies with furrowed brow. See Crane, “Rethinking Merger Efficiencies”, 110 Michigan Law Review 347, 356-358 (2011). 25 Another puzzle he described relates to the tension between the District Court’s approach in Lundbeck and a more standard mode of analysis in monopolization law whereby steps taken by a monopolist to eliminate a potential competitive threat (as in the US Microsoft browser case), irrespective of whether the source of the threat were found at this early stage to belong to the same relevant market as the dominant product, might constitute an infringement under Section 2 of the Sherman Act. See Hemphill’s oral remarks, pp. 21-22. 26 See Guidelines, cited above note 18, at 30-31: “The Agencies will not challenge a merger if cognizable efficiencies are of a character and magnitude such that the merger is not likely to be anticompetitive in any relevant market. […] [T]he Agencies consider whether cognizable efficiencies likely would be sufficient to reverse the merger’s potential to harm customers in the relevant market, e.g., by preventing price increases in that market. In conducting this analysis, the Agencies will not simply compare the magnitude of the cognizable efficiencies with the magnitude of the likely harm to competition absent the efficiencies. The greater the potential adverse competitive effect of a merger, the greater must be the cognizable efficiencies, and the more they must be passed through to customers, for the Agencies to conclude that the merger will not have an anticompetitive effect in the relevant market.” (footnotes omitted) 27 Practically any set of Guidelines, in the field of merger control or otherwise, is by nature incomplete and must be further tested and elaborated through implementation. With regard to the 2010 Merger Guidelines, see William Kovacic, “Dominance, duopoly and oligopoly: the United States and the development of global competition policy”, 13(11) Global Competition Review 39, 41 (December 2010).

xxii  Merger Control in European and Global Perspective acquisition eliminates the only possible competitive threat (a “2 to 1”, one might say), and where prices soar above costs, which caused Hovenkamp to conclude that Lundbeck “should have been an easy call”.28 This illustrates another kind of dissonance, two institutions talking past each other. If it is true that there has been a growing dissonance between the courts and the FTC in merger enforcement, still deeper misunderstandings may lie ahead now that the more challenging 2010 Guidelines have been adopted. This theme is addressed by Barry Hawk in “A Tale of Two Cities: Washington and Brussels Face the Courts”. The tension Hawk describes is a familiar one (not limited to merger control, but time pressure in this field may make it more poignant) between “sophisticated fact-specific economic analysis” – of which the Guidelines are an example par excellence – and “clear, predictable legal rules”, the kind of bright lines and safe harbors that used to predominate when economists’ prescriptions could more easily fit the less equivocal or even apodictic style of the law. There is some irony here in that, with hints of a political science story, Hawk (and others) portray the 2010 Guidelines – insofar as they broaden the discretion of the Agencies – as very possibly a strategic reaction to inhospitable treatment in the federal courts29 by embracing a larger analytical toolkit; yet the proliferation of refinements could conceivably put still more distance between judges and enforcers. It remains to be seen how relations between these institutions will evolve (after all, jurisprudence does change significantly when enough time elapses), but Hawk, in footnote 17 of his paper, tellingly cites a 2010 judgment in which the Southern District of New York refused to rely on a unilateral effects analysis that was based on the UPP approach. In 2011 the judgment was affirmed by the Second Circuit.30 Other federal courts have been skeptical as well.31 Hovenkamp, cited above note 21. High-profile agency losses include, among others: U.S. v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004): FTC v. Arch Coal, Inc., 329 F. Supp. 2d 109 (D.D.C. 2004); and FTC v. Whole Foods Market, Inc., 502 F. Supp. 2d 1 (D.D.C. 2007). 30 City of New York v Group Health Inc., 649 F.3d 151 (2nd Cir. 2011) (“[W]e find no error or abuse of discretion in the district court’s rejection of the Upward Pricing Pressure test. As the district court explained, and as we discussed above, the applicable case law requires plaintiffs asserting a claim under the Sherman Act, the Clayton Act, or the Donnelly Act to allege a market in which the challenged merger will impair competition. While the City explains the Upward Pricing Pressure test’s usefulness in assessing the impact of a merger, it does not explain how the test can substitute for a definition of the relevant market in the pleadings. Cf. Carl Shapiro, Deputy Ass’t Attorney Gen. for Economics, Antitrust Division, U.S. Dep’t of Justice, Update from the Antitrust Division, at 15 (Nov. 18, 2010), http://www.justice.gov/atr/public/speeches/264295.pdf (recognizing need to define relevant market in any antitrust challenge).”). In the cited speech, Shapiro had stated: “The Division recognizes the necessity of defining a relevant market as part of any merger challenge we bring.” This does not change the fact that the market definition is a means, not an end, and that misconstrued evidence can divert market definition from its purpose (see discussion of Lundbeck above); this seems to suggest that when market definition and other indicators point in opposite directions redoubled scrutiny and caution are appropriate, and the legalistic approach of the courts in Group Health may be insufficiently nuanced. 31 See, e.g., FTC v. Laboratory Corporation of America, No. SACV 10-1873 AG (S.D. Cal. 2011); U.S. v. H&R Block, 2011 WL 548955 (D.D.C. 2011) (where market power can be measured directly, “in theory market definition is superfluous, at least as a matter of economics […] [However, [a]s a matter of law, […] a market definition may be required by Section 7 of the Clayton Act”). 28 29

Introduction 

xxiii  

In part II of his paper, Hawk reports three basic criticisms that have been aimed against the 2010 Guidelines. First, the Guidelines tilt the field in favor of the Agencies and leave them with unfettered discretion, thus undermining certainty and predictability. They “increase the mystery of U.S merger control despite the declared intention to increase transparency”.32 The UPP approach that appears in the Guidelines is, of course, the main source of unease here, since the Guidelines do not offer much precision as to what diversion ratio and gross margin thresholds to watch for in order to stay on safe ground. Compounding this uncertainty seems to be a lack of clarity as to the role of UPP analysis in routine merger work.33 The second, related criticism is the “fanfare” point mentioned earlier – the reports of market definition’s demise. Hawk finds, as others noted at the Workshop, that market definition is not about to fade from sight in the US34 or, for that matter, in the EU35 (or in other jurisdictions). But some of the controversy over what was initially perceived or cast as a threat to market definition was likely fuelled by the suspicion (consistent with the political science story) that UPP puts another arrow in the Agencies’ quiver, and business hardly needs that. The courts may ultimately protect firms from the newfangled slings and arrows, but protection is costly; some incremental increase of intervention (allowing for defensive arguments based on, e.g., new entry, the counterstrategies of rivals and efficiencies) may thus follow from the adoption of the 2010 Guidelines even if market definition remains the touchstone for the courts, as Hawk expects. Related to the point about market definition is, once again, the greater level of “mystery” due to some uncertainty about when the Agencies will rely more or less readily on traditional tools, although intuition can probably be a good guide. The final criticism discussed briefly in the paper is related to the others: will firms across the vast sea of industries be able to collect sufficient, reliable pre-merger margin and diversion ratio data that will enable them to predict Agency findings? In part III of his paper Hawk morphs from detached reporter to agent provocateur and reformer, not to mention religious historian. The title of his paper is drawn Hawk, “Two Cities”, this volume, p. 54. See, e.g., Deborah Feinstein, “The Revised Merger Guidelines: Did the Agencies Heed the Lessons of the Past?”, Antitrust Source (October 2010), p. 3 (“[T]he UPP model is not intuitive to business people, nor particularly easy to implement as an initial screen. If it becomes a necessary initial step involving a data-intensive, months-long process for every transaction that raises unilateral effects concerns, the 2010 Guidelines will not have offered practical guidance.”). 34 In this sense, see, e.g., Gregory Werden, “Why (Ever) Define Markets? An Answer to Professor Kaplow” (2012), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2004655. 35 The European Commission quite often abstains in its decisions from reaching definitive conclusions as to the boundaries of relevant markets, but nevertheless it systematically carries out the market definition exercise, typically searching for narrowly drawn plausible markets in order to err on the side of caution. From the perspective of business operators, an unfortunate side-effect of the Commission’s approach is a lack of reliable precedents as to market definition, and a consequent degree of uncertainty. Narrow market definitions also sometimes imply costly data excavation. From the perspective of administrative costs and opportunity costs, the Commission’s approach may be questioned to the extent that it amounts to “continually reinventing the wheel”. See Jean-François Bellis, Porter Elliott and Johan Van Acker, “The Current State of the EU Merger Control System: Ten Areas Where Improvements Could Be Made”, in Barry Hawk, ed., International Antitrust Law and Policy: Fordham Competition Law Institute 2011, Juris Publishing, 2012, chapter 13, at p. 342. 32 33

xxiv  Merger Control in European and Global Perspective from the contrast he describes briefly between the way the US enforcers have responded to defeats in the courtroom (if this interpretation is accurate) and the way the European Commission responded to Airtours and its sister cases.36 But in the end his emphasis is not so much on the courts; it is on what Hawk sees as a dysfunctional procedural system in the US, the framework installed by the Hart-Scott-Rodino Act of 1976, now past its prime.37 He points, for example, to the insufficient publication of agency decisions not to challenge a merger, which compares poorly with other jurisdictions.38 Reform, he suggests, will be resisted by vested interests. The final paper explicitly covering a substance-related topic is Lars-Hendrik Röller’s “Efficiencies in EU Merger Control: Do They Matter?”, which to some extent seems to resonate with Dan Crane’s “asymmetry principle”.39 If the asymmetry is taken too far, efficiencies don’t matter. The background to this discussion in the EU context, as Röller points out, is the fundamental question of what goal or goals are driving law and policy. Lawyers may favor a range of very different views, but for many economists the answer often amounts to a (nontrivial) choice between a total surplus standard and a consumer surplus standard.40 Here there is a vivid contrast between the framework adopted by the EU for merger control and the Schumpeterian perspective mentioned earlier because in this field of law it is (threats, harm and benefits to) consumer surplus that steers the merger review process.41 Requiring credible proof of “consumer pass-on” would 36 On this comparison (predating the 2010 Guidelines), see also Mark Leddy et al., “Transatlantic Merger Control: The Courts and the Agencies”, 43 Cornell International Law Journal 25, 41-54 (2010). 37 There are other aspects of merger enforcement that could be discussed, including fundamental questions regarding the long-term sustainability (not only as regards jurisdiction over mergers) of the path-dependent two-agency system. Commentary abounds: see, e.g., Daniel Crane, The Institutional Structure of Antitrust Enforcement, OUP, 2011, chapter 2. 38 But see Rachel Brandenburger, “Merger Enforcement in the Americas: Update from the U.S. Department of Justice”, speech, New York, 17 July 2012, http://www.justice.gov/atr/public/ speeches/285173.pdf, at pp. 6-8 (emphasizing the issuance of eight closing statements since 2009 shedding light on the decision of the Antitrust Division in each case to exercise prosecutorial discretion). 39 See above note 24. 40 For further recent discussion, see Louis Kaplow, “On the choice of welfare standards in competition law”, Daniel Zimmer, ed., The Goals of Competition Law, Edward Elgar, 2012, chapter 1. The total surplus/consumer surplus debate has been questioned by economists who would rather put emphasis on how short-term effects analysis can best guarantee desirable long-term results, given the great difficulty of reliable evaluation of long-term effects. See Alison Oldale and Jorge Padilla, “For welfare’s sake? Balancing rivalry and efficiencies in horizontal mergers”, 55 Antitrust Bulletin 953 (2010). 41 While consumer surplus is the comfortable language of economists, the Court of Justice is more apt to use the expression “consumer interests”; but the concepts generally seem compatible, as the ECJ speaks in terms of price, choice, quality and innovation, “among other things”, from the perspective of consumers. See Ekaterina Rousseva and Mel Marquis, “Hell freezes over: a climate change for assessing exclusionary conduct under Article 102 TFEU”, forthcoming, Journal of European Competition Law and Practice. Innovation suggests a framework broad enough to evaluate dynamic efficiency, but the Court unsurprisingly does not make explicit distinctions between static and dynamic efficiencies. It seems likely that the scope for verifying long-term efficiencies and trading them off against short term adverse price or output effects in Article 101(3) and Article 102 cases will depend on the policy preferences of the European Commission. In Merger Control at least, one might infer that there is a “double asymmetry”

Introduction 

xxv  

sometimes be reconcilable with a consumer surplus standard, but not always, and its short-term bias means, once again, that there is little room for Schumpeterian arguments based on cumulative (and potentially massive) productivity growth. Röller proceeds with a review of the criteria employed under the EU’s 2004 Horizontal Merger Guidelines to determine whether efficiencies can justify clearing a merger that otherwise would be incompatible with (i.e., significantly impede effective competition in) the “common market”. These are the familiar conditions of consumer benefits, merger-specificity (causation) and verifiability which have now been made more explicit, albeit with different nuances, in Section 10 of the 2010 US Guidelines as well. Here Röller highlights the clear preference for static benefits, i.e., those that consumers can take advantage of in the relatively short run, and without too much intermediation (variable or marginal cost savings, or in the case of some non-horizontal mergers, avoidance of double mark-ups). The ground is thus laid for Röller to assemble an empirical answer to his question of whether efficiencies matter in EU merger control. His data set is composed of practically all EU-filed Phase II merger cases that arose in the first five years of the regime put in place in 2004, which amounts to 37 cases once withdrawn notifications are discarded. The cases are classified according to whether efficiencies were alleged by the parties in those cases, if so whether the claimed efficiencies were static or dynamic or both, whether the efficiencies were considered to be relevant to the analysis, and whether in the end they carried the day. A first finding is that in only 5 of the 37 cases did the parties assert that their transaction would yield static efficiencies (and dynamic efficiencies were asserted in only 6 cases). Röller finds this dysfunctional, and later argues that a better tailored incentive structure is necessary to encourage parties more frequently to highlight how a proposed merger will produce benefits, since if they do not the Commission is unlikely to do so at its own initiative. He suggests, for example, that the parties might be asked to explain why they have not submitted efficiency arguments, if they have not.42 Another finding is that in the five cases involving claims of static efficiencies, the Commission twice accepted that the cumulative conditions were satisfied – a ratio that seems significant if we do not balk too much at the N=5 problem. On the other hand, Röller points out that although static efficiencies were accepted as relevant in two cases (TomTom/Tele Atlas and Nokia/ NAVTEQ, both non-horizontals), efficiencies were not decisive in either of these. Dynamic efficiency claims fared poorly. While they are not rejected out of hand, the parties failed, in each of the 6 cases where they were asserted, to show that all of the relevant conditions (consumer benefits, merger-specificity and verifiability) were fulfilled. principle at work, which not only makes it relatively difficult to make a decisive efficiency claim, but which makes it particularly difficult to do so with regard to dynamic efficiencies. 42 To be clear, this does not mean that the Commission would be permitted to draw the opposite inference – i.e., a presumption that a merger is inefficient – if the parties failed to come forward with evidence of likely efficiencies. Probably, the sanction could be pitched low (even to the point of a nonsanction) so that it is just enough to provide the desired motivation.

xxvi  Merger Control in European and Global Perspective

2. Merger policy assessment and judicial review Contributions by Duso/Gugler/Szücs and Venit While Röller asks whether efficiencies matter, Duso, Gugler and Szücs ask the more fundamental question of whether the EU merger review process matters. Their paper, entitled “Merger Policy Evaluation: Where Do We Stand?”, is short but packed with empirical propositions that will not be independently verified here. The authors draw together several threads from studies published in the last few years in an effort to erect a multifunction four-part framework that could potentially be used as a standard (or as they put it tongue-in-cheek, “canonical”) tool by researchers going forward.43 The tool consists of the following chronologically inspired parameters: – the ability to predict with confidence how a notified merger will be received by the Commission (legal certainty, policy stability); – type I and type II enforcement errors (where errors are identified according to stock market movements in a time window straddling the announcement of a merger, and in particular shifts in share values of rivals who might stand to gain from more rigid post-merger market structures and softer competition); – effectiveness of remedies (which is distinct in methodology from, but nonetheless seems related to, the concern about type II errors); and – deterrence of “anticompetitive” merger notifications (where horizontal mergers, and the ratio of notified horizontal mergers to notified mergers overall appear to be used by the authors as proxies of anticompetitiveness). Although the authors’ method of measuring predictability – the first parameter – is not crystal clear, their finding that merging parties have been somewhat less certain in recent years (and especially since the introduction of Regulation 139/2004) about whether their transaction might be approved unconditionally is not terribly surprising, since theoretical and empirical techniques have advanced since the 1990s and since this may imply an added layer of complexity when anticipating review outcomes. With regard to the second parameter, decision errors, it may be a good idea to consider stock value-based event studies of merger control effects in a cautious vein. Apart from tricky causality issues and extraneous noise, there may be asymmetric information between those close to the investigation and the market, and other potential limitations (bounded rationality, biases, etc.) that constrain the market’s capacity to know whether a merger will add to or subtract 43 The reader may wish to compare this framework with the evaluative criteria suggested by Bill Kovacic in “International Convergence: Assessing the Quality of Horizontal Merger Enforcement”, this volume, pp. 253 et seq. (discussed below). For another recent study of the economic effects of merger review, focusing on the Greek regime in particular, see Panagiotis Fotis and Michael Polemis, “The Short Run Competitive Effects of Merger Enforcement”, 8 European Competition Journal 183 (2012).

Introduction 

xxvii  

from competitors’ bottom line – even if one may counter that the market takes its own limitations into account and assigns weight and probability values to all relevant factors.44 Be that as it may, the authors record a trend toward more frequent type II errors and a relative decline in type I errors, which is correlated with a relatively diminished rate of outright prohibitions since the adoption of the 2004 regime (with the authors’ time frame running 1990 to 2007). Stock market movements are also relied on in order to evaluate the third parameter, the effectiveness of remedies. Taking account of well-known ex post evaluation studies by DG Comp and by the FTC,45 the authors report conclusions they reached in another paper according to which: “only prohibitions achieve a full reversal of the anticompetitive rents generated at the announcement of a merger. Approvals subject to remedies are only partially effective in achieving rent reversion. Furthermore, remedies seem to be most efficient after the first investigative phase, while their impact is diminished after an in-depth investigation.”46 The sobering thought that Phase II remedies might not matter is complemented by the authors’ finding in relation to deterrence, the fourth parameter:47 outright prohibitions and withdrawals to avoid them generate significant (general) deterrence effects, but “[i]t appears that the Commission blocks too few mergers”.48 The method used by the authors relies to some extent on subsequent merger notification data, including the composition of mergers by type. Although it is not clear from the paper how extraneous factors which might also contribute to a decline in filings are controlled for, one can accept the intuition that an increase in the rate of prohibitions (at least in high-profile cases) may give boardrooms pause before See Gugler’s oral remarks at pp. 98-99. For more recent discussion of studies evaluating merger decision outcomes, see OECD Policy Roundtable, Impact Evaluation of Merger Decisions, DAF/COMP(2011)24 (18 September 2012), http://www.oecd.org/daf/competition/Impactevaluationofmergerdecisions2011.pdf; Fiorenzo Bivenzi and Anna Piesarkiewicz, “The Role of Efficiency Claims in Antitrust Proceedings: Mergers and Dominance Cases”, OECD Background Note, DAF/COMP(2012)16 (11 October 2012), http://search. oecd.org/officialdocuments/displaydocumentpdf/?cote=DAF/COMP%282012%2916&docLanguage =En, at pp. 27-35. 46 Duso, Gugler and Szücs, “Where Do We Stand?”, this volume, p. 108. On the patterns of remedy types (divestitures, access remedies, supply remedies, etc.) employed in horizontal, vertical and conglomerate merger cases, see, e.g., Wei Wang and Matti Rudanko, “EU Remedies and Competition Concerns: An Empirical Assessment”, 18 European Law Journal 555 (2012). 47 Andreas Mundt in his oral remarks opposed the proposition that a significant objective of merger control is to generate general deterrence effects in the general business sector. (See pp. 90-91.) This debate cannot be resolved here, but ideally a merger control system should deter firms from transactions that significantly impede competition while offering no appreciable quantitative or qualitative efficiencies (“naked” anticompetitive mergers). Remedies are futile in such cases, as they render harmless a merger that brings little or no value. If such mergers are notified they are likely to impose collateral costs and drain scarce enforcement resources. Putting aside the question of whether deterrence is an appropriate goal, as a concrete matter it seems clear that the prospect of scrutiny by agencies causes firms to abandon a small but significant proportion of merger plans even prior to notification. With reference to the UK, see Deloitte and Touche LLP, The deterrent effect of competition enforcement by the OFT, OFT962 (November 2007), http://www.oft.gov.uk/shared_oft/reports/Evaluating-OFTs-work/oft962.pdf, at pp. 27-49. The possibility that merger control has overdeterrent effects and dissuades firms from executing pro-competitive transactions is discussed at pp. 44-47 of the report. 48 Duso, Gugler and Szücs, ibid., p. 111. Compare the oral remarks of Mario Siragusa (p. 89, sympathetic to the claim of underenforcement) and Philip Lowe (pp. 93-94, suggesting the contrary). 44 45

xxviii  Merger Control in European and Global Perspective proceeding with transactions liable to raise serious competition concerns. (For example, 2-to-1, 3-to-2 or 4-to-3 mergers, or mergers thought problematic based on a range of market factors, might be passed over even without seeking advice from outside counsel.49) Again, however, the ultimate point of the paper is not so much to report the authors’ specific findings but to present their framework of interrelated yardsticks. In seven paragraphs, Jim Venit summarizes his chapter “The Scope of EU Judicial Review of Commission Merger Decisions” better than I could,50 so I’ll just jot down a few incidental observations. First, the issue of how the EU Courts review the Commission’s decisions in terms of scope and intensity seems to have become a permanent part of competition law discourse. This trend is likely to be irreversible given: (i) the constitutional requirement of effective judicial review (i.e., a fair trial by an independent and impartial tribunal) imposed by Article 47(2) of the EU Charter of Fundamental Rights and by Article 6(1) ECHR,51 and more abstractly by the subordination of the Union to the rule of law; and relatedly, (ii) certain salient features of EUlevel competition law enforcement, which include the potential imposition of stiff sanctions by an integrated, prosecutor-adjudicator agency. Second, Venit’s essay is a sequel to his leviathan chapter from the previous Workshop volume,52 in which he explored related issues but did so largely through the lens of abuse of dominance and restrictive agreement cases (i.e., cases involving conduct punishable by sanctions), rather than merger control. Of course, there are significant connections between judicial review in the field of antitrust See Deloitte and Touche, OFT962, cited above note 47, at p. 35. See Venit, “Scope of EU Judicial Review”, this volume, pp. 113 et seq., at pp. 113-114. A good deal of discussion on judicial review (in the EU system or otherwise) was had in Panel 2. See, among other interventions, the oral remarks of Enrique González Díaz (pp. 75-77 and 99-100), Venit (pp. 7779 and 101), Fanis Christoforou (pp. 81-82), Judge Forwood (pp. 82-84 and 103-104), Ian Forrester (pp. 86-87) and John Fingleton (pp. 88-89). 51 Article 6(3) TEU also provides that fundamental rights – as guaranteed by the ECHR and as derived from constitutional traditions common to the Member States – constitute general principles of EU law. Now that the EU Charter is directly binding and ranks as primary law, the Court of Justice has explained that its Article 47(2) gives expression to the general principle of effective judicial protection and serves to implement Article 6(1) ECHR, which requires States Parties to ensure that the right to a fair trial by an independent and impartial tribunal is respected. See Case C-386/10 P, Chalkor AE Epexergasias Metallon v Commission, judgment of the ECJ of 8 December 2011, not yet reported, paras. 50-52. This raises the question of whether Article 47(2) provides a higher level of protection than Article 6(1) (as elevated standards are permitted both by the Charter and by the Convention). Although the full implications of Article 47(2) are yet to be explored, the right to a fair trial under Article 47(2) applies regardless of whether a procedure is rightly characterized as “criminal” (hard core or, as in the case of administrative enforcement of competition law via sanctions, non-hard core, although debate rages over this distinction) from the perspective of the Convention. In other words, it applies even to the adoption of purely administrative measures falling short of the non-hard core category, and in particular to infringement decisions where no fine is imposed. See Heike Schweitzer, “Judicial Review in EU Competition Law”, in Damien Geradin and Ioannis Lianos, eds., Research Handbook on EU Antitrust Law, Edward Elgar, forthcoming, section III(A)(1). 52 Venit, “Human All Too Human: The Gathering and Assessment of Evidence and the Appropriate Standard of Proof and Judicial Review in Commission Enforcement Proceedings Applying Articles 81 and 82”, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2009: The Evaluation of Evidence and its Judicial Review in Competition Cases, Hart Publishing, 2011, pp. 191 et seq. 49 50

Introduction 

xxix  

and judicial review in contested merger cases.53 Third, while the title of the essay refers to judicial review, Venit also covers the closely related subject of the quality and sufficiency of proof when the General Court reviews the Commission’s merger decisions.54 He examines several significant judgments in these areas (Airtours, Schneider Electric, Tetra Laval, Sony/BMG, Ryanair), extracts the operative legal principles and articulates certain subtleties which are not obvious from the standard formulations of those principles. One sees from Venit’s paper and other sources – and this corroborates the general findings of ECLA 200955 – that the Commission’s margin of appreciation in relation to complex economic matters56 (as opposed to, e.g., errors of law, 53 Crudely speaking, the concepts of marginal review and comprehensive review “spilled over” first from the ECSC Treaty to the field of restrictive agreements under the EEC/EC Treaty (particularly in cases where the Commission assessed, under the defunct notification regime, whether Article 85(3) EEC or Article 81(3) EC applied); from restrictive agreements to EC merger control (especially given the need to evaluate the impact of mergers prospectively); and finally to the field of abuse of dominance (e.g., market definition, interoperability analysis, calculation of price squeeze, etc.). Pertinent cases in these three areas of law include, among others: Joined Cases 56 and 58/64, Consten and Grundig v Commission [1966] ECR 299; Joined Cases 29/83 and 30/83, Compagnie Royale Asturienne des Mines SA and Rheinzink GmbH v Commission [1984] ECR 1679; Case 42/84, Remia v Commission [1985] ECR 2545; Case C-441/07 P, Commission v Alrosa Company Ltd. [2010] ECR I-5949; Case T-427/08, Confédération européenne des associations d’horlogers-réparateurs (CEAHR) v Commission [2010] ECR II-5865; Case C-389/10 P, KME Germany AG, KME France SAS and KME Italy SpA v European Commission, judgment of the ECJ of 8 December 2011, not yet reported; Joined Cases C-68/94 and C-30/95, France and Société commerciale des potasses et de l’azote (SCPA and Entreprise minière et chimique (EMC) v Commission (Kali und Salz) [1998] ECR I-1375; Case C-12/03 P, Commission v Tetra Laval [2005] ECR I-987; Case T-210/01, General Electric v Commission [2005] ECR II-5575; Case T-342/07, Ryanair Holdings plc v Commission [2010] ECR II-3457; Case T-65/96, Kish Glass v Commission [2000] ECR II-1885, upheld: Case C-241/00 P, [2001] ECR I-7759; Case T-201/04, Microsoft v Commission [2007] ECR II-3601; Case T-271/03, Deutsche Telekom v Commission [2008] ECR II-477, upheld: Case C-280/08 P, [2010] ECR I-9555; Case T-321/05, AstraZeneca AB and AstraZeneca plc v Commission [2010] ECR II-2805, on appeal: Case C-457/10 P, not yet decided. See also the judgment of the EFTA Court in Case E-15/10, Posten Norge AS v EFTA Surveillance Authority, judgment of 18 April 2012, http://www.eftacourt.int/images/uploads/15_10_JUDGMENT.pdf. 54 It seems clear that the nature of the standard of proof imposed upon an administrative authority can have important de facto consequences for the standard of review of its decisions, however the latter standard may be denominated. See, e.g., Per Hellström, “A Uniform Standard of Proof for EU Competition Proceedings”, in Ehlermann and Marquis, eds., ECLA 2009: Evaluation of Evidence and its Judicial Review, cited above note 52, pp. 147 et seq.; and cf. Fernando Castillo de la Torre, “Evidence, Proof and Judicial Review in Cartel Cases”, in Ehlermann and Marquis, ibid., pp. 319 et seq., at p. 386. 55 See Ehlermann and Marquis, eds., ibid. For recent discussion, see Schweitzer, “Judicial Review in EU Competition Law”, cited above note 51. For further discussion, see also Renato Nazzini, “Administrative Enforcement, Judicial Review and Fundamental Rights in EU Competition Law: A Comparative Contextual-Functionalist Perspective”, 49 Common Market Law Review 971 (2012). Focusing on the context of infringement proceedings under Articles 101 and 102, Nazzini examines the EU Courts’ de facto deference to the Commission from several angles and notes a shift, particularly in light of the ECJ’s KME judgment (cited above note 53), from deferential review to “full” review (whereby the appropriate standard is “correctness”), but despite this trend finds that, “under the current system, deferential judicial review is incompatible with the principle of effective judicial protection”. Ibid., p. 998. This author advocates changes going well beyond a tightening up of the standard of review, and discusses means of establishing a first-instance decision taker which, unlike the European Commission, would satisfy “constitutional” standards of independence and impartiality. 56 Sometimes in merger cases the word “complex” is not used (as in “margin of discretion with regard to economic matters”), which might give the impression of a broader zone of judicial restraint but one may conclude on the basis of the case law that the distinction is inconsequential.

xxx  Merger Control in European and Global Perspective breach of essential procedural requirements or material errors of fact, to which “comprehensive” legality control applies; or fines, over which the Courts have unlimited jurisdiction according to Article 261 TFEU and Article 31 of Reg. 1/2003) is often a thin margin – sometimes razor-thin,57 unless the complex economic matter falls within a rather vague realm of economic policy choices within the Commission’s remit.58 The general narrowness of the Commission’s margin of assessment is evident from the requirement that the facts on which the Commission relies must be accurate, reliable and “complete”,59 and the requirement that the Commission’s conclusions must follow from that accurate and complete factual record.60 Venit suggests that this fine margin of appreciation 57 The Commission’s recognized margin of appreciation was not always so thin but in general it has become so, subject to nuances, in line with a general trend toward more detailed examination of evidence in most types of competition cases ever since the Court of First Instance opened for business in September 1989. See also the assessment of President Jaeger, who reviews the relevant jurisprudence and concludes: “first, the requirements surrounding the application of marginal review make this standard of review far from light; second, the General Court illustrates this understanding by constantly affirming its willingness to thoroughly review elements despite the fact that they belong to the Commission’s margin of discretion; and finally, the General Court’s actual control goes beyond its mere declared intention to conduct a deep examination of those elements in practice leading – if need be – to the annulment of the challenged decision.” Marc Jaeger, “The Standard of Review in Competition Cases Involving Complex Economic Assessments: Towards a Marginalisation of the Marginal Review?”, 2 Journal of European Competition Law and Practice 295, 303 (2011). Jaeger’s argument that the GC’s “actual control” is intensive may be contrasted with Venit’s oral remarks (see p. 101), where he qualifies his paper by claiming the GC often fails to bring the full intensity of its power of review to bear. See further Nazzini, “Administrative Enforcement”, cited above note 55, at pp. 992 et seq. 58 See Jaeger, ibid., pp. 310 and 313. In the past, matters of genuine policy orientation – to which only a marginal review (manifest error) standard applied – notably included the Commission’s judgement calls when applying Article 85 EEC or its equivalent. (See Nicholas Forwood, “The Commission’s ‘More Economic Approach’ – Implications for the Role of the EU Courts, the Treatment of Economic Evidence and the Scope of Judicial Review”, in Ehlermann and Marquis, eds., Evaluation of Evidence and its Judicial Review, cited above note 52, at pp. 255 et seq.) But the days of constitutive exemption decisions by a central enforcer are over, and since 2004 Article 101(3) TFEU has been directly applicable, which may imply an eradication of the zone of policy discretion the Commission once had in this context. (See Heike Schweitzer, “The European Competition Law Enforcement System and the Evolution of Judicial Review”, in Ehlermann and Marquis, ibid., pp. 79 et seq. at pp. 98, 102 and 140.) Other matters have been held to fall within the Commission’s policy remit, including in particular where its enforcement priorities lie. However, questions remain as to how to elaborate more precise guidelines with regard to the shades of gray between matters that constitute genuine questions of economic policy that may warrant institutional-balance deference (i.e., as Jaeger says at p. 313, marginal review “applied with the intensity described in Tetra Laval’s ‘forgotten’ paragraph”) and juridified matters that do not. Schweitzer maps out and discusses these issues in “Judicial Review in EU Competition Law”, cited above note 51, especially in sections II(C)(4)(b) and IV of her paper. 59 “Completeness” cannot refer to an absolute concept. In practice, the fact-finder will tend to gather pertinent information until the perceived marginal value of additional data collection is overtaken by its marginal costs. On appeal, the General Court has no jurisdictional impediment to concluding that the Commission’s decision is flawed by a material omission of fact, which amounts to a finding that the Commission’s cost-benefit calculation was from an ex post point of view deficient. 60 These significant qualifications emerge from what Jaeger (cited above note 57, at p. 300) describes as Tetra Laval’s “forgotten paragraph”, the forgetters being critics sometimes protesting too much about judicial deference for the Commission. The paragraph in question (para. 39) states: “Whilst the [ECJ] recognises that the Commission has a margin of discretion with regard to economic matters, that does not mean that the Courts must refrain from reviewing the Commission’s interpretation of information of an economic nature. Not only must the Courts, inter alia, establish whether the evidence relied on is factually accurate, reliable and consistent but also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of

Introduction 

xxxi  

corresponds to a “plausibility” analysis, and plausibility does not imply an obligation to tell the best story; the existence of more plausible conclusions does not justify annulment. In theory and in practice this seems to allow for more rigor than in the case of, say, a “substantial evidence” standard of review and less rigor than “full merits” review (where even a wholesale rewrite of a contested decision, i.e., a review going well beyond full or partial annulment, is theoretically possible). When the General Court controls the plausibility of the Commission’s conclusions, the ECJ can intervene and annul the decision of the GC if the latter has gone so far as to substitute its own preferences, including its own economic assessments, for those of the Commission;61 but in principle it is for the GC alone to assess the probative character, credibility and persuasiveness of the Commission’s evidence,62 and to verify that the evidence substantiates the Commission’s findings. The only other observation that will be made here concerning this paper is that Venit presents the descriptive and normative case for the proposition that, with regard to complex matters that are by nature speculative – theories predicting conglomerate effects (leveraging) or coordinated effects, and possibly other postmerger predictions, since merger analysis is ultimately prospective – the EU Courts (do and should) employ heightened scrutiny by the back door, i.e., by rigorously checking the quality and sufficiency of the evidence (or, in Venit’s terms, by raising the “burden of proof”).63 In this regard, Venit suggests that “the intensity of the Court’s analysis will be proportional to the speculative nature of the Commission’s theory of harm, with the intensity of judicial scrutiny being greater in cases involving conglomerate (and arguably coordinated) effects, as opposed to more straightforward cases of unilateral horizontal effects”.64 In such cases, the Commission will normally have to support its decision to intervene substantiating the conclusions drawn from it. Such a review is all the more necessary in the case of a prospective analysis required when examining a planned merger with conglomerate effect.” Although Tetra Laval was a merger case, the limits to marginal review imposed by the forgotten paragraph apply equally in non-merger cases, as can be seen in a growing number of judgments. 61 See Case C-441/07 P, Alrosa, cited above note 53, paras. 66-67 (finding that the “General Court” had gone too far in its proportionality review of the Commission’s findings as to the acceptability of (initially jointly) proposed commitments in an Article 9 case: by reaching “personal” conclusions regarding whether jointly proposed commitments could eliminate the competition concerns identified by the Commission, the GC had “put forward its own assessment of complex economic circumstances and thus substituted its own assessment for that of the Commission, thereby encroaching on the discretion enjoyed by the Commission instead of reviewing the lawfulness of its assessment”.). 62 See, e.g., Case C-136/92 P, Commission v Brazzelli Lualdi and others [1994] ECR I-1981, para. 66; Case C-237/98 P, Dorsch Consult v Council and Commission [2000] ECR I-4549, para. 50. The principle that the ECJ may generally only review points of law, if applied too expansively, may lead to concerns that the General Court is in some ways exempt from control. See Venit’s oral remarks at pp. 78-79. 63 Cf. Arjen Meij, “Judicial Review in the EC Courts: Tetra Laval and Beyond”, in Oda Essens, Anna Gerbrandy and Saskia Lavrijssen, eds., National Courts and the Standard of Review in Competition Law and Economic Regulation, Europa Law Publishing, 2009, chapter 2 (explaining that in Tetra Laval the CFI had endeavored to articulate the applicable standard of proof, as opposed to establishing a new standard of scrutiny going beyond the standard seen in the ECJ’s Kali und Salz judgment). 64 See Venit, “Scope”, this volume, pp. 125-126.

xxxii  Merger Control in European and Global Perspective with particularly convincing evidence, a proposition consistent with the evidence needed to prove that a lion feasted on daisies in Regent’s Park last Sunday.65 Venit astutely notes a tension between the intuitive lion-in-the-park, sliding scale approach when assessing evidence and the odd “symmetry” principle established in Sony/BMG.66 If theories of adverse effects defy normal expectations when a conglomerate merger is consummated (see the CFI in Tetra Laval67), wouldn’t it be natural to allow such a merger to benefit from a rebuttable presumption of compatibility with the common market?68 In adopting the symmetry principle in Sony/BMG, the ECJ in this sense seems to have established an asymmetry between merger control cases and antitrust cases. But again, Venit suggests that while the standard of proof for prohibitions and approvals is the same,69 one sees upon scratching the surface differences in the evaluation of evidence; and from a functional perspective one sees something roughly comparable to a presumption embedded within the sliding scale.

65 Several parkgoers claimed to have spotted Lord Hoffmann’s stately beast firsthand in Ehlermann and Marquis, ECLA 2009, cited above note 52. See, e.g., Luis Ortiz Blanco, “Standards of Proof and Personal Conviction in EU Antitrust and Merger Control Procedures”, in ibid., pp. 175 et seq. 66 Case 413/06 P, Bertelsmann AG and Sony Corporation of America v Independent Music Publishers and Labels Association (Impala) [2008] ECR 4951, para. 43. 67 Case T-80/02, Tetra Laval BV v Commission [2002] ECR II-4381, para. 155 (“Since the effects of a conglomerate-type merger are generally considered to be neutral, or even beneficial, for competition on the markets concerned, as is recognised in the present case by the economic writings cited in the analyses annexed to the parties’ written pleadings, the proof of anti-competitive conglomerate effects of such a merger calls for a precise examination, supported by convincing evidence, of the circumstances which allegedly produce those effects (see, by analogy, Airtours v Commission, paragraph 63).”). Cf. Case C-12/03 P, Commission v Tetra Laval [2005] ECR I-987, paras. 40-41 (endorsing the need affirmed by the CFI for a “close examination” of effects on the reference market in light of the nature of the notified merger and rejecting the contention that the CFI was introducing a more rigorous standard of proof to be met by the Commission, opining instead that the CFI had merely drawn attention to the “essential function of evidence, which is to establish convincingly the merits of an argument or, as in the present case, of a decision on a merger”); Case C-413/06 P, Bertelsmann AG and Sony Corporation v Impala, cited previous footnote, paras. 50-51 (quality of evidence in conglomerate merger cases particularly important, and again, a reference to the essential function of evidence). 68 Coordinated effects cases may arguably raise somewhat different issues. Rigorous analysis to avoid type I errors is certainly commendable. However, vigilance seems equally important, all the more so in specific instances such as maverick-absorption cases. To some degree, the intensification of civil and criminal punishment for cartel activity may be driving firms seeking supracompetitive profits to consider horizontal concentration and competition-softening strategies as a better risk compared to cartel activity. The latter point is raised, for example, in William Kovacic, “Criminal Enforcement Norms in Competition Policy: Insights from US Experience”, in Caron Beaton-Wells and Ariel Ezrachi, Criminalising Cartels: Critical Studies of an International Regulatory Movement, Hart Publishing, 2011, chapter 3, at p. 55. 69 In essence, the Commission must show that its treatment of the merger in question is justified given the balance of probabilities (economic outcome “most likely to ensue”) absent its decision to (conditionally) approve or prohibit the transaction. Cf. Aindrias Ó Caoimh, “Standard of Proof, Burden of Proof, Standards of Review and Evaluation of Evidence in Antitrust and Merger Cases: Perspective of the Court of Justice of the European Union”, in Ehlermann and Marquis, ECLA 2009, cited above note 52, pp. 271 et seq., at 281-282 (quoting Case 413/06 P, Bertelsmann AG and Sony Corporation v Impala, cited above note 66, para. 52).

Introduction 

xxxiii  

3. Merger enforcement across jurisdictions: Procedural issues Contributions by Völcker, Forrester, Goldman Sven Völcker’s paper – “Dare to Defer? Towards Greater Procedural Efficiency in Multijurisdictional Merger Remedies” – is a polite call to action for the world’s antitrust agencies with regard to coordinated remedy design and/or agency restraint based on work sharing, comity or “pure” deference. Outside of special cases such as the EU, competition problems arising from multijurisdictional mergers cannot be resolved via a one-stop shop remedy procedure, just as a one-stop global filing procedure is unrealistic. However, de facto coordination of remedies is already possible, particularly where two or more agencies and the merging parties all cooperate in good faith. When effective work sharing is achieved (or even when acting on a more unilateral basis), Agency X can close its file with little fuss on the ground that the merging parties have agreed divestitures with Agency Y (and maybe Agency Z, etc.) which fully address any likely competition problems in X’s jurisdiction that flow from the merger.70 In this regard, Völcker reviews some of the relevant practice, points out the potential difficulties which may on a formal level interfere with cooperation, and then shows – using the EU merger rules as an example – how practical solutions can overcome formal obstacles (and how the Commission’s enforcement powers can allay fears of system-gaming). Agencies with well developed remedies policies, or perhaps the OECD and/or ICN, could take the lead in distilling practice into non-binding principles relating to “core safeguards” and work allocation in order to build trust and encourage more frequent deference in appropriate cases. Such initiatives could build on the existing degree of bilateral guidance already provided by the EU and the US in the form of a more general Best Practices document.71 A range of issues concerning cooperation and comity are taken up again in section 4 (“International convergence, etc.”) below. 70 For example, the Canadian Competition Bureau decided to take no action in the (US $18 billion) United Technologies/Goodrich case on the ground that remedies secured by the DOJ and the European Commission sufficed to alleviate competition concerns in Canada. See CCB Press Release of 26 July 2012, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03483.html. The cooperative and remedial aspects of the case are discussed in Rachel Brandenburger, “Recent Developments in Merger Control: Views from the U.S. Department of Justice’s Antitrust Division”, speech, Florence, 14 September 2012, http://www.justice.gov/atr/public/speeches/286981.pdf, at pp. 13-16. Another frequently cited example is the Cisco/Tandberg case. See DOJ Press Release of 29 March 2010, http://www.justice.gov/atr/public/press_releases/2010/257173.htm (closing investigation after taking account of commitments made to the European Commission). Concerning this case, Christine Varney made pertinent remarks shortly before the DOJ published its press release. See Varney, “Coordinated Remedies: Convergence, Cooperation, and the Role of Transparency”, speech, Paris, 15 February 2010, http://www.justice.gov/atr/public/speeches/255189.htm. 71 US-EU Working Group, Best Practices on Cooperation in Merger Investigations (October 2011), http://ec.europa.eu/competition/mergers/legislation/best_practices_2011_en.pdf (recommending that merging parties coordinate the timing and substance of remedy proposals and to participate in joint discussions on remedies with the Commission and with the FTC or DOJ).

xxxiv  Merger Control in European and Global Perspective Like Völcker, Ian Forrester – in his paper, “POST PLURES UNUM: Streamlining and Simplifying Merger Procedures in an Era of Multijurisdictional Merger Filings” – writes chiefly about Europe but also raises some procedural issues that need to be addressed by jurisdictions around the globe. The problems stem from a generalized parochial bias which leads legislatures to adopt merger regimes (in particular, notification thresholds) that do not take account of the full brunt of global regulatory costs, and which are sometimes disproportionate in light of limited links between the merging parties and the jurisdiction in question. Perhaps empirical measurements (independently conducted and financed) have been made of the true additional (static and dynamic) private and social costs caused by clearly gratuitous filings that serve little purpose,72 but in any case Forrester’s more qualitative analysis seems to confirm the intuition that the global patchwork of filing rules constitutes an appreciable drain on the global economy that tends to dampen investment.73 Forrester begins his paper with an unexpected but engaging historical recollection of the hard road that had to be traversed in order to secure a (post plures unum) European-level procedural framework for the vetting of large mergers and acquisitions.74 This first part of the paper might even have been extracted as a fitting preface to this volume: considering the economic importance and competitive impact of such transactions, one can easily point to the merger control system as one of EU competition law’s capital achievements, ranking with the adoption of the competition rules of the Treaty itself. But the remainder of the paper is equally interesting; it proceeds to highlight the role of the Merger Regulation (i.e., Regulation 139/2004) in combating against economic nationalism and political interference that would stifle cross-border investment in a manner contrary to an ever closer Union. The key federalesque instrument used to check this kind of “spillback” is Article 21 of the Regulation.75 72 For a study commissioned by the private bar which focuses on private costs, see PricewaterhouseCoopers, “A tax on mergers? Surveying the costs to business of multi-jurisdictional merger review” (June 2003), http://www.pwcglobal.com/uk/eng/about/svcs/vs/pwc_mergers.pdf, cited by Adam Fanaki, “Convergence in Multi-Jurisdictional Merger Reviews: A Canadian Perspective”, this volume, pp. 261 et seq. Since 2003 the level of this “tax” in real terms has undoubtedly risen. 73 On the other hand, there is surely a body of “inframarginal” merger activity that is unlikely to be deterred by high aggregate regulatory costs in absolute terms due to the potentially enormous scale and value of some international transactions. Any chilling effect that may result from extra regulatory and compliance costs occasioned by a multiplicity of merger regimes seems more likely to affect transactions of lesser value. 74 For an in-depth retrospective on the contentious and lengthy efforts to convince the Council of Ministers to approve the Community’s first merger regulation – which in the end was only possible with the aid of the ECJ, as Forrester explains – see Hubert Buch-Hansen, Rethinking the history of European level merger control. A critical political economy perspective, Copenhagen Business School Press, 2009. For a variety of legal and historical perspectives on the history of competition law and policy (and industrial policy) in the Community (i.e., the EEC) and its largest Member States, see Kiran Patel and Heike Schweitzer, eds., Historical Foundations of European Competition Law, Oxford University Press, forthcoming. 75 In principle, concentrations with a “Community [now a Union] dimension” are subject, so far as the EU is concerned, to the exclusive administrative jurisdiction of the European Commission. However, Article 21(4) of the Merger Regulation provides that “Member States may take appropriate measures to protect legitimate interests other than those taken into consideration by this Regulation

Introduction 

xxxv  

The most specific parts of Forrester’s essay concern notification thresholds and the EU’s pre- and post-notification referral mechanisms (Articles 4(4), 4(5), 9 and 22 of the Merger Regulation).76 With regard to notification thresholds, the issues include, as noted above, jurisdictional nexus (stressed often in international fora), possible perverse incentives arising from inordinate filing fees, and occasionally unjustified complexity and opacity (or even irrationality) of filing thresholds, leading to uncertainty and pointless risk. As for (total or partial) referrals to and from the Commission, which occur relatively frequently,77 Forrester provides elaborate comments that highlight their unsatisfactory overall performance. The issues here generally relate to: needlessly excessive time windows in which the Commission or the Member States, as the case may be, can mull over the acceptability of a mooted referral; and needlessly onerous questions that could be struck out from the Form RS. In brief, Forrester finds significant room for improvement in these technical but practically important areas, and makes numerous suggestions to that end. In Cal Goldman’s paper, “Contemporary US-Canada Cross-border Merger Review: Tradeoffs in Policy Objectives between Harmonization, Simplification and Accountability”, both the bright side and the potentially dark side of procedural convergence are seen. Canada, orbiting as it does in the gravitational field of the United States, continually faces substantive and procedural choices that are to some extent thrust upon it by US policy preferences; and yet the latest procedural alignments78 occurred only in 2009-2010, while the Hart-Scott-Rodino Act was adopted in 1976 and NAFTA entered into force in 1994. It may be that US/Canadian convergence is driven almost as much by global trends, and by the gentle prodding of the ICN, as it is by gravity – which after all is the weakest of the fundamental forces. [i.e., essentially competition-based criteria] and compatible with the general principles [fundamental rights protection, proportionality, legitimate expectations, etc.] and other provisions of Community [Union] law”. Legitimate interests automatically include public security, plurality of the media and prudential rules, but “[a]ny other public interest must be communicated to the Commission by the Member State concerned and shall be recognised by the Commission after an assessment of its compatibility with the general principles and other provisions of [Union] law before the measures […] may be taken. The Commission shall inform the Member State concerned of its decision within 25 working days of that communication.” 76 For further critical remarks concerning case referrals under the Regulation 139/2004, see also Ioannis Kokkoris and Krisztian Katona, “Critical Analysis of the ECMR Reform”, in Ioannis Lianos and Ioannis Kokkoris, eds., The Reform of EC Competition Law, Kluwer Law International, 2010, chapter 16, at pp. 439-444; Bellis, Elliott and Van Acker, “Ten Areas”, cited above note 35, at pp. 335337 and 348-350 (questioning, inter alia, the tendency of case teams to use the submission of the (draft) Form RS as an occasion to explore substantive issues; and calling – unless harmonizing steps are taken to address persistently divergent processes among EU Member States – for an expanded concept of referral whereby transactions could be referred if notifiable in just two or more Member States). 77 See Thomas Deisenhofer, “International Cooperation in Merger Cases – An EU Practitioner’s Perspective”, this volume, pp. 227 et seq. 78 Further substantive convergence was prompted by the 2010 US Horizontal Merger Guidelines. The Canadian Competition Bureau responded in October 2011 with new Merger Enforcement Guidelines (briefly covering non-horizontal mergers as well). See http://www.competitionbureau. gc.ca/eic/site/cb-bc.nsf/vwapj/cb-meg-2011-e.pdf/$FILE/cb-meg-2011-e.pdf.

xxxvi  Merger Control in European and Global Perspective In his paper, Goldman reports Canada’s most significant procedural reforms, which were largely modeled on the “second request” process in the US.79 Although he finds certain aspects of the reforms to be helpful (e.g., nominally more compatible 30-day + 30-day waiting periods), his reservations concern the rather fundamental matter of whether second requests (or in the Canadian style, “supplementary information requests”) are a suitable model for merger review. It is intimated in the paper that second requests can involve disproportionately burdensome data requirements, and that such a costly and extensive process is hardly conducive to simplifying and streamlining the Canadian regime (one of the “tradeoffs” referred to in Goldman’s title).80 While Canada’s reforms were directly inspired by the US model, Goldman also explains how certain local practices will continue to reflect Canada’s specific institutions, customs and available resources.81 But in importing the central pillars of the framework, Canada has cast its lot with the Procrustean approach.82

4. International convergence: Substantive and procedural issues and the scope for comity Contributions by Mundt/Bardong, Deisenhofer, Brandenburger, Kovacic, Fanaki, Shiraishi, Kameoka/Marquis, Jeon, Zhang/Zhang, Mateus, Dabbah Some of the international-dimension themes seen already – such as managing complexity, diversity and time pressure in a global and multipolar transaction 79 These procedural reforms are also discussed by Fanaki, “A Canadian Perspective”, this volume, pp. 261 et seq. 80 Arguably, when measuring red tape, treatment in the US of cases that do not provoke a second request may compare favorably with other significant jurisdictions. It is not unusual to see criticism of the information requirements contained in the Form CO in Europe, and the EU’s “simplified procedure” as currently configured is regarded as underinclusive and ultimately of limited value. See, e.g., Bellis, Elliott and Van Acker, “Ten Areas”, cited above note 35, at pp. 327-334. 81 See also Fanaki, “A Canadian Perspective”, at pp. 267-268 (referring to certain distinctions remaining between the Canadian and US procedures, such as Canada’s one-year time window for ex post merger challenges and the possibility of obtaining an Advanced Ruling Certificate (i.e., premerger no-action ruling) from the Commissioner. 82 The tradeoff seems to be whether gains arising from enhanced synchronicity exceed the additional costs imposed by supplemental information requests – taking account of the low percentage of notified mergers subject to such requests. Some of the benefits of the broad alignment of the US and Canadian regimes are reviewed by Fanaki in his paper, and he observes that the scope of data and document requests in issued SIRs has been more narrow than that of the model Second Request, but he stops short of claiming that Canada’s reforms have in fact reduced the relative cost of multijurisdictional merger review where one of the jurisdictions in which the merger must be cleared is Canada. Instead, he calls for further study (noting also that other factors not causally linked to the procedural reforms, and in particular a turn toward effects-based merger analysis) tend to increase the costs of review). The latest stocktaking suggests that the Canadian Competition Bureau has exercised some restraint with the new procedure, and that overly burdensome use of SIRs has been avoided. See Mark Katz and Erika Douglas, “The Canadian ‘SIR’ Process: A Progress Report”, CPI Antitrust Chronicle, September 2012(1).

Introduction 

xxxvii  

arena – are also treated in the many contributions to this session 4, which was more explicitly dedicated to international perspectives and issues. The first of these contributions is Andreas Mundt and Andreas Bardong, “Comity, Cooperation and International Convergence – Recent Developments in German Merger Control”. Mundt and Bardong cover, among other things, the exercise of jurisdiction over foreign-to-foreign mergers, comity and the “nexus” principle, inter-agency cooperation and (procedural and substantive) convergence. With respect to foreign-to-foreign mergers, Mundt and Bardong describe the (fully litigated) Phonak/ReSound case, where the parties came forward with several quaint arguments as to why public international law should have forestalled the assertion of jurisdiction (and a fortiori intervention) by the Bundeskartellamt. These arguments have the superficial appeal of being linked to classical ideals of sovereignty and territoriality. Although public international law may be criticized for sometimes (not always) being too faithfully wedded to paradigms of the past,83 this is not so in relation to the extraterritorial application of domestic law where a cognizable harm is suffered in a State that otherwise does not appear to have a plausible claim to exercise jurisdiction on the basis of classical territorial links. Indeed, the foundations of the idea that jurisdiction necessarily presupposes such links84 were unsettled, to some extent, 85 years ago by the Permanent International Court of Justice (i.e., the forerunner of today’s ICJ).85 Since then the doctrines of significant effects and “implementation” (which are distinct86) have taken various twists and turns,87 and they certainly have not rendered territoriality irrelevant,88 83 See recently Germany v. Italy, judgment of the International Court of Justice of 3 February 2012, not yet reported, http://www.icj-cij.org/docket/files/143/16883.pdf. 84 See American Banana Co. v. United Fruit Co., 213 U.S. 347, 356-357 (1909) (Sherman Act capable of application only to conduct occurring within US borders; with the Supreme Court claiming that “the general and almost universal rule is that the character of an act as lawful or unlawful must be determined wholly by the law of the country where the act is done”). 85 See France v. Turkey (Affair du « Lotus »), judgment of the PICJ of 7 September 1927, Series A No. 10, http://www.icj-cij.org/pcij/serie_A/A_10/30_Lotus_Arret.pdf (confirming proper exercise of criminal jurisdiction by Turkey over French ship officer Lieutenant Demons, accused of an allegedly criminal act, i.e., allowing his ship to tear a Turkish ship apart, causing death and injury, on the high seas. Arguably, this could be called a hybrid case that could be read as either an extension of territorybased jurisdiction (the “territory” being a sailing ship) or as an embryonic effects-doctrine case, although there are passages, such as at page 23, that leave doubts). See further the Opinion of Advocate General Darmon in Joined Cases 89, 104, 114, 116, 117 and 125 to 129/85, Ahlström Osakeyhtiö and others v Commission [1988] ECR 5193, paras. 19-58. 86 See, e.g., Yves Botteman and Agapi Patsa, “The Jurisdictional Reach of EU Anti-Cartel Rules: Unmuddling the Limits”, 8 European Competition Journal 365, 379 (2012). 87 There is abundant literature on the subject, and the point of departure is, of course, Judge Hand’s Opinion in Alcoa. See, e.g., Eleanor Fox, “Competition Law”, in Andreas Lowenfeld, International Economic Law, 2nd edition, OUP, 2008, chapter 14, pp. 423-431; Mark Popofsky, “Extraterritorial Jurisdiction: Legal Foundations and U.S. Jurisprudence”, in Wayne Dale Collins, ed., Issues in Competition Law and Policy Vol. III, ABA Section Antitrust, 2008, pp. 2417 et seq.); and some of the contributions in Andrew Guzman, ed., Cooperation, Comity and Competition Policy, OUP, 2011. The subject is also discussed in Tadashi Shiraishi, “Effects on Domestic Purchasers: A Descriptive Theory for Competition Law in Cross-Border Cases”, this volume, pp. 275 et seq. 88 Sovereignty may be “in transition”, but it is sticky, and it takes offense. Historically, many States defended their prerogatives by adopting blocking statutes (which remain in place; see, e.g., Erica Davila, “International E-Discovery: Navigating the Maze”, 8 Pittsburgh Journal of Technology Law and Policy 1, 11-14 (2008)). Good ol’ Westphalian manoeuvring was exemplified recently by a Russian

xxxviii  Merger Control in European and Global Perspective but with time they have been adopted, subject to limiting principles, as common “State” practice. Barring a wildly improbable turn toward a global merger control system, these widely applied doctrines will continue to be of practical relevance. Mundt and Bardong accept that in some scenarios the exercise of merger control jurisdiction may have to be weighed against actual or likely harm to another State’s important interests (e.g., where national security is under threat), but such cases are far out of the ordinary. On the other hand, it is evident that in some instances a State may voluntarily decline to exercise jurisdiction irrespective of such balancing. Here the authors refer to the (non-binding) principle of comity, which is reflected in the sound notion that only jurisdictions with a sufficient nexus to a given transaction should exercise jurisdiction over it.89 In this regard they show how the nexus principle influenced the 2009 refinements of the German merger regime, including in particular the adoption of a second domestic turnover test which probably contributed (along with the economic crisis) to a decline in notifications in Germany. The strengthening of the local nexus requirement is portrayed by the authors as one feature among others of a svelte, state-of-the-art merger regime; to substantiate the “competitiveness” of German merger control, they refer to ICN documents and international (product) comparisons. Turning to the related subjects of inter-agency cooperation and convergence, Mundt and Bardong first of all review some of the multijurisdictional merger cases on which the Bundeskartellamt cooperated with other authorities, such as the curious case of BHP/Rio Tinto. Here the authors suggest that the modalities of cooperation between competition authorities can be more important than procedural convergence as such. They also stress the need for substantive convergence (to avoid conflicting decisions and to facilitate cooperation90), which may already be an acquis among some jurisdictions as far as horizontal merger cases are concerned,91 although one can still occasionally observe varying outcomes even executive decree of 11 September 2012 that looks likely to interfere with the European Commission’s ongoing antitrust investigation of Gazprom (see Press Release 12/937 of 4 September 2012). Even if a practical solution were ultimately to be found in this case, the intervention of the government sends a grim message about State-owned enterprises and uneven playing fields. 89 See, e.g., ICN Recommended Practices for Merger Notification Procedures, http://www. internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 90 Cristina Caffarra gives an example of how extra costs may result from divergent theories of harm in her oral remarks at pp. 15-16. For a fuller account of the problems associated with substantive divergence in merger control, see, among others, Daniel Crane, “Substance, Procedure, and Institutions in the International Harmonization of Competition Policy”, 10 Chicago Journal of International Law 143, 148 (2009) (discussing negative externalities, potential overdeterrence and opportunistic behavior on the part of regulators). On convergence and divergence in general, see Cheng, “Convergence and Its Discontents”, cited and quoted above note 7. 91 Gauging US/EU convergence with regard to non-horizontal issues in recent years (in particular, post-GE/Honeywell) is less straightforward. (Moreover, it is not clear what future directions the FTC and DOJ might take given ongoing personnel changes, such as, e.g., the departure of Commissioner Rosch.) On the one hand, the form used to notify a merger under the Hart-Scott-Rodino Act does not (any longer) inquire as to the merging parties’ vertical relationships, a sign of nonchalance. On the other hand, nonhorizontal aspects have been cited in enforcers’ speeches and publications as a potentially promising line of development. See J. Thomas Rosch, “Terra Incognita: Vertical and Conglomerate Merger and Interlocking Directorate Law Enforcement in the United States”, speech, Hong Kong, 11 September

Introduction 

xxxix  

in cases of horizontal effects.92 Embracing a kind of “evolutionary convergence” approach for substantive matters, Mundt and Bardong provide an update with regard to the preparation of merger guidelines in Germany, now adopted in their final form,93 and other envisaged or possible initiatives. The traditional dominance test in Germany is probably living out the last of its days, but experience suggests that its successor will not provoke a radical change of practice. Thomas Deisenhofer is next with “International Cooperation in Merger Cases – An EU Practitioner’s Perspective”. Senior-level contacts between the European and US competition enforcers date back to the 1960s, when Hans von der Groeben flew to Washington and Don Turner later paid a visit to DG “IV”. But cooperation in merger control, which only goes back about two decades as far as Brussels is concerned, and cooperation in international cartel cases have necessarily been of an intensity and urgency that requires particularly close contacts at the level of operations – cooperation in real time.94 Deisenhofer’s paper is illuminating in this 2009, http://www.ftc.gov/speeches/rosch/090911roschspeechunivhongkong.pdf, at pp. 7-15 (discussing the sparse FTC practice since the Reagan/Baxter era with regard to non-horizontal mergers, but taking some inspiration from the Commission’s Guidelines on Non-Horizontal Mergers and signaling a possibly more assertive approach in non-horizontal merger cases); J. Thomas Rosch and Darren S. Tucker, “Emerging Theories of Competitive Harm in Merger Enforcement”, Antitrust Source (October 2011). As regards the DOJ, recent cases suggesting increasing attention to non-horizontal mergers include Google/ ITA, GrafTech/Seadrift Coke and the Comcast/NBC joint venture. See also DOJ, Antitrust Division Guide to Merger Remedies (June 2011), http://www.justice.gov/atr/public/guidelines/272350.pdf, at p. 5 (“vertical mergers can create changed incentives and enhance the ability of the merged firm to impair the competitive process. In such situations, a remedy that counteracts these changed incentives or eliminates the merged firm’s ability to act on them may be appropriate. Accordingly, in appropriate vertical merger matters the Division will consider tailored conduct remedies designed to prevent conduct that might harm consumers while still allowing the efficiencies that may come from the merger to be realized. The Division also will consider structural remedies in vertical merger matters – they may be particularly effective when the vertical integration is a small part of a larger deal.”). 92 See Deborah Feinstein, “Process Divergence As an Obstacle to Substantive Convergence?”, Antitrust 5, 5 (Summer 2012) (contrasting the handling of Intel/McAfee by the European Commission, which required commitments, and the FTC’s unconditional clearance; and citing cases involving other significant jurisdictions). Of course, some jurisdictions, including now most famously China, have merger control provisions that permit the relevant agency (generally MOFCOM’s Anti-Monopoly Bureau in China’s case) to bring non-competition criteria (notably industrial policy, among others) into the review process, sometimes in opaque ways. To the extent that non-competition criteria override a strictly effects-based assessment, divergence with other jurisdictions is to be expected. For a recent assessment of merger control in China, see Xinzhu Zhang and Vanessa Yanhua Zhang, “China’s Merger Control Policy: A Three-Year Milestone”, this volume, pp. 303 et seq. Much has been written on the specific subject of the explicit and implicit industrial policy features of Chinese competition law and policy. See, e.g., Xiaoye Wang and Jessica Wu, “China’s Anti-Monopoly Law: agent of competition enhancement or engine of industrial policy?” in Zimmer, ed., The Goals of Competition Law, cited above note 40, chapter 20. 93 The Guidelines cover vertical and conglomerate mergers as well as horizontal mergers. See BKA, Guidance on Substantive Merger Control (29 March 2012), http://www.bundeskartellamt. de/wEnglisch/download/pdf/Merkblaetter/2012-03-29_Guidance_final_neu.pdf. See also Andreas Bardong, “The Bundeskartellamt’s New Merger Guidelines”, Journal of European Competition Law and Practice (2012) doi: 10.1093/jeclap/lps042. 94 Initial steps toward the establishment of a Framework for Merger Review Cooperation were taken in April 2012. The presently humble ambitions of the non-binding, opt-in Framework include the compilation of a list of liaison officers and the development of modes of information exchange, subject to confidentiality. The Framework may thus facilitate inter-agency discussions of, in particular, time schedules, market definition, theories of the case and remedies. See http://www.jftc.go.jp/en/ policy_enforcement/speeches/pdf/1Framework%20for%20Merger%20Review%20Cooperation.pdf.

xl  Merger Control in European and Global Perspective respect, as it contains both statistical and qualitative analysis of both “external” cooperation, i.e., cooperation with third countries, and internal cooperation, i.e., cooperation with EEA jurisdictions. The statistical analysis draws on two merger case datasets from the periods of January 2007 to 30 September 2009 and November 2008 to January 2011. The first of these is comprehensive (926 merger decisions), the other is tailored to include only those cases that require the special attention of senior management (“significant” cases) and those requiring substantial resources (“heavy” cases), which combined amounted to 94 cases. Of these 94 cases, if one removes the 53 cases where the merging parties were both EU-established undertakings, it appears that DG Comp’s case teams cooperated with their international counterparts (typically but not only from the US and Canada) in 25 of the remaining 41 cases (i.e., about 61% of those cases). In significant and heavy cases, therefore, international cooperation is the norm if the merger involves at least one non-EU party.95 This is undoubtedly as it should be, and one wonders if the remaining 16 cases did not involve cooperation because they were not notifiable (or notified) in any other jurisdiction, or whether there are other reasons. Deisenhofer provides a variety of other useful statistical information on both global cooperation and intra-EU referrals which must be skipped over here, and then offers (in section 4 of his paper) some observations regarding the challenges faced by agencies in the age of global business and the latticework of merger control regimes. These include, among others: tight and non-linear statutory review periods that differ in different jurisdictions, which can invite opportunistic behavior (although Deisenhofer later warns that attempts to exploit divergent timetables will be self-defeating); restrictions on the exchange of information among authorities, typically overcome when parties furnish waivers (although third parties may have fewer incentives to do so); and language and cultural barriers, not to mention yawning time lags between, say, Brussels and Ottawa.96 In a context of ever-growing complexity, and with relatively high numbers of mergers processed each year, aren’t conflicts inevitable? It seems they are not. A key finding of the paper is that, of the 27 significant or heavy cases involving international cooperation,97 tensions only boiled over in one case, namely, Oracle/ Sun, and even then the DOJ and DG Comp were able to smooth things over.98 95 This is true a fortiori in cases involving only firms established in third countries. In this context, DG Comp cooperated with a foreign agency in 19 of the 23 relevant cases (83%). 96 For further discussion of obstacles to inter-agency cooperation (in merger control and antitrust), see OECD Secretariat, “Limitations and Constraints to International Co-operation”, DAF/COMP/WP3(2012)8 (12 October 2012), http://search.oecd.org/officialdocuments/ displaydocumentpdf/?cote=DAF/COMP/WP3%282012%298&docLanguage=En. 97 The number 27 includes the 25 cases mentioned above plus two of the 53 cases that involved only parties from the EU. 98 The paper does not make any reference to GE/Honeywell, or to Boeing/McDonnell-Douglas. However, it was recognized at the Workshop, and has equally been recognized by others, that the quasi-acrimony sparked by GE/Honeywell led to redoubled efforts on the part of the European Commission and the US agencies (above all, the DOJ) to achieve common understandings and to minimize the politicization of high profile cases. The rapid development of the ICN and general proliferation of contacts may also have been factors nurturing an esprit of cooperation and serving as

Introduction 

xli  

Deisenhofer compares this paradoxical conflict avoidance rate with the much greater uncertainty faced by a competition agency (or its legal service) walking into court to seek an injunction or to defend a merger decision, and hypothesizes that agencies may be more risk averse in relation to international conflicts than they are when deciding to request a court order to prevent or unwind a merger, or when adopting an administrative decision likely to trigger an appeal. Section 5 of Deisenhofer’s paper is more qualitative. Here he reviews the modalities of cooperation witnessed in several noteworthy “global” cooperation cases (Panasonic/Sanyo, BHP Billiton/Rio Tinto, Oracle/Sun, Microsoft/Yahoo! Search Business and Cisco/Tandberg) and he outlines, more concisely, several referral cases within the EU. In the final section of the paper he concludes that international cooperation is not just advisable on account of the benefits but essential on account of the high costs of non-cooperation. This entails a need for significant investments (which in some cases might in turn pose prioritization choices for a number of competition agencies). It also requires good faith effort, cooperative gestures and appropriate timing of notifications by the merging parties to avoid blindsiding one authority or the other. As if to offer proof that the world’s leading competition authorities are on the same page, Rachel Brandenburger endorsed Deisenhofer’s positive assessment in her oral remarks and underlines the virtues of cooperation in her paper, “Promoting International Convergence: Substantive and Procedural Challenges – the Scope for Comity”. Brandenburger’s very appointment was a gutsy symbolic gesture by the DOJ’s Antitrust Division. Hiring a foreign citizen is not unprecedented for the DOJ: Georgetown’s Heinrich Kronstein once worked for the Antitrust Division (Special Attorney for international/German cartels) under Thurman Arnold, and there may well be other examples. Nevertheless, the US Agencies had never hired anyone fresh off foreign shores for a Deputy-level post; this and comparable hiring practices by certain other agencies worldwide is to be commended and emulated.99 But coming to the paper, Brandenburger emphasizes a number of themes that define many of her public speeches. These begin with one of her primary tasks at the Antitrust Division, namely, to ensure that international cooperation is not left on a partial buffer against conflict. Perhaps this means that agency cooperation is not akin to a “bubble” that bursts on a cyclical basis. For all of the challenges faced by agencies today when handling international matters, the common recognition of a need for dialogue and mutual operational support, all in the light of reputational effects in a “repeated game” and of Deisenhofer’s costs of failure, might be enough to avoid public meltdowns for a good while. However, agencies need to keep the issues of unmanaged public conflict avoidance and informed divergence on their permanent agenda. Furthermore, arrangements should be made with a minimum of delay to bring the Chinese agencies formally into the ICN fold. 99 There are, of course, less politically daring but nevertheless equally commendable ongoing exchange programs (e.g., the DOJ’s Visiting International Enforcer Program) that allow foreign enforcers to take up residency for a while at the host agency. Such programs do not substitute for senior-level cross-border appointments but they are another tool that facilitates mutual learning; a degree of socialization/integration and consequently intermediation skills and better communication; convergence in some areas; and ultimately enhanced conditions for conflict avoidance.

xlii  Merger Control in European and Global Perspective the sidelines for the internationally inclined eccentrics of the antitrust community but integrated into operational and strategic work as a core part of the Division’s thinking. This too can only be commended, although again, as Deisenhofer suggests, it does require material investment. With references to the DOJ’s case practice, Brandenburger’s other themes include the rise of multipolarity; convergence, cooperation and transparency (i.e., the principles endorsed in the ICPAC Report presented to AAG Jim Rill in 2000); and associated ICN-friendly ideas promoted by Christine Varney (while at the DOJ) and articulated more specifically by Brandenburger in various fora, namely, “mindfulness, respect, trust and dialogue”. Mindfulness, of which comity would be a more specific expression, refers first to greater and constant awareness of the potential external impacts of an agency’s actions and decisions (e.g., as regards remedies), possibly imposing costs, and second, to an awareness of the impact on the said agency of the actions and decisions of other agencies. Moreover, as Brandenburger explains, an agency should be mindful of other authorities’ jurisdiction, practices and traditions (and this might extend to mindfulness of the jurisdiction of non-competition authorities with competences that bear on antitrust or merger control). Respect involves openness to the ideas of others, i.e., of other agencies, and respect for the differences between agencies in a very heterogeneous global environment. Respect might be seen as an end in itself or as a condition that facilitates communication and persuasion in the deliberative processes typical of the largely informal sphere of inter-agency contacts. An equally important condition is trust, which according to Brandenburger has two dimensions. The first concerns trust between agencies, which tends to vary with frequency and intensity of interaction and cooperation. Second, since cooperation on cases entails “triangular” interdependencies involving regulated parties whose activities straddle two or more jurisdictions, Brandenburger (like Deisenhofer) stresses the need for trust on the part of the agencies vis-à-vis the parties as a key to effective cooperation. The last ingredient of Brandenburger’s recipe for global good will à la crème is dialogue.100 Importantly, the idea of dialogue “includes a willingness of competition agencies to revisit their own policies and practices over time to reflect new learning and the experiences of others”.101 Dialogue therefore includes communicative exchanges with a view to finding ways to develop better ideas and habits, and ways to build on success. Humility, self-critique, adaptability, a sense of future history and, where appropriate, openness to watershed reform should be seen as hallmarks of top-notch agency culture, and here again it is necessary to think in terms of proactive investment of resources (where feasible) and prioritization. All such 100 This orientation (non-coercive discourse with a view to transcending subjectivity in order to discover superior ideas and solutions) recalls the communicative rationality discussed by scholars of various stripes. See, e.g., Tanja Börzel and Thomas Risse, “The Transformative Power of Europe: The European Union and the Diffusion of Ideas”, KFG Working Paper 1/2009, http://userpage.fuberlin.de/kfgeu/kfgwp/wpseries/WorkingPaperKFG_1.pdf, with references; and Thomas Risse, “Let’s Argue!: Communicative Action in World Politics”, 54 International Organization 1 (2000). 101 Brandenburger, “Promoting International Convergence”, this volume, p. 250.

Introduction 

xliii  

self-discovery and self-improvement is closely linked to what Muris or Kovacic might call an agency’s “research and development and innovation” portfolio. Since I have mentioned Bill Kovacic I might as well turn to his paper, “International Convergence: Assessing the Quality of Horizontal Merger Enforcement”. One can divine the main messages of Kovacic’s papers before reading them, and while that could be a cruel fate for some commentators, here it reflects characteristics determined by a fox/hedgehog double helix: adroit and quick, knowing many things; but also tireless and nocturnal, with a big, unrelenting idea. In his contribution to this volume Kovacic proposes, drawing on his previous work,102 three criteria that might be used to measure how well or poorly a given merger control regime performs.103 The objectives of the essay thus recall those of the more empirically oriented paper delivered by Duso, Gugler and Szücs.104 The latter paper (reviewed above) responds to some extent to Kovacic’s call for more attention to the economic effects of merger control itself, or at least the economic effects of a particular merger regime’s substantive, procedural and institutional configuration. Economic effects, Kovacic points out, involve questions quite different from the typical methodologies used to assess how well agencies are doing. In classic Muris/Kovacic terms, he points out first of all that “[t]o say that an agency is doing a lot of things or only a few things does not tell us whether it is doing the right things”.105 He then presents his evaluative criteria. First, he would ask whether merger policy has improved economic performance by achieving benefits for consumers such as reduced prices, better quality and so on. If the answer is yes, then the policy is yielding the opposite of type I and type II errors: it is approving, reforming and prohibiting the transactions that it should.106 Shifting from a consumer welfare to social welfare perspective and from a single-jurisdiction to global perspective, Kovacic’s second criterion is whether particular merger regimes minimize pointless implementation costs within a given jurisdiction and across jurisdictions. This might be done by shredding red tape and alleviating other burdensome system features, converging toward better and best global investigative techniques and standards, and enhancing trans-system interoperability. The final criterion is another signature preoccupation for the author: it is the question of whether a given regime and its competent agencies take 102 The essay is also partly inspired by Dennis Carlton, “The Need to Measure the Effect of Merger Policy and How to Do It”, 22 Antitrust 39 (Summer 2008). 103 An expanded discussion of regime performance criteria going beyond merger control is provided in William Kovacic, Hugh Hollman and Patricia Grant, “How Does Your Competition Agency Measure Up?”, 7 European Competition Journal 125 (2011); for a concrete application of the six criteria discussed in the latter article, see also Daniel Sokol, Christine Wilson and Joseph Nord, “Grading the Professor: Evaluating Bill Kovacic’s Contributions to Antitrust Engineering” (2012), http://ssrn. com/abstract=2119676. On regime evaluation methodologies in relation to different parameters (accountability, ex post intervention reviews and the “broader impact of competition policy”), see also Peter Ormosi, “Stocktaking on Evaluation”, OECD Discussion Note, DAF/COMP/WP2(2012)5 (22 May 2012), http://search.oecd.org/officialdocuments/displaydocumentpdf/?cote=DAF/COMP/ WP2%282012%295&doclanguage=en, with references. 104 “Merger Policy Evaluation: Where Do We Stand?”, this volume, pp. 105 et seq. 105 Kovacic, “Assessing the Quality”, this volume, pp. 255. 106 Sensibly enough, the attempt to measure enforcement errors forms part of the framework advocated in the paper by Duso, Gugler and Szücs (see above).

xliv  Merger Control in European and Global Perspective seriously the need for permanent cycles of reassessment and improvement.107 A prerequisite for this third criterion, which Kovacic describes in further detail, is an institution’s commitment to self-study, to engage with independent critics and with the private bar, and to participate in multilateral surveillance and benchmarking.108 The three criteria discussed in the paper are also distinguished from less rigorous and possibly useful but also possibly biased or incomplete techniques often used to assess system quality such as practitioner’s surveys and isolated studies of highprofile cases, or strictly “radar screen” measures such as rates of activity and the number of blocked mergers or of other interventions. With greater attention to the pitfalls of current techniques, and with more dedicated investments and budgetary allocations to pursue the diagnostic and self-improvement measures discussed in Kovacic’s essay, agencies will be able to upgrade to “broadband quality”. Adam Fanaki contributed a paper entitled “Convergence in Multi-Jurisdictional Merger Reviews: A Canadian Perspective”, which comments on the convergence process in general but also focuses on Canada’s procedural reforms and in this sense the paper picks up on the themes addressed in Cal Goldman’s paper.109 Fanaki begins by reviewing some of the virtues of convergence between merger control regimes in terms of, for example, a relative reduction of compliance and transaction costs, mitigated risk of divergent outcomes and remedies, and so on. From that perspective, the opposition of the business community to at least some aspects of the 2009 reforms in Canada aimed at enhanced convergence might seem puzzling. However, it is clear from both Goldman’s paper and this one that the controversy amounted to merger regime users saying: be careful what you wish for; Second Requests in the US can entail “long delays and hefty costs”,110 so why not converge toward something at the cutting edge? However, the fact that the US model was chosen as a template is hardly a surprise; as suggested earlier, the only surprise, to an outsider, is that this harmonization exercise came so late in the day. Following an interesting discussion of the “challenges” of multijurisdictional merger review, Fanaki proceeds in parts III and IV of the paper to review the background to and details of those aspects of the Canadian regime that were affected by the harmonization move, highlighting the controversy over the perceived prospects of the higher regulatory costs and burdens of supplementary information requests. His early assessment of the reformed system is in general positive, particularly from the perspective of more efficient agency cooperation on such things as timing, investigative steps, sharing thoughts and information, remedy design and implementation, etc. Several cases are cited with regard to coordinated remedies, the common theme being a willingness to defer to remedies 107 Kovacic’s second and third criteria are not fully captured by the Duso/Gugler/Szücs framework, which suggests (to the extent one accepts Kovacic’s approach) that the latter can be a useful or essential but not exhaustive tool. 108 Kovacic also envisages, at p. 258, “collaborative projects by which competition authorities engage in intensive side-by-side discussions of commonly reviewed completed matters”. 109 Goldman, “Contemporary US-Canada Cross-border Merger Review: Tradeoffs in Policy Objectives between Harmonization, Simplification and Accountability”, pp. 165 et seq. 110 Fanaki, “A Canadian Perspective”, p. 262 (citation omitted). See also ibid., pp. 266-267 (discussing the high relative costs of US review where a Second Request is issued).

Introduction 

xlv  

negotiated elsewhere if the Bureau’s independent investigation confirms that the package is sufficient to resolve the competition problems with respect to Canada.111 In part VI of his paper, Fanaki raises the question of whether, in light of reforms and trends, multijurisdictional merger reviews are less costly than before. The answer is not straightforward. He notes first of all that, “[a]lthough evidence on this point is not currently available, it is likely that the costs to merging parties of the merger review process in Canada have continued to increase following the implementation of the amendments […]”,112 especially in those (proportionately rather few) cases where the parties must respond to supplementary information requests. But he then points out that increasing costs are also due to factors unrelated to the reforms; in this regard, higher costs result not from procedural convergence but from substantive convergence toward more sophisticated and more data-intensive effects-based merger review. Although procedural economies can potentially help to offset the additional costs of more rigorous substantive analysis, Fanaki reserves judgment on whether such economies have been generated by the procedural reforms, calling for continued examination of the issue as more experience accrues under the revised framework. The next four contributions concern merger control in the Far East (Japan, South Korea and China). The first of these is Tadashi Shiraishi, “Effects on Domestic Purchasers: A Descriptive Theory for Competition Law in Cross-Border Cases”. The interest of this paper goes beyond merger control and addresses the more horizontal theme of the assertion of extraterritorial jurisdiction where business activity has global dimensions, whether in the field of antitrust or merger control. The paper begins with the observation that, although the “effects” basis for exercising jurisdiction is often discussed, the nature of the doctrine remains imprecise. One may refer, for example, to substantial and reasonably foreseeable effects within the territory of the State or region in which the authority concerned is competent to act – but what are the principles that govern the application of this effects test? Shiraishi suggests – as a descriptive matter – that the relevant effects in this context are effects on domestic purchasers.113 From this point of view, the location of suppliers is immaterial. With this straightforward descriptive tool in hand, Shiraishi proceeds literally to illustrate several factual scenarios from international cartel and merger control cases and shows how his “effects on domestic purchasers” (eDP) theory can explain the assertion of jurisdiction or, in some cases (particularly with regard to the Japan Fair Trade Commission), the non-assertion of jurisdiction. Shiraishi begins by pointing out that, despite some katana-rattling in the early 1990s by the Department of Justice, and the purported renunciation of footnote 159 contained in the 1988 version of the international operations 111  On coordinated remedies and willingness to defer, see also Sven Völcker, “Dare to Defer? Towards Greater Procedural Efficiency in Multijurisdictional Merger Remedies”, this volume, pp. 133 et seq. 112  Fanaki, “A Canadian Perspective”, p. 272. 113  This suggestion logically excludes jurisdiction where, in particular, domestic sellers suffer anticompetitive harm due to market access impediments in a foreign jurisdiction.

xlvi  Merger Control in European and Global Perspective guidelines,114 the DOJ did not in fact pursue, vis-à-vis Japan, pure foreign market access theories of harm where there was no evidence of harm to US consumers (i.e., purchasers).115 This left the eDP theory intact. Several other examples are then discussed, including the Empagran case, where foreign (i.e., non-US) cartel victims sought, understandably, to avail themselves of the US courts; and the Marine Hoses cartel, in relation to which Shiraishi notes the quite different result yielded by Japan’s surcharge system (kachokin) compared to that of the European Commission’s Fining Guidelines but then shows how these contrasting results can both be consistent with the eDP theory. Two further applications of the theory involve the issue of whether, in the context of certain merger cases, the JFTC defines geographic markets as Japan-only markets or global markets. The scope of geographic markets has been controversial in Japan, but the JFTC has generally preferred, as Shiraishi points out, to delineate Japan-only (or occasionally regional) markets – a reflection of the objective of protecting purchasers in Japan.116 Shiraishi concludes his paper with a different kind of scenario to show where the eDP theory breaks down. As he explains, the theory presupposes that the location of purchasers can be identified. But some purchasers move across borders. Some engage in “borderless business”. The example cited by Shiraishi is the Cathode Ray Tubes cartel, where purchasers were located outside Japan in Southeast Asia but the corporate parents of these purchasers were situated in Japan. The purchasers were in one sense foreign but in another sense domestic (since the parent companies in Japan sometimes negotiated deals with the Japanese suppliers, who were secretly colluding, and directed their Southeast Asian subsidiaries to buy the cartelized inputs). In a “frontier” case of this kind – which can be transposed to the merger sphere mutatis mutandis – Shiraishi suggests that one must choose between a literal application of the eDP theory (which might conceivably lead to a regulatory gap, depending on the circumstances) and an extended version of it which presumably would broadly construe the concept of a “purchaser”. Next comes my awkward and incentive-incompatible summary of the short paper by Etsuko Kameoka and Mel Marquis, entitled “Recent Developments in Japan’s Merger Control System”. The formulation and application of 114 See, e.g., Marina Lao, “Jurisdictional Reach of the U.S. Antitrust Laws: Yokosuka and Yokota, and ‘Footnote 159’ Scenarios”, 46 Rutgers Law Review 821 (1994). 115 Gradual action to strengthen domestic Japanese antitrust enforcement, following the Strategic Impediment Initiative (SII) talks (encouraged to some extent by more weapon-rattling, but on the part of Congress and the US Trade Representative), likely served as a pressure valve to ease (not to say eliminate) tensions in the early 1990s between the US and Japan. There is a small mountain of journal articles and essays regarding the trade frictions between these countries, the SSI talks and the consequences for Japanese competition law and policy. See, e.g., Michael Beeman, Public Policy and Economic Competition in Japan: Change and Continuity in Antimonopoly Policy, 1973-1995, Nissan Institute/Routledge, 2002, chapter 8. 116 In the rarer body of cases where the JFTC has identified global markets, each supplier has generally charged a uniform price irrespective of where the product is sold globally; in this scenario the global market serves as a proxy for the Japanese market, and effects on domestic purchasers are thus taken into account even though the substantive analysis examines the impact of a merger on the global market.

Introduction 

xlvii  

antimonopoly policy in Japan is marked by permanent incremental reform that adds up to transformative change when longer time periods are considered. In this paper, however, the authors focus on just a thin slice of time to report the refinements of the JFTC’s merger procedures in the summer of 2011. Going back over the decades, the German model was a source of some degree of inspiration for the JFTC’s rules and procedures (and indeed for Japan’s entire underlying administrative law infrastructure); and, more commonly recognized, dialogue with United States federal authorities (in the shadow, in those pre-WTO days, of the infamous Section 301, the trade retaliation statute117) has triggered evolutions in the JFTC’s rules, procedures and enforcement posture. The authors here suggest that the European Commission seems to have had some influence as well, but they could also surely attribute the most recent reforms to (apart from the Government’s growth strategies) the intensified global antitrust conversation of the last decade.118 Ultimately, the unspoken motivator in Japan is: let us do whatever will help to resuscitate the pulseless economy.119 In the paper the authors describe, first of all, how the JFTC’s merger review procedures have been altered. For example, new discipline has been imposed to rein in the “prior consultation” system, which was not very transparent and which at least potentially warped the JFTC’s timetables. Although the old system has been abolished, it remains possible, as is routine in many other jurisdictions,120 117 See, e.g., Mitsuo Matsushita, “Structural Impediments Initiative: An Example of Bilateral Trade Negotiation”, 12 Michigan Journal of International Law 436, 440 and 445-446 (1991); Industrial Structure Council, Uruguay Round Committee, Subcommittee on Unfair Trade Policies, Complete Report on Unfair Trade Policies by Japan’s Major Trading Partners, Nova Science Publishers, 1993. 118 The JFTC certainly understands itself to be a leader in the global antitrust community (with active participation, for example, in the ICN’s Merger Working Group), and this is so above all in Asian/APEC circles, where the JFTC routinely conducts training events and dialogues (e.g., with China, Indonesia and Thailand, among others). This self-conception is hardly surprising given that Japan’s grandfatherly antimonopoly framework has been in place for more than six decades. The desire to lead is also linked to a certain degree of institutional isomorphism: to be an outlier is to risk irrelevance, whereas mingling with the mainstream presents certain empowerment advantages. Commentators have noted, for example, that ICN-validated projects and trends carry weight back home, and two-level or multi-level games may be played. 119 On this motivation, the interests of the JFTC and METI converge. Although the authors catch the scent of some possible degree of reversion to an industrial policy orientation in Japan, the context is quite different today compared to Japan’s industrial policy heyday. With the progressive abolition (mid-90s to 2003) of a stunning variety of exemptions from the Antimonopoly Act (AMA), industrial policy today must be conducted with a more horizontal, pan-industrial orientation than the bad old days of vertical (sector specific, discrimination-based, cartel-friendly) industrial policy. Studies of Japan’s industrial policy and the clashes between the then-MITI and the JFTC abound, but for an introduction see Mitsuo Matsushita, “The Intersection of Industrial Policy and Competition: The Japanese Experience”, 72 Chicago-Kent Law Review 477 (1996). For a summary of the radical reduction of exemptions from the AMA, see Michael Wise, OECD Country Study: Japan – The Role of Competition Policy in Regulatory Reform (1999), http:// www.oecd.org/regreform/liberalisationandcompetitioninterventioninregulatedsectors/2497259. pdf, pp. 31-39. By 2004, the OECD Competition Committee opined that “[t]he process of eliminating and narrowing exemptions has been substantially completed”. Monitoring Review of Japan (15 July 2004), DAFFE/COMP(2004)11/REV1, http://www.oecd.org/regreform/ liberalisationandcompetitioninterventioninregulatedsectors/34728075.pdf, at p. 14. 120 For a critical view of the pre-notification préliminaires between would-be merging parties and DG Comp’s case teams, see Bellis, Elliott and Van Acker, “Ten Areas”, cited above note 35, at pp. 327-328; and cf. the oral remarks of John Fingleton, p. 204.

xlviii  Merger Control in European and Global Perspective to engage in pre-notification contacts with the JFTC – contacts which might be particularly important depending on the character of a given merger.121 Other reforms include transparency measures designed to help the merging parties understand the precise concerns of the regulator, and more detailed explanations as to the framework employed by the JFTC when delineating between Japanonly, regional and worldwide geographic markets (the background to which is discussed in Professor Shiraishi’s paper). In a subsequent section of the paper the authors report on three mergers reviewed by the JFTC with reference to the recent reforms, namely, the NSC/Sumitomo Metal and the hard disc drive (HDD) cases.122 To the extent that international convergence in merger procedures is a good thing,123 the JFTC appears to be making its contribution to this process. Seonhoong Jeon reports on South Korea’s merger review apparatus in his contribution, “International Convergence and Recent Korean Experiences in Merger Control”. From this paper a nice picture emerges of another merger system in transition. Jeon focuses on two convergence-related dimensions of this transition. First, he notes a trend toward greater use of state-of-the-art economic thinking about mergers. This apparent substantive turn presents a challenge to conventional merger control habits in Korea, which have depended to a large extent on market shares and concentration indicators. Indeed, until recently, difficult-to-rebut structural presumptions based on these criteria have generally been decisive. But the global trend toward predicting likely post-merger effects in this field is now also having an effect in Korea.124 Jeon cites two examples of this phenomenon. In the TDK/Alps merger of 2007, despite post-merger shares of more than 50%, the transaction was approved unconditionally – most likely, according to Jeon, on account of countervailing pressure from strong buyers (Western Digital, Hitachi Global Storage, Fujitsu, Samsung and Toshiba). In the 121 Such pre-notification contacts may at least partially alleviate concerns about the possible impact of the reform on the coordination of remedies. See Ilene Knable Gotts, “Preface”, in Knable Gotts, ed., The Merger Control Review, 3rd edition, Law Business Research Ltd, 2012, at p. viii (“When combined with the inability to ‘stop the clock’ on the review periods, counsel may find it more challenging in transactions involving multiple filings to avoid the potential for the entry of conflicting remedies or even a prohibition decision at the end of a JFTC review.”). 122 These cases are also discussed in Yusuke Nakano, Vassili Moussis and Tatsuo Yamashima, “Japan”, in The Merger Control Review, 3rd edition, cited previous footnote. 123 As already seen in footnote 7, “blind” convergence for its own sake adds no value; “informed” convergence in merger control should proceed on the basis of identified net benefits and net avoidance of costs, and taking account of good reasons for diversity. If it is accepted, on the basis of such reflection, that convergence on substantive standards is desirable, then procedural convergence (apart from any independent benefits it may have) is likely to be useful as well, insofar as divergent procedures may impede consistent outcomes in substance. Cf. Feinstein, “Process Divergence”, cited above note 92. 124 Such a trend can only be reinforced with the roll-out of the EU-Korea Free Trade Agreement, which entered into force on 1 July 2011, and its competition chapter (Chapter 11). According to Article 11(6)(2) of the Agreement, the Parties “shall cooperate in relation to their respective enforcement policies and in the enforcement of their respective competition laws, including through enforcement cooperation, notification, consultation and exchange of non-confidential information based on [the EU/Korea bilateral cooperation agreement of 23 May 2009]”. On a less formal level, the KFTC and DG Comp cooperate in accordance with their Memorandum of Understanding on Cooperation between the Fair Trade Commission of the Republic of Korea and the Competition Directorate-General of the European Commission (October 2004), http://ec.europa.eu/competition/international/legislation/korea.pdf.

Introduction 

xlix  

eBay/G market two-sided market merger of 2009, concerns were identified on the seller-side of the market, as e-Bay and its affiliate company Auction held a combined share of 87.5% of the open market by transaction volume. However, the KFTC allowed the merger to proceed, subject to temporary behavioral remedies. As Jeon explains, the possibility of price or output effects due to the merger was sufficiently diminished by prospective new entry and product repositioning, presumably by virtue of negotiable market barriers. Jeon singles out this case in particular as reflecting a significant shift in analytical approach. The second trend highlighted by Jeon is the active cooperation between the KFTC and other enforcers with regard to multijurisdictional transactions. The implicit point here is that cooperation and coordination among authorities can promote (and be promoted by) international convergence, to some degree as concerns procedure and especially in terms of substantive outcomes. In this regard, Jeon again cites two cases. First, in the TDK/Alps case, cooperation between the KFTC and JFTC seemed likely, since TDK and Alps were both Japanese companies and since both the KFTC and JFTC cleared the merger free of conditions within a month of each other. Second, in the 2010 BHP Billiton/ Rio Tinto production joint venture (treated as a 3-to-2 merger in the market for worldwide seaborne lump ore), Jeon notes that international coordination played a “very important” role. The JV-merger was abandoned in October 2010 when it became clear that several competition agencies were anxious about both unilateral and coordinated effects likely to result from the parties’ combined operations. With regard to this case, the KFTC conducted its examination in close cooperation with the JFTC and other authorities, exchanging information and coordinating “comprehensively” on possible remedies. A volume on “global” merger control issues would not be very global without devoting at least some modest attention to China, and to the rather intense merger work of MOFCOM’s Anti-Monopoly Bureau. Xinzhu Zhang and Vanessa Yanhua Zhang have ably monitored and reported on MOFCOM’s merger activities from the time the Antimonopoly Law entered into force in August of 2008. They provide an updated contribution here entitled: “China’s Merger Control Policy: A Three-Year Milestone”. The paper takes account of the eleven concentrations between 2008 and December 2011 in which MOFCOM’s Anti-Monopoly Bureau intervened in order to impose remedies or, in the oftdiscussed Huiyuan Juice case of 2009,125 to prohibit the transaction.126 This group 125 The would-be acquirer in that (2.4 billion-dollar) case was Coca-Coca, who owned, among other assets, Minute Maid orange juice. For discussion, see inter alia Wang and Su, “China’s AntiMonopoly Law”, cited above note 92, at pp. 387-391). Notably, at p. 388, Wang and Su state: “based on the information disclosed during the MOFCOM Q&A session and our discussions with the case handlers, our observation is that although the lack of detail resulted in ambiguity, the Coca-Cola/ Huiyuan decision is mainly based on the protection of competition and consumer choice and on careful competition and economic analysis.” Nevertheless, these authors call for more transparency in MOFCOM’s decisions (see ibid., pp. 390-391) in order to avoid fuelling suspicions about hidden (protectionist) motives. 126 For the year 2011, see also, e.g., Jim O’Connell, “The Year of the Metal Rabbit: Antitrust Enforcement in China in 2011”, Antitrust 65 (Spring 2012), at pp. 68-72. For further recent discussion and a general appraisal of Chinese merger control, see Ping Lin and Jingjing Zhao, “Merger Control

l  Merger Control in European and Global Perspective of a dozen cases should be seen in the light of the number of cases processed each year, which despite globally stalled merger activity has been ramping up steadily (17 cases in 2008, 80 in 2009, 117 in 2010 and 168 in 2011).127 Although these numbers suggest an appreciable degree of “buy in” by users,128 the quality of MOFCOM’s decisions remains difficult to assess with much precision due to the current transparency deficit and the all-too-concise reasons given for interventions. Notwithstanding this data limitation, Zhang and Zhang provide a helpful overall view of the functioning of the system, commenting first of all on the rules and guidelines MOFCOM uses when assessing a merger (section II) before discussing the case practice (section III) and finally outlining what they like to call “patterns and implications” (section IV). Beginning with the rules and guidelines, the authors consider two key documents: the Anti-Monopoly Commission’s “Market Definition Guidelines” of 2009 (which cover more than market definition), and MOFCOM’s “Interim Rules on the Assessment of Competitive Impacts of Concentrations of Undertakings” of 2011. The Guidelines and the Interim Rules are criticized for different reasons, but in each case the authors suggest that further elaboration and specificity is required, as in competitive analysis the value of general maxims and broad principles only goes so far. The author’s discussion of MOFCOM’s decisional practice covers both horizontal and non-horizontal mergers. With regard to horizontal mergers, MOFCOM’s analysis is not very dissimilar to EU and US standards, and it takes account of unilateral and coordinated effects, although as Zhang and Zhang point out, the Interim Rules do not explain how assessments of these distinct types of effects differ. As economists, they also express concern at MOFCOM’s heavy reliance on qualitative economic evidence (in particular, substitutability analysis) to the apparent exclusion of quantitative economics when defining markets. There is also too much uncertainty, we are Policy under China’s Anti-Monopoly Law”, 41 Review of Industrial Organization 109 (2012); Yee Wah Chin, “The High-Wire Balancing Act of Merger Control under China’s Anti-Monopoly Law” (August 2012), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2120280; Dan Wei, “China’s Anti-Monopoly Law and its Merger Enforcement: Convergence and Flexibility, 14 Journal of International Economic Law 807 (2011). 127 By the summer of 2012, MOFCOM had reviewed over 450 merger cases under the AML. Subsequent to the time period covered in Zhang and Zhang’s paper, MOFCOM intervened in: the Tiande/Henkel joint venture in the chemicals sector (February 2012; approval with remedies); Google/ Motorola affecting markets for smart mobile devices and operating systems (May 2012; idem); and Wal-Mart/Newheight which affected online retail sales and certain telecommunications services (August 2012; idem). Among these cases, Google/Motorola was noteworthy in that, as in GM/Delphi and Seagate/Samsung (both discussed by the authors), MOFCOM insisted on (five-year) behavioral remedies even though the transaction had already been cleared unconditionally in both the US and the EU as well as other jurisdictions. 128 “Sales” are compulsory if the relevant thresholds are met. On 5 January 2012, MOFCOM adopted regulations that envisage sanctions for companies that fail to notify transactions under the Anti-Monopoly Law. Under the regulations, non-compliant companies may be fined up to RMB 500,000 (i.e., around 61,000 euros) and, potentially more serious than the fine, may be forced to divest assets to restore the status quo ante. For a flow chart of the notification and review process, see page 3 of Herbert Smith, “Merger Control: China” (June 2011), http://www.herbertsmith.com/NR/ rdonlyres/0150AA9B-7A7D-40A2-BB22-5E4330A73375/22810/Merger_Control.PDF.

Introduction 

li  

told, about what exact impact market shares and concentration indicators have in determining market power; how rebuttal evidence is assessed; and whether, among other things, safe harbors apply and how wide or narrow they are. Other critiques emerge as the authors discuss MOFCOM’s decisions, but in the absence of more details in the published decisions the overriding concern may be the unfortunate need to engage in guesswork and to speculate about MOFCOM’s theory of the case. Presumably, over time – and assuming China does not, for the foreseeable future, reshuffle and rationalize its trio agency structure – MOFCOM may be expected to grow increasingly confident in this regard and may ultimately open its hefu (I’m told this is Chinese for kimono) a bit wider. MOFCOM’s relatively close scrutiny of non-horizontal mergers is worth noting. Providing case examples, Zhang and Zhang observe that MOFCOM has in fact focused on vertical foreclosure and on portfolio effects in several of the cases where it has intervened. To some degree a certain affinity may thus be perceived with merger assessment in the EU, where non-horizontal effects are treated rather seriously.129 However, the authors are also critical, on robustness and post-merger incentive grounds, of how MOFCOM has executed its examination of input foreclosure in particular cases (specifically, in GM/Delphi and in the GE/Shenhua joint venture case). MOFCOM’s loose analysis is then used as an illustration of the need to embrace a more rigorous evidentiary analysis, i.e., to adopt a higher standard of proof when assessing vertical and conglomerate mergers. Doing so would be good practice in any event but it would also align the Chinese approach better with the standard in Europe, where the European Courts have insisted on particularly persuasive evidence when the Commission intervenes in non-horizontal cases.130 Overall, it is perhaps no surprise to see that the authors detect the influence of the US and EU merger systems but also the signs of indigenous “characteristics”. Some commentators have started to call this eclectic blend a fusion-style approach to merger analysis, as one finds in art or food or music, although a catch phrase such as this may pose difficulties in certain European languages (since fusion means “merger”). Nor is it surprising to see Zhang and Zhang praising and criticizing MOFCOM and China’s merger regime at the same time. The half-full, half-empty appraisal of how far China has progressed in such a short time (in the field of merger control, since 2003 and in the field of restrictive agreements and abuse of dominance, since 2008) reflects both the youth of and unique challenges faced by a (qualified) competition regime131 With regard to trends in the US and a hint of convergence, see above note 91. See Venit, “The Scope of EU Judicial Review of Commission Merger Decisions”, this volume, pp. 113 et seq. 131 Many commentators have noted that MOFCOM has basically (questions of correct application aside) stuck to competition criteria in its substantive merger practice. On the other hand, this regime is “qualified” in the sense that the potential for political interference and for non-competition issues to determine outcomes is something that may be inherent in China’s present legal and political framework. For example, Article 27 of the AML obliges the relevant competition authority to consider, inter alia, “the effect of [the transaction] on the development of the national economy” when deciding whether to approve a merger. There is also the matter of national security review (again, a potential inroad for political decision making, especially where there is tit for tat with the US) and many other 129 130

lii  Merger Control in European and Global Perspective superimposed on a largely State-run economy with high levels of hybridity and ambiguity.132 Not all the BRICS are represented in this volume (even if there have been plenty of recent newsworthy developments in India and Russia133), but Abel Mateus delivered a useful and succinct examination of “The New Brazilian Merger Control Regime”, which entered into force on 29 May 2012. Distinctive “characteristics” have also long defined the Brazilian regime, particularly in its eccentric institutional and procedural dimensions, not always universally admired;134 but now these characteristics have been substantially transformed by Law 12529 of 30 November 2011. Many of the key points in this discussion of the reforms are summarized in the first paragraph of the paper. First, Brazil has to a large extent rationalized the institutional design of the merger review process, whereby full responsibility for merger investigations is consolidated and resides with the Conselho Administrativo de Defesa Econômica (CADE),135 now in need of additional case handlers. As Mateus points out, this rather bold institutional reform – which implies that only two national agencies (and for core work, just one) will handle the application of Brazil’s competition law – responds to longstanding criticisms about overlapping agency turf and its consequences. Second, Law 12529 inverts the former Brazilian regime, which allowed parties to close transactions prior to the conclusion of merger investigations, and establishes a pre-notification regime with related accoutrement such as gunjumping penalties and, for non-notifiable mergers, a one-year “clawback” features of the AML and of China’s political structure could be discussed in this context. Ping Lin and Jingjing Zhao have argued that China should adopt a policy of giving “lexicographical” priority to competition-oriented analysis and that it should only bring in industrial policy considerations, if at all, in relation to competition-neutral transactions. Lin and Zhao, “Merger Control Policy”, cited above note 126, at pp. 128-131. 132 For recent discussion of a range of subjects regarding Chinese competition law (including merger control), see Michael Faure and Xinzhu Zhang, eds., Regulation and Competition Policy in China: New Developments and Empirical Evidence, Edward Elgar, forthcoming. 133 India’s new merger control regime entered into force in June 2011. See Avinash Sharma, “Merger Control under India’s New Competition Law: A Comparative Perspective”, 32 European Competition Law Review 602 (2011) (more generally, see also Mark Sansom and Priya Christian, “India’s New Competition Regime: The Elephant Who Became a Tiger”, 1 Journal of European Competition Law and Practice 62 (2010)). In Russia, the Third Antimonopoly Reform Package took effect in January 2012. For an early summary, see Mathieu Fabre-Magnan, Liza Ivanova and Valeria Ponomareva, “Russian Federation”, PLI, Multi-Jurisdictional Guide 2012, www. practicallaw.com/6-504-6542, pp. 1-5. For further discussion of the BRICS, see Terry Calvani and Karen Alderman, “BRIC in the International Merger Review Edifice”, 43 Cornell International Law Journal 73 (2010). 134 The main concerns were noted in a 2005 OECD Peer Review (http://www.oecd.org/daf/ competition/prosecutionandlawenforcement/35445196.pdf). See now also (covering, among other topics, merger policy and the institutional structure of the agencies), John Clark (OECD Rapporteur), Competition Law and Policy in Brazil (2010), http://www.oecd.org/daf/competition/45154362.pdf. For extended discussion of the former system, see Marco Botta, Merger Control Regimes in Emerging Economies: A Case Study on Brazil and Argentina, Wolters Kluwer, 2011. 135 The other surviving competition authority in Brazil is the Secretary of Economic Monitoring, i.e., the SEAE (a part of the Ministry of Finance). Following the reforms, the SEAE does not function as a true enforcer but becomes a special organ that carries out competition advocacy work vis-à-vis Brazil’s regulatory authorities and other units of government.

Introduction 

liii  

clause (i.e., a limited possibility to unscramble the parties’ eggs).136 With regard to notification thresholds, Brazil has also abolished the use of deviant market share-based criteria.137 Parties may now rely solely on turnover triggers (which now include minimum turnover for the second party) to determine whether their “concentration act” must be notified to CADE. A joint decision in 2012 by the Ministries of Justice and Finance has provided further specifications as concerns the relevant turnover conditions.138 Overall, Mateus provides a chiaroscuro appraisal of the new system. On the one hand, he notes that Law 12529 “brings the legal regime largely in line with OECD and ICN guidelines and recommendations”.139 Yet the Brazilian legislator could not relinquish certain tokens of exceptionalism. Potentially, for example, investigations may languish for long periods (with a waiting period of 240+60 or 240+90 days), and while under the old regime the duration of the review had limited impact since merging parties could consummate before the authorities finished their work, dilatory inquiries will now impose more costs and risks.140 Other potential problems may result from the lack of a clear two-stage review (i.e., the odd lack of an initial 25-30 day waiting period), where most untroubling transactions can be expedited while genuine risks to competition can be filtered into an in-depth investigation.141 Among other subjects, Mateus proceeds to discuss a number of relevant procedural aspects (such as CADE’s internal administrative investigation/adjudication procedures) and substantive details (pointing out the influence not just of the EU’s substantive standard under Regulation 139/2004 but of Article 101(3) as well). With these significant reforms the expunged bath water did not abscond with the baby, but the grimy tub requires further scrubbing. Flaws in the legal framework might be compensated for, to some degree, if case teams and management work efficiently and concentrate their efforts on nexusbased cases; this would be a positive turn away from the days when Brazil attracted notification of many foreign-to-foreign transactions of dubious import for Brazilian consumers. 136 Ironically, the flexibility that used to prevail as far as timing was concerned has been superseded, in the new regime, by a new apparent rigidity: namely, the obligation to notify within 15 days of the signing of an agreement. Such obligations are becoming increasingly anachronistic, largely because they potentially complicate the coordination of multijurisdictional filings. 137 Worldwide, some of the countries with merger review systems that still hinge to some extent on market shares include, among others, Columbia, Lithuania, Morocco, Portugal (even after its substantial 2012 reform), Spain, Ukraine and the UK. 138 For the thresholds and various other details, such as the (broad) definitions of the terms “corporate group” and “concentration act”, see Fiona Schaeffer and Michael Culhane Harper, “A Fundamental Shift: Brazil’s New Merger Control Regime and its Likely Impact on Cross-Border Transactions”, Antitrust Source (August 2012). 139 Mateus, “The New Brazilian Merger Control Regime”, this volume, p. 315. 140 In this regard, moreover, Law 12529 did not provide for automatic approval of transactions once the statutory review period expires. This may distort incentives to some morally hazardous extent, and may generate additional uncertainty. 141 Superintendent General Carlos Ragazzo has felt obliged to provide public assurances in this respect, indicating that “simple” cases would be reviewed in less than 30 days, and that the 240-day limit would be reached, if ever, only in a case of extraordinary complexity. See Schaeffer and Culhane Harper, “Fundamental Shift”, cited above note 138, at p. 7, footnote 33.

liv  Merger Control in European and Global Perspective To close this long section 4 let us modulate to a minor key, as it were – leaving Europe, Asia and the Americas behind to consider not just a different part of the world but also an area that features countries smaller and often less resourced than those we have considered thus far. Maher Dabbah kindly obliges with his paper, “Merger Control in Middle Eastern Countries: A Perspective on Challenges and Opportunities”. The countries belonging to this geographic group (including, for example, Egypt, Jordan, Morocco, Yemen, Tunisia and Turkey) are heterogeneous, but some exhibit strong features of state involvement in the economy.142 Where this is so, merger policy is unlikely to be seen in a “pure” form. Either the rules themselves will allow for political intervention or, if not, the gap between rules and practice may be wide. In any case, some Middle Eastern countries seem to provide good examples of merger control systems on the outskirts of the ICN community. ICN Recommended Practices have not frequently been designed for these economies, or by them. The distinctive political context of many Middle Eastern countries helps to explain why in some areas the ICN cannot progress at a uniform rate but must move “pluralistically” at multiple speeds. It is therefore understandable that Dabbah observes that Middle Eastern countries, taken as whole, have not yet succeeded – despite some modest progress – in establishing effective merger control. It is this mixed success that leads him to adopt a “challenges” and “opportunities” frame of analysis, which may be a normative way of saying that more principled merger control will not go down easy in many of these countries, but that it is worth the effort. Among the challenges are the need to choose rational and balanced notification thresholds and the need, among some countries (e.g., Jordan and Morocco) to renounce thresholds based on market share. It follows from the considerations above that another fundamental challenge in the Middle East is to ensure that the merger control authority, and as the case may be the courts reviewing the authority’s decisions, have solid institutional and budgetary independence and cannot be unduly swayed by the government or regulators or business. Another challenge is that of national legislators, who will have to face up to the need to progressively (or more resolutely) eliminate unjustified sectoral exclusions and the hangovers of bacchanalian rent-seeking. As for opportunities, Dabbah points out that merger control allows the competent authority to help determine how competitive markets are, and though he does not say so explicitly he seems to assume that an effective merger regime produces benefits (all else equal) for both consumers and society in general. Such opportunities and benefits may be more accessible if today’s global fora (e.g., the ICN, the OECD Global Forum and regional agency networks) intensify general trends toward greater inclusiveness, regular dialogue and constructive debate. 142 A gradual and qualified retreat of the State, owing to factors such as WTO obligations, bilateral and regional trade agreements and strategic economic policy, is observed in certain countries such as Jordan. See Abdullah Nawafleh, “The Contribution of Business Law Reform to Economic Development: Lessons from the Middle East”, [2012] European Business Law Review 309.

Introduction 

lv  

5. Merger control and best practices Contributions by Boyce, Burrichter/Zandt, Drauz This last session was largely inspired by the proliferation of “best practice” and “recommended practice” documents to guide competition agencies and notifying parties through the merger review process in an effective, efficient manner. John Boyce is the first author here with the musical title of “Best Practice in Merger Control: It Ain’t What you Do It’s the Way That You Do It … And That’s What Gets Results”. Boyce begins by providing several definitions of best practice and then offers reasons why the best practice concept can be valuable in the context of merger control. In his view these relate, in essence, to the utility of best practices in promoting a common understanding between a reviewing agency, the merging parties and, in many cases, sister competition authorities. Common “nodes” of understanding can ideally reduce transaction costs and enforcement costs, and can help minimize the potential for conflict. Boyce then selects three general areas in which, he suggests, further harmonization of best practice standards are desirable. Specifically, these are: (i) best practice to facilitate the merging parties’ “preparatory work” (i.e., at the stage before pre-notification contacts have begun; relevant here is the relationship between strategic decisions and the rules governing potential trigger jurisdictions and the related practice); (ii) best practice for the conduct of merger investigations by reviewing agencies; and (iii) best practice when (draft) remedies are designed, pitched to the agencies and beefed up or refined. As regards the pre-notification stage, Boyce underlines the importance on the parties’ side of taking account of all the relevant risks and costs associated with applicable merger filings and review. This best practice for merging parties should be matched, he says, by a strong commitment to transparency and convergence toward internationally accepted best practices. The emphasis seems to be on how agencies can reduce unnecessary search and information costs, regulatory uncertainty and risk; and Boyce provides specific suggestions to that effect.143 In the second stage, i.e., during the pendency of the agency’s (agencies’) investigation, there are numerous guidelines, most of them by now incorporated in the ICN’s Recommended Practices, which agencies are encouraged to follow in their merger control practice in order, once again, to reduce unnecessary costs for business, although many of them can also enhance the efficiency of the agencies themselves. Many of these guidelines (some of which are discussed in other contributions as well, including in particular Ian Forrester’s paper – see above) are a matter of common sense in a globalized context and are aimed at minimizing potential negative externalities. With reference to these streamlining measures Boyce once again sets forth a series of specific suggestions, and points out where the merging parties too must assume responsibility.144 A few 143 144

See pp. 346-347, this volume. See pp. 347-349.

lvi  Merger Control in European and Global Perspective final observations concern the third phase, remedies. Here Boyce briefly outlines strategies for developing remedies that are coordinated and consistent across jurisdictions and tailored to address relevant competition concerns. As an adjunct to his paper, Boyce provides three helpful annexes. Annex 1 (pp. 350-352) itemizes, in chronological order, notable guidelines from diverse sources (ICN, OECD, FTC, OFT, etc.) which cover various aspects of the merger review process as well as substantive merger analysis. Annex 2 (pp. 353-356) lists the ICN’s Recommended Practices for Merger Notification Procedures.145 And Annex 3 (pp. 356-359) contains a set of best practices used internally by the author’s firm which may assist companies and their advisors where merger approval is sought from agencies in multiple jurisdictions. Jochen Burrichter and Manuel Zandt’s paper, “Merger Control and Best Practice”, provides a three-part taxonomy of best practice documents, highlights the substantial variety and lacunae in “best practice” practice in key regions (EU, North American and Asia/Oceana), and evaluates their benefits and usefulness. According to the authors, most best practice (and recommended practice) documents can be divided into “external” and “internal” categories, depending on whether they are internally generated by a competition authority with reference to the relevant laws, policies and practice or whether, by contrast, they are produced externally – typically by an international organization such as the ICN or the OECD, or occasionally by private interests (e.g., BIAC or the ICC).146 The external and internal categories both promote best practices that should be adopted or pursued by competition authorities, but internal best practices feature a higher degree of specificity with regard to particular merger control systems. A third category consists of best practices in the sense of recommendations for merging parties. The authors note that only one serious set of guidelines of this kind has been promoted (namely by the private-initiative Merger Streamlining Group). One can hardly ignore the glut of practice guides published each year, some but not all freely available, which provide some practical advice to help merging companies to manage filings in at least the major jurisdictions, BRIC countries and various other strategic systems (Mexico, South Africa, Turkey, etc.). However, such guides may be only partial substitutes for a more nuanced and/or (if issued by a competition agency) authoritative document. Part II of the paper maps out existing best practice documents.147 The emphasis is on Europe and North America but reference is also made to Singapore, Australia 145 See also the corresponding Implementation Report (April 2005), http://www. internationalcompetitionnetwork.org/uploads/library/doc324.pdf. 146 Conceptually, bilaterally agreed best practices (notably, the best practice document adopted by the US-EU Merger Working Group on cooperation in merger investigations) belong to the “external” category because they do not address to merging parties the way various procedures or process elements will be handled but rather make transparent the principles on which inter-agency cooperation is conducted. 147 Again, the various documents can be seen systematically in Annex 1 to John Boyce’s paper (see main text above). However, unlike Boyce, Burrichter and Zandt exclude guidelines on substantive analysis from their definition of best practices. From their perspective, best practices relate fundamentally to procedural and process issues. Many competition authorities separate these dimensions in their published guidance as well, although a few do not.

Introduction 

lvii  

and Japan, the latter for the absence so far of any best practices document as defined by the authors, i.e., a document focusing on procedures, process and due process, as opposed to substance.148 The authors regret to find that agencies worldwide – including some of the EU’s NCAs – have issued rather few internal best practices, or none at all, although many have issued guidelines on substantive analysis. Part III discusses the scope of the currently available best practice documents and here one sees considerable variety across jurisdictions, but a point of commonality is that comprehensive guidance is quite rare, and where best practices exist they are seldom in consolidated form. There is as yet no international best practice standard about the appropriate structure, content and level of specificity of best practices. Having established that the overall current state of affairs is disappointing from the perspective of transparency and predictability, Burrichter and Zandt proceed in Part IV to outline the aims and benefits of best practice documents. With regard to internal best practices, it stands to reason that the ultimate objective should be the promotion of effectiveness of merger control regimes at reasonably low social and private cost (efficiency) while respecting due process. Although the specific utility of internal best practices is somewhat different for agencies and for business, they each share a common interest in having in place a set of best practices that the other finds useful. To the extent that convergence is desirable (a subject discussed earlier in this chapter), there are also benefits to be gained from aligning internal practice with the generally accepted best standards agreed at the level of, in particular, the ICN. Part V of the paper concerns the “effectiveness” of best practices. Burrichter and Zandt point out that no comprehensive survey has yet been undertaken to assess the effectiveness of internal best practices for merger control procedures. As an initial framing step in that direction they endorse, as a standard of success, a basic concept of “balance” between public and private interests. They then give reasons why internal best practices may be to liable fall short of the correct balance, particularly because in their view such documents seldom provide adequate certainty and predictability. This provides a background for the authors to discuss DG Competition’s Stakeholder Survey of 2010 and the praise and critiques it elicited.149 In Part VI the authors collect their conclusions. Their findings are generally positive as far as external best practices are concerned. However, they point out that multilaterally prepared and propagated best practices (basically, by the ICN) present benefits that can’t be matched by bilateral documents (as the latter are too limited in scope and potentially out of step with global standards) or, still less, private initiatives (lacking in authority). As for internal best practices, the current state of play is unsatisfactory for the reasons noted above. In order to arouse a stronger commitment on the part of competition agencies worldwide to assemble 148 A 2002 document, revised in 2007, provided guidance concerning the JFTC’s prior consultation procedure but this procedure was formally laid to rest in 2011. Some elements of the 2011 guidelines provide information relating to “process” issues including notably section VI, which discusses the JFTC’s approach to remedies. 149 See Eurobarometer results (18 October 2010), with various compiled reports as well as an aggregate report at http://ec.europa.eu/competition/publications/reports/surveys_en.html.

lviii  Merger Control in European and Global Perspective and publish more useful guidance on procedural matters, the authors recommend that the ICN should actively encourage the development and adoption of internal best practices and should advocate a standard minimum set of subjects to be addressed (including pre-notification contacts, scope of requests for information, conduct of meetings, access to file and confidentiality). According to the authors, such best practices should also have a harder edge, in the sense that commitment softeners such as “this guidance is in no way binding on the Authority” should be avoided as far as possible for the sake of legal certainty. Only the most sanguine agencies are likely to fully embrace this advice. Götz Drauz kindly contributed the twenty-fourth and last paper to be reviewed in this introductory chapter; the title, “Promoting Best Practices within and among Competition Authorities and with Business”, reflects the paper’s scope and structure. As an example of the “within” part of the paper (internal best practices, to borrow the term used by Burrichter and Zandt), Drauz discusses the European Commission’s merger-related best practice documents as a leading example.150 In particular, he describes the Commission’s Best Practices for merger control proceedings of January 2004151 and, briefly, the 2008 Remedies Notice (and the related Standard Models on divestitures and trustee mandates).152 With regard to the former, Drauz finds that the Commission has been a leader in experimenting with open and flexible dialogue with merging parties, and that such pragmatism helps to expedite the resolution of complex cases. However, he laments a certain creeping inefficiency in EU merger reviews, attributable, he says, to case teams whose lack of experience leads them to make expansive requests for data of slight probative value. Other critical remarks are provided with regard to the Commission’s then-ongoing consultation on best practice in antitrust proceedings (investigations, submission of economic evidence and the Hearing Officer). Drauz turns next to best practices and cooperation “among” competition agencies (external best practices). These include multilateral cooperation (facilitated by the work of the ICN’s Merger Working Group and of the OECD) and bilateral cooperation. With regard to the ICN and the OECD, Drauz reviews the various initiatives of these organizations and underlines their value but at the same time observes, with reference to the ICN in particular, a substantial gap between its Recommended Practices and their actual reception and effective implementation by its members. This point corroborates the findings in the papers of Boyce and of Burrichter/Zandt, and serves as a reminder that, while the “take-up” of the Recommended Practices has in many ways been a success story over the last decade, global merger control still faces growing pains and is destined for some time to lumber forward at variable speeds. Turning to bilateral cooperation, the friendly 150 For comments on the Commission’s best practice documents, see also the oral remarks of Mario Siragusa, pp. 333-336. With regard to their “bindingness” vis-à-vis the Commission (and in particular the relevance of the principle of good administration set forth in Article 41 of the EU Charter of Fundamental Rights), see the remarks of Judge Forwood, pp. 337-338. 151 See http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf. 152 Commission Notice on remedies acceptable under the Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, 2008 OJ C267/1.

Introduction 

lix  

rivalry between the European Commission and the US Agencies provides an irresistible example. In this context, Drauz reviews the essential points of the 1991 bilateral cooperation agreement and then reminisces about the flare-up over GE/ Honeywell.153 The important point here is that high-profile disagreement need not poison working relations but should rather be seen in the wider context of generally well-functioning cooperation between the Commission and the Agencies, and that the media-stoking fallout from GE/Honeywell has given way to more intensive and generally positive transatlantic cooperation despite what appear to be tremendous institutional challenges.154 Given the strategic importance and leadership role of the Commission and the Agencies, success in overcoming cross-jurisdictional conflict may also be essential in the sense that breakdowns between these authorities would hardly augur well for global cooperation more generally, and could conceivably cause collateral damage for future merging parties. Following a short discussion of how competition authorities can involve businesses and their advisors in the process of developing the content of best practices and thereby enhance their quality and usefulness, Drauz comes back to the point that “there is still a long way to go before true convergence is reached in the way the world’s competition authorities conduct their review of mergers”.155 Notable differences remain in virtually all dimensions: substance, procedure, institutional design and interstitial norms of practice. He concludes with a rather grim assessment of the global state of affairs, emphasizing the high costs of today’s maze of merger control regimes; but he sees the continued advancement of best practices in their quality and concrete impact as a crucial means of managing diversity and complexity.

153 For further discussion of the transatlantic dimension of this case, see inter alia the sources cited in this volume at p. 201. 154 On the surprising success, all things considered, of the European Commission’s cooperative efforts internationally in the field of merger control, see Deisenhofer, “International Cooperation in Merger Cases – An EU Practitioner’s Perspective”, this volume, pp. 227 et seq. 155 Drauz, “Promoting Best Practices”, this volume, p. 388.

Introduction and Panel I Merger Enforcement Across Jurisdictions: Substantive Issues (Market Definition, Entry Barriers/Potential Competition, Unilateral/ Coordinated Effects, Innovation, Efficiencies)

1 INTRODUCTION and PANEL DISCUSSION

Chair: John Fingleton

Speakers: Simon Bishop Cristina Caffarra Kirsten Edwards Amelia Fletcher Barry Hawk

Scott Hemphill Johannes Lübking Alison Oldale Howard Shelanski Irwin Stelzer



2  Merger Control in European and Global Perspective

Introduction John Fingleton: Good morning, and welcome to Florence. We are discussing merger control this year, and there is probably no other area in competition law where we’ve seen more convergence than in merger control over the last several years. So some of our discussion will likely be about fine tuning, and some of the discussion will be technical in nature. We won’t be talking about the initial questions such as, why do we need merger control. But those questions never go away, and we see political intervention in the merger control process in some instances, including the Potash case in Canada.1 So as we go through some of the narrower, more technical issues we will also want to bear in mind that there are broader issues as well. We may well hear speakers talking about false positives and false negatives in merger control. Some may argue that in high tech markets, false negatives are not costly, whereas what we see when we look back at auditing markets, where mergers were allowed about ten years ago,2 false negatives are seriously costly to the economy in some markets. The balance between false positives and false negatives, and the costs of each, seems to be a theme running through several of the papers. The recent revisions to the merger guidelines in the US and in the UK will be a strong theme this morning. There is quite a lot of discussion in the papers about the evidence that is used in merger cases. As Chairman of the ICN, I worry about whether some competition agencies will be able to match the evidentiary requirements that the US and the UK are setting, and we need to think about how to make some of these requirements administrable. So there are some convergence issues there as well. Irwin Stelzer: If I can steal someone else’s line, the views I express might not even be my own. I’d like to make several points in relation to merger control. First, everybody is unhappy with market definition. Some economists have done some deep thinking about this, but in the United States, we have the Court. And as Barry Hawk points out in his paper,3 so long as the statute refers to a line of commerce, there will be some struggling with market definition. I know a lot of people want it to go away, and I am among them, at least partly so. But the fact is we’re going to have to grapple with the intersection of law and economics. The great 1 BHP Billiton/Potash Corporation of Saskatchewan. On November 3, 2010, Canada’s Minister of Industry, acting under the Investment Canada Act, announced that the proposed (US $ 40 billion) acquisition by BHP Billiton of Potash was unlikely to be of ‘net benefit to Canada’. This effectively halted the transaction. For more details on the Investment Canada Act, see, eg, Anthony Baldanza, Huy Do and Douglas New, ‘Canada: Foreign Investments’, in Global Competition Review, The AsiaPacific Antitrust Review 2012, Section 3, Country Chapter 5. 2 For details on how the ‘Big Eight’ auditors active in the 1980s became the ‘Big Four’, see Niels Philipsen, Background Note, in OECD, Competition and Regulation in Auditing and Related Professions, DAF/COMP(2009)19, http://www.oecd.org/dataoecd/55/46/44284186.pdf, at pp 17–25. 3 Hawk, ‘A Tale of Two Cities: Washington and Brussels Face the Courts’, this volume, pp 53 et seq.

Introduction and Panel 1 

3  

legal scholar Kingman Brewster once said that if we economists would concede that law is a social science, he would concede that economics is a profession. (laughter) The second point is that I’ve been trying to figure out where the next set of merger problems is going to come from, and if you look at corporate balance sheets you see about 50 billion dollars in excess cash that Apple is sitting on, 22 billion in excess cash that Google is sitting on, and 8 billion in excess cash for News Corporation. And if you consider the antipathy our corporate boards have to returning money to shareholders, you’ve got to assume that this is where the merger action is going to be. And DG Comp is also expecting mergers to be happening in these sorts of industries, as I understand it. The problem that I have with the alphabet soup described by Amelia Fletcher is that I’m not sure it’s going to be very useful for appraising mergers in those industries. The kind of mergers we’re going to be confronting are not horizontal mergers. They’re going to be acquisitions by firms with very substantial market positions of firms in cognate industries. The sole purpose of these transactions is to transfer to the target firm the benefits of being connected to a much larger firm with a much larger market position in its own field. And I came here to be instructed by all of you about how I should be guided through that kind of situation. Now my suggestion, and I’ll conclude with this, comes from watching my client acquire MySpace, which was a dominant firm by the usual measures, only to then be obliterated by Facebook. In these sorts of markets we’re not talking about contestable markets in the way that Will Baumol first developed it.4 We’re talking about the ‘destructive’ part of creative destruction.5 Obliteration would be a better word. And we should look, for instance, at what Android is doing in the smartphone market, going from 3% of the market to 30% of the market. So whenever the subject of market power comes up we should really be asking how durable that power is. How do we factor that in to merger control analysis? We need to know how to do that, because that’s where the next wave will come from. We’ll see banking consolidation, of course. In the current atmosphere banks will be an enormous problem. But we need to be thinking about the kinds of markets I have referred to. In the US, corporations are holding two trillion dollars in excess cash, and a lot of it is in these industries. Do we need to shape merger control in order to account for this? Or are the tools we have adequate, and am I completely wrong to worry? 4 See William Baumol, John Panzar and Robert Willig, Contestable Markets and the Theory of Industry Structure, Harcourt Brace Jovanovich, 1982; William Baumol, ‘Contestable Markets: An Uprising in the Theory of Industry Structure’, 72 American Economic Review 1 (1982). 5 See Joseph Schumpeter, Capitalism, Socialism and Democracy (1st edn 1942), Routledge, 1994, pp 81–85. Schumpeter’s vision of creative destruction is adapted from the rather different, but not too different idea advanced by Karl Marx to the effect that capitalism continually destroys existing wealth while at the same time creating new wealth. One of the interesting aspects of Schumpeter’s arguments, of course, is that although creative destruction is an important element of capitalism that brings significant long-term economic gains and enhanced living standards, capitalism itself – the best form of economic organization – is predicted to decline, as an intellectual class turns its back on it, and succumb to a form of socialism (famously: ‘Can capitalism survive? No, I do not think it can.’ – ibid. p 61).

4  Merger Control in European and Global Perspective Fingleton: Thank you, Irwin, that’s a nice, provocative way to start the morning. We’ll move on to the other speakers now, and Irwin has already set the standard with his concise set of points to get us going. Irwin’s presentation functioned as an introduction for us, and now we continue with a session devoted to ‘merger enforcement across jurisdictions’, and in particular to substantive issues. Our first speaker is Johannes Lübking of the European Commission.

PANEL 1: Merger enforcement across jurisdictions: substantive issues (market definition, entry barriers/potential competition, unilateral/coordinated effects, innovation, efficiencies Johannes Lübking: Thank you. Well, this year we have the 20th anniversary of the adoption of the EC Merger Regulation, and of course one of the major events since that original Regulation was adopted was its revision in 2004. I’d like to look at what has happened in EU merger control since 2004. Two related developments I should mention were the Horizontal Merger Guidelines of 2004, and the Non-horizontal Merger Guidelines of 2007. Also relevant is the Remedies Notice of 2008. Now, in 2004 when we adopted the new substantive test, the focus was on horizontal cases, typically in oligopolistic markets, and there was a lot of discussion about ‘gap cases’, that is to say, non-coordinated effects cases which did not result in post-merger single dominance but which had significant competitive effects. It’s worth noting that most of our cases over the last six years would somehow fit into the ‘dominance’ category, so they were not gap cases. But with our increasing emphasis on effects what we saw was that there were a number of mergers involving firms with significant market shares which were nevertheless cleared. For example, we had the Rotogravure printing joint venture involving Bertelsmann, and we cleared that despite high market shares due to a lack of competitive effects. And as you know we’ve been looking closely at the role of differentiated products, and there we had the chocolate case, the Kraft/ Cadbury case.6 We cleared that unconditionally for France and for the UK, again despite high market shares. In the UK we found that consumers preferred British chocolate, and didn’t like Continental chocolate very much. On the other hand, we looked at Romania and Poland, and there we saw close competition between the merging parties and so we asked the parties kindly to divest one of their traditional brands. But we have seen gap cases, the first of these was T-Mobile/Tele-ring,7 in the Austrian mobile telecoms sector. That was a merger between the number 2 6 7

M.5644, Kraft Foods/Cadbury (6 January 2010). M.3916, T-mobile Austria/Tele.ring (26 April 2006),

Introduction and Panel 1 

5  

and number 4 firms in that sector, but it did not create a market leader, instead the result was a strengthened number 2. Since Tele-ring had been a very active competitor (a maverick, if you like), we found that the transaction was likely to change incentives, and we cleared the merger but T-mobile divested, in essence, most of Tele-ring’s frequencies and network. We’ve had other cases as well, such as Fortis/ABN Amro.8 It took us some time to implement the remedies in that case, but we did so at the beginning of 2010. More recently, we had EDF/ Segebel in the Belgian wholesale electricity market.9 There we had a large margin between Electrabel, the market leader, and the merged entity but again we found competitive effects due to changed incentives. EDF had been planning to install new capacity in Belgium, and we found that the merger would have removed the incentive to do that. Then there were coordinated effects cases. There we had guidance from the Court of First Instance in the Airtours judgment,10 and then we had refinements to the relevant criteria in the Impala saga.11 In 2008 we had the ABF/GBI case in the compressed yeast sector.12 This was a textbook coordinated effects case, the final customers were bakeries in Spain and Portugal, with local distributors, and the market was very transparent in terms of prices, volumes and capacities. We found that the distributors were visiting the customers several times a week, with information being channelled to suppliers, and we found that retaliation in case of aggressive pricing would also be possible. Turning to the non-horizontal merger cases following the adoption of the Nonhorizontal Merger Guidelines, we had, first of all, Google/DoubleClick,13 and then we had the two maps cases, TomTom/Tele Atlas14 and Nokia/Navteq.15 The Guidelines were important in all these cases. They set out a three-step analytical approach, which focuses on the ability to foreclose, the incentive to foreclose, and the impact on the downstream market. This was a learning experience for the Commission, and those criteria were applied quite strictly. The case concerned the integration of digital maps and portable navigation devices, and we found that the merged entity would have no significant incentive to restrict the supply of digital map databases to its competitors because a loss in sales of maps would not be compensated by increased sales of mobile telephones. And third, there was no impact on the downstream market. The Commission took account of efficiencies, and specifically the elimination of double marginalization. But it was difficult to establish the margins and the profits in this case. We put our Chief Economist’s team to work there, but it was not easy. M.4844, Fortis/ABN Amro Assets (3 October 2007). M.5549, EDF/Segebel (12 November 2009). 10 C-342/99, Airtours plc v. Commission [2002] ECR II-2585. 11 Case C-413/06 P, Bertelsmann AG and Sony Corporation of America v. Independent Music Publishers and Labels Association (Impala) [2008] ECR I-4951. 12 M.4980, ABF/GBI Business (23 September 2008). 13 M.4731, Google/Double Click (11 March 2008). 14 M.4854, TomTom/Tele Atlas (14 May 2008). 15 M.4942, Nokia/Navteq (2 July 2008). 8 9

6  Merger Control in European and Global Perspective We also had a Phase I vertical case in 2008, which was IPIC/MAN Ferrostaal.16 IPIC had a melamine producer while MAN Ferrostaal had a minority shareholding in a supplier of high-quality melamine. Again we ran through the three main criteria in the Guidelines, and we found they would all be satisfied. So we intervened, and divestments were made as a condition for clearance. I would also like to mention two cases that reached the EU Courts, first, Sun Chemical v. Commission,17 and second, the Ryanair judgment.18 In these cases the General Court looked very closely at the Guidelines. In the Sun Chemical case, the court said, first, Commission, you are legally bound to respect those guidelines, and second, we are going to go through and apply the criteria in those Guidelines ourselves. And in Ryanair the court took the Guidelines very seriously, in relation to both entry and efficiencies. Now, I see that my time is up, so I will stop there. Thank you very much. Amelia Fletcher: I’m going to address what some have referred to as ‘alphabet soup’, which is a reference to expressions such as UPP, that is, upward pricing pressure. Now this approach is not for the kinds of mergers that Irwin Stelzer was describing. This methodology is for more standard horizontal mergers in differentiated goods markets. The UPP concept has been incorporated very firmly in the US Merger Guidelines, and we have put it, somewhat less firmly, in the UK Guidelines as well. And in this presentation I want to discuss a ‘sister’ concept, which is the illustrative price rise, or IPR. But I want to begin with a short and stylized history lesson… Once upon a time, assessing horizontal mergers was fairly straightforward. It was essentially a counting game. The analysis would start with market definition, in some cases more rigorously with the use of critical loss analysis, and very often less rigorously, particularly at Phase I. Then we looked at concentration. We counted fascia, we counted market shares, maybe we did Herfindahls. And essentially, if the post-merger concentration was too high, and if the increment was too high, then we blocked or we remedied the merger. We did look at barriers to entry, we did look at durability, and we looked at buyer power, but that was the basic approach. The problem with that approach is that it didn’t really allow for an assessment of differences in the closeness of competition between market players. And that factor is crucial for mergers because a merger between close competitors will tend to have a greater impact on prices than a merger between more ‘distant’ competitors. Now in order to take this factor into account, Farrell and Shapiro have given us a giant leap forward. It might not be like putting a man on the moon, but it’s still pretty impressive. And on the basis of this literature we have UPP, we have IPR, and we have GUPPI, which means gross upwards pricing pressure index. All of these have pros and cons. The GUPPI is the most general and simple, 16 17 18

M.5406, IPIC/MAN Ferrostaal AG (13 March 2009). Case T-282/06, Sun Chemical Group v. Commission [2007] ECR II-2149. Case T-342/07, Ryanair v. Commission [2010] ECR II-3457.

Introduction and Panel 1 

7  

and it’s very attractive for that. It’s simply the margin multiplied by the diversion ratio. Margins and diversion ratios are the two key elements of all of these measures. Margin times diversion ratio means the value of diverted sales that is internalized as a result of the merger, and this is what changes pricing incentives. What it doesn’t tell us anything about is what the likely impact on prices is. For that we can look at the illustrative price rise, or IPR, and this is its great beauty. The trouble is, just like Dorian Gray, that great beauty comes at a cost. And the mirror in the attic in this case is that we have to make quite strong assumptions about the nature of competition and the shape of demand. Now the UPP concept is more general than the IPR, but it is better than the GUPPI because it allows us to look at efficiencies, and in fact it generates a figure for the efficiencies that would be required in order to offset a merger’s upwards pricing pressure. The key point is that these are all very closely related, and they all provide a better proxy of merger effects than before. They are also much easier than merger simulation, and can be done at Phase I, whereas merger simulation is essentially impossible at Phase I without substantial pre-notification work. Let me tell you about the UK’s experience with the IPR methodology. This was first used by the UK Competition Commission in 2005 in its second-phase inquiry in the Somerfield/Morrison merger, involving supermarket chains. This involved a two-stage analysis. First, you define local catchment areas, on a fairly cautious basis, and you count fascias to identify potential problem areas. And then within those you calculate illustrative price rises. Since 2005 the OFT has refined the IPR in various mergers, at the first phase, and recently we’ve looked at asymmetric illustrative price rises, where you have asymmetric diversion ratios. There are four key elements to the IPR methodology, I’ll just go through them quickly, and there would be a fifth element in a case involving asymmetric illustrative price rises but I’ll leave that to one side. So the main elements are diversion ratios, gross profit margins, assumed demand function and the intervention thresholds. Starting with diversion ratios, it isn’t completely simple to derive these. There’s a debate to be had about how to get them and what are the right questions to ask, and Kirsten Edwards is going to talk about that in her presentation. But they can be derived in various ways: surveys, econometric analysis of demand, win/loss data and event studies. The OFT has most commonly used surveys, and because we don’t think that most consumers are good with percentages, we tend to ask what the consumer would do if a given product were unavailable. Then we need gross profit margins, again, this isn’t easy. We generally use variable costs as a proxy for marginal costs, and there we need to be careful with regard to the time periods and the output increments over which we consider the variability of costs. The principle is that we should be using the time periods and output increments that reflect the relevant process of competition. There is a tricky issue in the case of a multi-product firm where the merger affects only one of those products, but that’s not insurmountable. In grocery mergers, for example, we look at the grocery margins rather than the non-grocery margins.

8  Merger Control in European and Global Perspective If you have your margin and your diversion ratio you can work out most of the UPP measures, but in order to measure IPR you also need to assume a demand function. In practice we don’t know a lot about the shape of demand, at least at the first phase. We tend to say the iso-elastic demand is an upper bound on our IPR estimate and the linear demand is a lower bound. At the OFT we’re typically cautious, and we use the iso-elastic demand. Finally, once we have our margins and our diversion ratios and we’ve assumed a demand function, we can easily generate our illustrative price rise. But then there’s a question: what kind of IPR do we worry about? So we need an intervention threshold. The OFT has tended to apply a 5% threshold for the IPR, although we do some sensitivity analysis around that. Now that doesn’t mean that the OFT has any tolerance for price rises. But arguably, the 5% figure allows some credit for unmeasured efficiencies, and for possible measurement errors, and also for the illustrative nature of the model that is used. These models are not perfect reflections of reality, they are more like metaphors. The question of where to set the intervention threshold could be a useful line of research, bearing in mind that the intervention threshold may vary depending on which indicator is used. Just to conclude, I’m very attracted to this approach, mainly because it’s so much better than the simple counting approach. But not everyone likes it. Some claim that there’s a lot of work and costs involved, lots of surveys of local areas, lots of implicit assumptions being made about the nature of competition, it’s simplistic, there’s no allowance for competitor reaction, product re-positioning, buyer power, durability, and so on. I do think these concerns carry some weight, and caveats are needed. But I’d rather that we refine the methodology where we can, rather than junking it too soon. Fingleton: Thank you, Amelia. Now we have not one but two Chief Economists from the UK with us, and next we’ll hear from Alison Oldale of the Competition Commission. As we worked on the UK Guidelines, we certainly worked very closely with her and with our American colleagues as well. And our merger work at the OFT is very much a joint product with the CC. So here now is Alison. Alison Oldale: Thank you very much, John. In light of Irwin’s remarks I’ll begin, like Amelia, by stressing the limited scope of this presentation. I’m going to be talking about the immediate effects, from the loss of horizontal competition, in differentiated product markets. Cleary, that does not extend to all issues arising in all kinds of mergers. Now in the revised Horizontal Merger Guidelines in the US and in the UK, there is less emphasis on the SSNIP test for the early identification of markets and more emphasis on the direct assessment of unilateral effects. At least in the UK, the problem that we had was not so much with markets as with the SSNIP test, and not so much with market structure as useful information, but rather as attempts to use market structure all on its own. What I’d like to do is to put the recent developments in historical context, as another step in the process of accepting the collapse of the structure-conduct-performance paradigm.

Introduction and Panel 1 

9  

To do that we can go back to 1968, when the Department of Justice adopted the original Merger Guidelines. Those guidelines were based squarely on structure, conduct and performance. Under that approach, if you knew the structure of the market, that told you how firms were going to conduct themselves, and that was all you really had to know in order to understand whether post-merger prices would be high. At the time it was believed that there was firm empirical support for this paradigm, but it was already under attack. People were pointing to scale efficiencies as alternative reasons for why prices might be high when concentration was high, and people were also looking more closely at the data that supposedly supported the prevailing structural approach. More recently, John Sutton suggested that causation was completely the other way round.19 If you have strong competition, it means that firms have a harder time staying in the market, and you wind up with higher concentration. So observing concentration can indicate strong competition. As time went on, and as the US adopted successive versions of its guidelines in 1982 and then 1992, empirical support for the structure-conduct-performance paradigm had evaporated. By this time, it was evident in the guidelines that competitive analysis was framed in terms of market power, and not just in terms of market structure. The 1982 guidelines introduced the SSNIP test as a means to introduce this shift from structure to market power. The SSNIP test, as you all know, is used to identify the narrowest set of products over which firms can exercise market power. It’s a means of calibrating market shares, if you want to use market shares as a stand-alone indicator. If you know what the effect of a merger to monopoly is, then you stand a chance of knowing what the effects of a merger to X% market share are, if X is the only bit of information that you’re going to use. But the price of this calibration is very heavy. Firstly, the markets defined in this way are not very obvious, and the newly revised US Guidelines discuss some cases illustrating how the use of the SSNIP test to define markets might result in markets that are rather odd. Now, this isn’t a problem if the only point of defining markets is to calculate market shares. But it’s not – we use markets for other things, and I might get to that later. The other problem with the SSNIP test is that it mixes up the SSNIP test with the competitive assessment. Applying the SSNIP test involves estimating the competitive effects of a merger to monopoly. This second point ought to have led to a decisive break with the idea that you can first define markets and then do the competitive assessment. It was finally this year that the change was made explicit in the new US Merger Guidelines. But in many places, such as Europe, the US and the UK, the practice had already evolved and it is only now that the guidelines are catching up. In the UK Guidelines we’ve kept markets, but we’ve explained that we don’t use the SSNIP test as the sole basis for defining them. There will be more focus on the 19 John Sutton, Sunk Costs and Market Structure: Price Competition, Advertising and the Evolution of Concentration, MIT Press, 1991. See also John Sutton, Technology and Market Structure – Theory and History, MIT Press, 1998.

10  Merger Control in European and Global Perspective natural boundaries between products and between segments. We’ll use the SSNIP test only if we’re going to use market shares as a stand-alone indicator of concerns. And if we do that it will be a separate exercise using a market defined especially for that purpose. We still need markets, they are part of the organizing framework for assessing a merger. You need some idea of what the markets are in order to think about what the theories of harm are, and to identify overlaps. And you need markets in order to think about things that may require a second assessment. But markets can also be helpful as analytical tools, and I’ll come back now to market shares and market structure. These elements can in fact be informative, it’s just that you need other bits of information as well in order to interpret them. One loose indication of a price effect from a merger can be derived by multiplying margins times some measure of market structure times the recapture rates (that’s the amount of diversion that is retained within the market when the price of one of the products goes up) times the pass-through rate. So market structure is in there, and it’s telling us something useful, it’s just that it doesn’t tell us anything useful on its own, you need to put it together with other information, including the competitiveness of the market, which affects margins and pass-through rates, and how tightly the boundaries of the market are drawn, which affects the recapture rate. Markets are useful for thinking about the basis of new entry. You need some idea about what sorts of products would count, if you saw entry of a new one, in order to counteract the effects of a merger. Is the set of products that forms that basis the same set of products that you would get from the SSNIP test? I’m not sure. My bet, not having thought about it hard enough, is that the markets would be a bit broader, actually, than the markets you would get from the SSNIP test. And markets are very useful for thinking about competitive dynamics. It is quite useful to know the set of products you’re interested in if you’re trying to track over time how shares have changed, and how the patterns of entry and exit may have evolved, and just generally to understand how stable market positions might be, and what sorts of firms and products switch market share over time. In conclusion, the aim of the SSNIP test was very noble. It was to recognize, as a failure, the structure-conduct-performance paradigm, and to shift the focus of merger control to assessing market power. But the SSNIP test mixed up market definition and competitive assessment. The alternative is to drop the SSNIP test, define markets in a more natural and obvious way, but to make sure that when you do that you’re combining market share information with other evidence in order to draw conclusions about competitive effects. Kirsten Edwards: I’m going to talk about the practicalities of estimating diversion ratios for use in these models. And these models are particularly appealing because they are simple to implement once you have the data. But trying to get diversion ratios in the first place can be quite tricky, especially if you don’t have price or quantity data that enable you to estimate them on the basis of actual consumer behaviour. In many cases we have to use surveys to get an understanding of what the diversion ratios are.

Introduction and Panel 1  11   I’ve been reviewing cases, and the sample is biased because most of the time it has been cases that I’ve worked on, or cases that I’m familiar with. But I’ve been looking at the kinds of questions that are being asked to elicit this information about consumer switching behaviour. The questions are normally asked in three formats. The first kind of question is: would you switch to another product if faced with an X% price increase? For those respondents that say yes, there’s usually a follow-on question, which is: what product would you switch to? This kind of question has been asked more in Europe than in the UK, actually, and an example would be recent airline mergers. A second type of question is: what would you do if this product was unavailable? This was the kind of question that was endorsed by the OFT and the Competition Commission in their consultation on the Merger Guidelines, and it has been used in a number of cases in the UK. And the third sort of question is a stated choice question. This is more of an experiment than a question. The question posed is: out of this selection of products, which would you choose? Some simplified product choices are then offered to the consumer. So there is product A, which is big and green and it costs five euros. Product B is small and blue and costs 6 euros, and the consumer is asked to choose between those products. And if you give the respondents a range of these questions, switching patterns will emerge depending on changes in price or product characteristics. And this approach does not come up very often, although I did see that the Competition Commission used this kind of question in National Express/Greater Anglia, which was back in 2004.20 What interested me was whether the type of question that is used really matters. I think the answer is yes, it does matter. I’ve been looking at the literature on surveys, and there is a vast body of work on the way surveys should be designed. There’s not so much relating specifically to competition policy, but there’s a huge amount in fields such as health and political polling and pensions. In these kinds of areas you find massive, nationwide surveys. In this literature on survey design, the emphasis is on avoiding four types of error, namely: sample error, coverage error, non-response error and measurement error. Now how do these types of error affect diversion questions? Well, sample error poses a particular problem in relation to the first type of question, that is, would you switch if there were a price increase, with the follow-on question of where would you switch? There you’re only asking the diversion question of a small sample of people. Sample error, by the way, is to some degree inherent in surveys because you’re only addressing a subset of the population. The larger the sample of respondents, the less risk of error, and conversely the smaller the sample the greater the risk. The problem with sample bias was in fact noted by the Commission in KLM/Martin Air.21 They asked this type of question, and only 14% of the people they asked said that they’d switch. So on one of the routes, where they calculated the diversion ratio as 20%, they noted that the 95% confidence interval around this diversion ratio was 8% to 32%, which was very wide, and as a consequence the interpretation of the result was limited. 20 21

See CC Press Releases of 50–04 of 17 September 2004 and 64–04 of 4 November 2004. M.5141, KLM/MartinAir (17 December 2008).

12  Merger Control in European and Global Perspective The second type of error which needs to be minimized is coverage error. Coverage is where either a portion of the population who should be in the sample are excluded, or where there are people included in the sample that are not of interest. This is a problem that comes up in relation to the way the OFT and the Competition Commission formulate their question, because they ask everyone, what would you do if this product were unavailable. So everyone is ‘forced’ to answer this question, but what we’re really interested in is whether marginal consumers are going to switch. Does it matter? Well, it might matter if the preferences of marginal consumers and inframarginal consumers are different, and that’s a plausible scenario. Marginal consumers are more price-sensitive, and so they may be more likely to switch to an alternative that is cheaper as compared with the alternatives that would be preferred by inframarginal consumers. The Competition Commission had the NEG/Thameslink case, a case in the train sector.22 In order to identify diversion ratios, two sets of questions were asked. They were looking at substitution between the Gatwick Express and the Thameslink train, both of these run from central London to Gatwick Airport. In the survey the CC first asked: what would you do if this service were unavailable. And then it asked: what would you do if there were a 5–10% price increase? And then they compared the diversion that could be estimated from the two answers. When they asked the marginal consumers, marginal consumers were, unsurprisingly, more likely to switch to coach, which was cheaper, and less likely to switch to the more expensive Gatwick Express. Two other types of error. One is non-response error, I won’t say much about that. This error comes up when people don’t respond to a question and we cut them out of the data. Potentially that could be a problem if those not responding have a particular preference that we’re missing. The final type is measurement error, which is where responses to the question are inaccurate. People might misunderstand the question, or the question might allow for ambiguity and might therefore be misinterpreted. There can be framing biases too, particularly if response prompts are given, and at least in the Ryanair/ Aer Lingus survey, customers were asked whether they would consider flying with Ryanair, with Aer Lingus, or with ‘other’.23 Questions like that might have a framing bias problem because customers might just accept the response that they see. Now all of this may sound like doom and gloom, since I’ve been talking about all the errors that may blemish a survey, but I do think surveys help. It is like IPR, or UPP, surveys take us a step beyond where we once were in merger analysis. But when we use surveys we need to approach them more critically, and more research in this area would be helpful as well. One thing we can do is to look at stated choice-type surveys because those are thought to create fewer bias problems. The cost there is that they’re more complex and time-consuming to administer. But if we start doing it we’re only going to get better at it. Finally, we also need to look at other evidence, of course. You may get outlying diversion ratios, and there may 22 23

CC Press Release 84/05 of 22 December 2005. M.4439, Ryanair/Aer Lingus (27 June 2007).

Introduction and Panel 1 

13  

be ways to check that, and in general these data need to be considered in their proper context. Fingleton: Thank you, Kirsten. When I used to hear economics being described as the ‘dismal science’, I always used to think it was macroeconomics they were referring to, but now it seems that it’s not macroeconomics but survey economics. (laughter) But thank you for that, and there’s another question there as well, which is whether competition authorities can rely on the merger parties to do the survey work. If we are concerned when the authorities do it, we might want to be doubly cautious when the parties are doing it. Let me turn to our next speaker now, Simon Bishop. Simon Bishop: Thank you. I’d like to begin with the basics. All horizontal mergers eliminate a direct competitive constraint, by definition. But it’s also clear that not all horizontal mergers are anticompetitive. This raises a big issue of how we can distinguish anticompetitive mergers from pro-competitive mergers. There are numerous questions we need to ask. What is the intensity of the competitive constraint between the merging parties? What is the intensity of the competitive constraints provided by other competitors? How easy is it for firms to expand their output? How easy is it for firms to adjust their product offers? How easy is it for firms to enter the market? How easy is it for buyers to underwrite new entry? What impact do any merger-specific efficiencies have? It’s a long list, it’s a complicated assessment, and in some cases a very complicated assessment. And not long ago, some people were saying we can get to the right answer very easily, we’ll do a simulation, and out pops the answer. We can take all of these factors into account in one go. There are a number of issues with merger simulations. I won’t rehearse them here, but one of them is that it takes a lot of time. So as a practical matter there has to be a filter. Which mergers can we waive through directly, and which require more attention? So whether you’re in the traditional camp, or whether we’re you’re in the UPP camp, in both cases we’re talking about filters. Under the traditional approach, you define the relevant market and you count shares. And if there is high concentration, we should conduct a more detailed investigation. What’s the problem with that? One can argue that it all depends on whether the market is properly defined. Well, yes, but that’s an operational point. Accuracy always has to be a top priority. Putting that aside, the most common criticism of the traditional approach is: how are we to interpret market shares? The proponents of UPP point that out that market shares can either understate or overstate competitive significance, because they don’t reflect enough information about competitive relationships in differentiated markets. Market shares implicitly assume that all products within the market are equally substitutable. Well, true, but that doesn’t mean we should stop using these tools. As long as we recognize the potential problem with market shares, we can take that into account. Moreover, there is increasing awareness now that high market shares by themselves don’t

14  Merger Control in European and Global Perspective mean that there will automatically be a prohibition or competition problems, instead they should function, and they do function, as a filter to show us when we should be looking at a variety of factors more closely. Now with the UPP, you don’t look at market shares, the idea is to look directly at the competitive constraints between the merging parties. There are two issues here. First, the results of the analysis that you do has to accord with the way the market really works. But the UPP is fairly agnostic about what is happening in the real world because you just calculate diversion ratios and gross margin, and that test doesn’t really care about anything else. Now is that consistent with pre-merger outcomes? Well, there the problem is: who decides? If two different analysts come up with two different diversion ratios, it’s not clear how to discriminate between those conflicting data. With regard to gross margins, you have to take into account the time over which marginal costs are assessed. That’s an extremely complicated issue. Now Carl Shapiro has said, well yes but we measure variable costs routinely in predatory pricing cases. That’s true, we do, but we have to devote a lot of time to doing that, and that might not be suitable as a Phase I filter in merger control. We need something quick and easy to do. Measuring gross margins is not easy to do. The second element is that we need to look closely at how competitors respond. What can we conclude once we know that two firms are close competitors. I would say, in and of itself, nothing. You need to take into account competitive constraints provided by other players. Even if there’s a diversion ratio of 40% between the merging parties, the post-merger firm can be subject to very effective pressure from remaining competitors. So that analysis is central, and the traditional approach does that, UPP doesn’t. So then one can say well alright, we’ll put that in at the second phase. Let’s consider that. How does UPP perform? Well, absent efficiencies, UPP will always tell us that a merger should be subject to more investigation. So to get around that, when we apply the test we assume mergerspecific marginal cost efficiencies of 10%. But even if we do that – and here I’ll refer to a paper by Joe Simons and Malcolm Coate24 – if the margins are over 50%, UPP will be positive. In the paper they try to translate the diversion ratios into a structural analysis by making some assumptions about symmetry. They’re saying that if you’ve got high margins, even in markets with ten firms a 10 to 9 merger would fail the test, and a 6 to five would fail the test. So the UPP test is actually suggesting a much greater area for intervention. Is there evidence supporting that change of policy? I haven’t seen it. But my criticism of the UPP test isn’t unqualified, because with more intervention and more second-phase cases, the likelihood of me being on the dole is reduced. (laughter) So is this implied policy shift warranted? That’s an empirical question. Are there are a lot of mergers out there which have been permitted when they should have been prohibited? Finally, I’m always surprised when arguments are made to downplay market definition in merger cases when in Article 102 cases it’s absolutely critical. Of 24 See Coate and Simons, ‘Continuity and Change in the 2010 Merger Guidelines’, CPI Antitrust Journal (October 2010/2). More recently, see Coate and Simons, ‘In Defense of Market Definition’ (forthcoming, Antitrust Bulletin, 2012), http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1967208.

Introduction and Panel 1 

15  

course, market definition is easier in merger cases, and more difficult in Article 102 cases, because of the Cellophane fallacy. So if we move away from market definition in merger control, what implications does that have for the dominance test in Article 102 cases? For that matter, what are we doing with all those market share thresholds in the various block exemptions based on Article 101(3)? Cristina Caffarra: This morning I want to look at standards of evidence used in connection with large, multi-jurisdictional mergers. What we see, for various reasons, are differences in the standard of evidence across a range of jurisdictions. Those differing standards can lead to different theories of harm in relation to the same facts, or different positions regarding the same issues. As a consequence, ironically, predicting outcomes is not becoming easier despite a great deal of convergence in many jurisdictions. The point I’d like to make is that the empirical evidence necessary to show what is likely or unlikely to occur varies quite significantly, and I’ll refer to some examples. Starting with unilateral effects, simulation of unilateral effects in difficult cases is rather standard now, at least in the UK and in the US, and certainly at the European Commission. When simulations are done there is a lot of room for debate and disagreement, but the approach is broadly settled. And simulations can be done in a wide range of industries, and in some cases you might see quite tricky questions come up, such as the rate of investment in a market. But there are many jurisdictions which do not have the necessary expertise or resources to evaluation these models, and this raises questions about how to handle mergers in these diverse jurisdictions. Apart from unilateral effects cases, though, we have to think about coordinated effects cases, and this is where variations in the standard of evidence across jurisdictions is create particularly challenging issues. Now in the EU, we had Airtours,25 and a lot of economists welcomed that judgment, not only because it provided clarity with regard to the relevant framework for coordinated effects, which is tacit collusion, but also because it was very clear that the standard of evidence in that particular case was not satisfied. So then we started to look at transparency, monitoring, punishment and entry, although these criteria were sometimes applied in a fairly mechanistic way. Now the more sophisticated agencies have taken a rather restrictive approach to this. But a rigid checklist approach still dominates in many jurisdictions. Let me give you an example of this relative divergence across agencies, I’m thinking of the BHP Billiton/Rio Tinto joint venture, of course this case was widely reported in the press.26 The joint venture was notified in the country of the producers, Australia, but it was also notified in countries where the major customers were located, and in Germany, the parties had limited sales there Cited above note 10. The joint venture, ultimately abandoned, ran into trouble with several merger authorities including, among others, the European Commission, the Japan Fair Trade Commission, the Korea Fair Trade Commission and the Australian Competition and Consumer Commission. See, eg, Commission Press Release IP/08/1798 of 26 November 2008. 25 26

16  Merger Control in European and Global Perspective but it was considered that the market was a world market and there was some concern that the price effects would travel. When you look at the theories of harm adopted by the different authorities in that case, you see that they were not exactly consistent and homogeneous. One issue was whether we should worry about unilateral effects or coordinated effects. The market was concentrated, and it was described, in a caricature, as a three-to-two merger. The rate of demand growth was very high, and in the short term there were capacity constraints. In those circumstances, the key variable is how much investment needs to be made to meet future demand, as opposed to short-term production decisions. And here, coordinated effects don’t really work, they just won’t be credible. But the point is that some cases that are more properly regarded as unilateral effects cases will sometimes be framed as coordinated effects cases. When that happens it implies a multiplication of efforts. A multiplication of models, a multiplication of work and time, and so on. Sometimes agencies will say that the parties have plenty of resources for these purposes, and much more than the agencies do themselves, but that is not really a complete answer. You do not have constant returns to scale having even larger teams of economists trying to answer multiple questions. So to summarize, we have a ‘jurisdictional thicket’ out there, and this creates a resources issue at least for merger parties. So we can applaud the considerable progress that has been made in developing criteria and harmonizing procedures, but we do need more discussion of what constitute good standards of assessment across jurisdiction, and Bill Kovacic might touch on this later as well. Fingleton: I think if Vince Cable were here he’d be happy because he has said that there should be some grit thrown in the merger system, and you’re suggesting that there might be a lot of it in certain types of mergers. But wearing my ICN hat again, we’re working there to develop capacity as well as standards. This is an important aspect of the ICN’s work that is not always seen. The objective is to make sure that agencies are better organized and better able to manage their work, and all agencies should be considering their effectiveness and efficiency in that regard. Now we’re going to hear from Howard Shelanski, who played a key role in the revision of the US Merger Guidelines, and he also had quite a hand in handling the international coordination aspects during that revision process. Howard, over to you. Howard Shelanski: Thanks, John. I’d like to discuss what we were doing in the US with the Merger Guidelines, and what the role of market definition is, and then I want to shift gears and talk about a non-price issue, in particular, innovation. There is an unfortunate myth that is developing about the revised Guidelines that we’ve issued in the United States. The myth is that we’ve dispensed with market definition. That’s absolutely wrong, market definition still plays an important role. If you gathered together the six drafters of the revised Guidelines, you might get six different opinions about how great a role market definition

Introduction and Panel 1 

17  

plays, but that doesn’t mean market definition has been tossed out the window. In fact, we now have a rather expanded discussion of market definition compared to the 1992 Guidelines. There will be many cases in which we really don’t have a better option than to proceed with a 1992-style assessment of competitive effects. So market definition will be used either as a screening device or to generate some presumptions about the effects of a merger. I happen to think that market definition is a tool that can work well in relation to many mergers, I agree with Simon Bishop about that. And in the revised section on market definition, following the revised discussion of HHI-thresholds, you’ll see a sentence that says these thresholds should not be interpreted as triggers for enforcement, but rather as indicators of when the Agencies will move to Phase II, or Stage II in our parlance, and issue a second request. Narrowing down a transaction to a limited set of plausible market boundaries thus has a very valuable function, both for businesses who want to anticipate whether there will be a second request, and for the internal work allocation of the Agencies. Upward Pricing Pressure, or UPP, is meant to be an additional instrument to add to the tool basket. Of course, we do give this additional tool very prominent treatment in the Guidelines. That is a reflection of the powerful intellectual influence that Carl Shapiro and Joe Farrell had on the revisions we made, and it also reflects the fact that other members of the drafting group including myself had already been substantially in agreement with Carl and Joe as to the usefulness of UPP analysis long before we launched the revision process. The important role of UPP, and of the related concepts that Amelia Fletcher talked about, is to restore competitive effects as the primary objective of merger analysis by the Agencies. When one thinks about the SSNIP test, that test should really be a benchmark that triggers concern about a merger’s competitive effects. The ability to adopt a profitable, non-transient, small but significant price increase is an indication of market power that concerns us. But instead of being a benchmark to be proved through a variety of forms of evidence, the SSNIP became – as Alison Oldale laid out very well – a mechanism used to apply the hypothetical monopolist test in order to define the relevant market. So we go through this rather odd sequence of using the SSNIP test, which is more naturally about competitive effects, to reach conclusions about a hypothetical test, only to go back and enter into an assessment of a quite non-hypothetical merger. If we define the market in that way, we derive market shares and that takes us back to a case based on market structure, which may not necessarily be a good basis for reaching conclusions about competitive effects. So in revising the Guidelines we were influenced by the notion that this traditional approach based on market definition and market shares may not be the best way to identify a merger’s likely market effects. Why should we fight about the boundaries of a market if we have more direct evidence of effects? Simon pointed out that this is not a new insight – this is what merger simulation tried to do. But we felt that UPP is a very important additional tool that can at least in some cases work better than merger simulation and indeed better than market definition.

18  Merger Control in European and Global Perspective Now we’re not always going to have the margin data or the diversion ratios necessary to apply UPP or any of its analogues. Kirsten Edwards highlighted some of the difficulties of getting diversion evidence. In some cases, the information falls right out of the parties’ win/loss records, and in a number of cases we’ve been able to calculate diversion ratios without surveys, and without an enormous amount of work. But then again, one can analyse cross-elasticities of demand when you have scanner data, for example. The best evidence that is available to us is the evidence that we should use. In some cases that will lead us into a more conventional market definition effort, and in some cases the data will spare us that effort and allow us to go directly to an effects analysis. So the utility of the new methodologies is that they restore the idea that determining competitive harm is the ultimate objective of merger analysis, and that the ritual of finding market boundaries is really only a means, sometimes necessary sometimes unnecessary, of getting to that objective. On the other hand, once again the use of UPP does not mean that market definition vanishes. If we calculate UPP, or a GUPPI, in a case involving a merger between A and B, there is going to be some upward pricing pressure. As Simon mentioned, and as Joseph Simons and Malcolm Coate have pointed out, any time you have positive margins and a positive diversion ratio, you get a positive value for upward pricing pressure.27 So given this fact, how do we know what an appropriate level of upward pricing pressure should be in order to assess whether the Agencies should intervene? This is an open and challenging question. We can give a standard deduction of 10% for efficiencies, or there might be other approaches, but this is something that is not resolved by the Guidelines. As Steve Salop puts it, what level of GUPPI equals a SSNIP?28 We don’t really know at this point. We certainly don’t want to be intervening in a merger that reduces eight players to seven, or something like that, just because we get a positive value for UPP. Then again, under the 1992 Guidelines, we haven’t intervened every time we find an HHI of more than 1800 and a delta of more than 200. So there again, collateral considerations are going to come in. But one of those collateral considerations will be whether we’ve calculated the diversion ratios correctly, and indeed, whether there would be rapid entry or competitive responses that we haven’t considered. This gets us back to where market definition might still be relevant even when we conduct more direct, effects-oriented analysis. We can certainly expect that the parties will bring this into the discussion. But notice that, here, the debate over market boundaries comes only after credible evidence of effects has been put on the table, and the debate proceeds with reference to those identified effects. Traditionally, the debate preceded the identification of effects, and it was a largely hypothetical debate. There are many cases in the US – Oracle/PeopleSoft29 might See Coate and Simons, cited above note 24. See, eg, Steve Salop and Serge Moresi, Hearings Statement, 3 December 2009, http://www.ftc. gov/os/comments/horizontalmergerguides/545095-00032.pdf, at pp. 8–9. 29 United States v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004). 27 28

Introduction and Panel 1 

19  

be the most extreme example but we can see it in Whole Foods30 and other cases – many cases where there is a hypothetical discussion of market boundaries quite divorced from any discussion of effects. In these cases, the judge might not even let you get to your effects analysis until you’ve gotten over the hump of specifying your market boundary. By contrast, when we start with effects before proceeding to that debate, the burden is shifted to the defendants. But this shift of the burden is purchased by the Agencies only with credible evidence of effects, and this is a very healthy and defensible enforcement approach. Let me finish with some brief remarks about how we have treated innovation in the Guidelines. We have an expanded discussion in Section 6.4 of non-price effects. That section talks about product variety and innovation. Now we’ve had some recent cases at the FTC that illustrate how we will calculate effects on innovation, and one such effect is what I call ‘DIP’, that’s my new acronym – downward innovation pressure. What is downward innovation pressure? You can see it in some mergers between A and B, such as in the case of a proposed medical device merger that we blocked between Thoratec and Hardware,31 where the incentive of the merged entity to bring B’s new product to the market is quantitatively much less than the standalone incentive of B to bring that product that it was developing to the market in the absence of the merger. In some cases you can show that the profit incentives will be significantly reduced if the merger goes forward, and this follows on largely from the cannibalization effect that Arrow showed many years ago – the merged firm will have to take into account not only the profitability of the new product but also the degree to which sales of the incumbent product are cannibalized by the new product.32 So the entry conditions for the new product from the perspective of the merged firm will be more stringent than they are for the standalone firm. There are many caveats to that – many cases where the reduction in incentives won’t happen, or cases where the diminished innovation incentive cannot be accurately calculated. But when it can be done, it is a unilateral innovation effects analysis that is analogous to the UPP analysis. We had another merger that we cleared at the FTC that highlights the kinds of issues that Irwin Stelzer raised. What do you do in a highly dynamic market where the number 1 and number 2 players propose to merge, but the market is two years old, and you have no idea how that market is going to evolve? This is a very difficult kind of case for the Agencies. The case I’d refer to is Google/Admob.33 Admob was leading the field in the placement of advertisement in global applications, and Google had a pretty good track record. The question was should we allow these  FTC v. Whole Foods Mkt., Inc., 548 F.3d 1028 (D.C. Cir. 2008).  Immediately after the FTC filed an administrative complaint on 30 July 2009 the parties announced that they were abandoning the planned merger. See http://www.ftc.gov/os/adjpro/d9339/ index.shtm. 32  Kenneth Arrow, ‘Economic Welfare and the Allocation of Resources for Invention’, in The Rate and Direction of Inventive Activity: Economic and Social Factors, National Bureau of Economic Research, 1962, pp. 609 et seq. 33  See FTC statement closing the investigation, http://www.ftc.gov/os/closings/100521googleadmobstmt.pdf. 30 31

20  Merger Control in European and Global Perspective two firms, who are number 1 and number 2 in the global applications ad market, with no one close to them, to merge? We were quite hesitant, and we worried that there might be platform tipping. But we also knew that there was an 800-pound gorilla in the room, Apple. And Apple helped us out in this case. We were having a big hypothetical debate about what Apple might do. But given our uncertainty about that we had a philosophical debate within the FTC about whether we should authorize the merger. That philosophical debate remains unresolved; we didn’t have to resolve it because Apple then announced that it was going to favour its own advertising, so we attributed to Apple a substantial market share, we treated them as a market participant and we let the merger go through. As for the best approach in such cases when we don’t have convenient market developments such as Apple’s announced business plans, I personally sympathize with Irwin’s point that the Agencies should proceed with an appropriate measure of modesty. That’s all, thanks. Fingleton: Thank you, Howard. You rightly made the point that UPP is about a subset of mergers. With retail and bricks and mortar mergers we can do that kind of analysis on a fairly predictable basis, but then as you and Irwin point out we also have a broader canvass that includes cases that require an adapted approach, and one that encourages product and service innovation. Our penultimate speaker is Scott Hemphill of Columbia University. Scott Hemphill: Good morning, everyone. As a preliminary point let me raise the question of how educators in the law field might try to teach students about concepts like upward pricing pressure, because I do think it’s more challenging than the more familiar tools like market definition. But coming to my remarks, I’d like to address, first of all, two defences that sit at the intersection of economics and law, and two puzzles that we can think about. I’ll do this by reference to a US merger case called Ovation.34 Ovation was a pharma company that owned a drug called Indocin, it’s a drug used to treat a disease afflicting premature babies called PDA. But Ovation also bought the rights to a second drug which was the only other treatment, apart from surgery, for that disease. That second drug is called NeoProfen. And there were some shocking facts in this case, because after the merger the price for the treatment jumped from less than a hundred dollars to more than a thousand dollars, a price increase of 1200 or 1300 percent. Now the Federal Trade Commission and the Attorney General of the State of Minnesota challenged this acquisition of NeoProfen, and lost. I’ll come back to why they lost in a minute. These facts help to set up two puzzles about US merger policy. The first puzzle was not something I’m drawing from the Ovation case, but the facts allow us to think about this. Let’s assume that Ovation decided to acquire NeoProfen because otherwise its profits would be lower. Lower profits mean less innovation, and 34 FTC v. Lundbeck, No. 08-6379, 2010 WL 3810015 (D. Minn. 2010), aff’d, 650 F.3d 1236 (8th Cir. 2011).

Introduction and Panel 1 

21  

that’s bad for society. Now at first glance that seems kind of crazy. In general, we don’t wave through mergers that eliminate price competition just because it might promote innovation. Consider the following statement from the revised US Horizontal Guidelines: ‘The Agencies also consider the ability of the merged firm to appropriate a greater fraction of the benefits resulting from its innovations’.35 That language seems to provide for a defence based on increased appropriability. And if we took that in a certain direction it would generate unusual results. Instead of liking maverick firms and disliking their removal from the market, we would come to opposite conclusions because mavericks increase price competition and that might be bad for appropriability. It seems unlikely that the FTC and the DOJ would take the quoted language in that direction. I’m not sure about how judges might take it. But there are some variations on this that we might want to think about. Suppose instead that Ovation says ‘I’m great at making sales. The guys at Neo are terrible at this; Neo’s a pure technology company.’ Or Ovation might say, well, Neo doesn’t know very well how to get regulatory approval for its product, but we’re good at that. And those kinds of situations might fall within an appropriability defence. Or a third scenario, which comes closer to the actual facts of the case, is where Ovation is having problems appropriating the value of its original product, Indocin, and indeed Indocin was not covered by a patent. But NeoProfen is patented, and Ovation wants to execute a classic pharma industry product switch. So my appropriability strategy would be to acquire NeoProfen and to make sure that prescribing doctors and customers move from Indocin to NeoProfen. So that’s the first puzzle, and I’d interested to hear views about whether the Guidelines might open the door to those kinds of arguments. Puzzle number 2 is closer to the way the Ovation case has been argued. The District Court in Ovation basically argued in three steps. First, there is essentially no price competition between Indocin and Neo. Doctors are the decision makers, they’re sticky, they wouldn’t switch even if offered a 10, 20, 30 percent discount. Therefore, step 2, these drugs are not in the same market. Therefore, step 3, the FTC loses. I should say that I’m not convinced that the District Court got that right, either as a matter of understanding price competition or in its evaluation of nonprice competition, because there might well have been a reduction of non-price competition. But there is a doctrinal puzzle here, so I want to accept everything the District Court thought about price competition. It seems to me that the way the District Court sees this is that Indocin is a monopoly. If there is no other competitor besides Indocin, then Ovation has serious market power. But to view that as a defence justifying a merger seems to me really quite anomalous. To me it seems a bit like Microsoft.36 Imagine that, instead of excluding Netscape, Microsoft has just offered to buy Netscape. I find it difficult to believe that a District Court judge would say, well in light of the fact that Microsoft has a monopoly in the operating 35 US DOJ and FTC, Horizontal Merger Guidelines (August 2010), http://www.justice.gov/atr/ public/guidelines/hmg-2010.html, § 10, at p. 31. 36 United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001), remanded to 231 F.Supp.2d 144 (D.D.C. 2002), affd sub nom. Massachusetts v. Microsoft Corp., 373 F.3d 1199 (D.C. Cir. 2004).

22  Merger Control in European and Global Perspective system market, it’s fine for them to buy this threatening complement that might become a substitute, given that Netscape is outside the product market. If we adopt the perspective of Section 2 of the Sherman Act, I would have trouble seeing how, in the Ovation case, it should be ok for a monopolist to buy its nearest rival, albeit one that technically fell outside of the relevant market. So my question there is: is there anything right about this? Thanks. Fingleton: Thank you very much. Again, some interesting issues concerning how we should apply merger policy in innovation markets. We turn finally to Barry Hawk, a man who I think has seen it all by now, probably, in antitrust. Barry Hawk: Thanks, John. I’m just going to raise two questions. And I’ll relate each of these questions to historical figures. First of all, we’ve talked about UPP, GUPPI, and now we have DIP … then there’s IPR, which I always thought meant intellectual property rights. And Amelia Fletcher told us this is a great leap forward. Well, one leap forward was Neil Armstrong landing on the moon, but the other great leap forward was Mao Zedong. (laughter) I can only report that there has been some severe criticism by economists in the US about the ripeness of these economic theories. If Dennis Carlton were here, and if we asked him is it Neil Armstrong or Mao Zedong, he’s say it’s Mao Zedong.37 But this takes us back to a fundamental problem in antitrust. How do we use economics to generate clear, cost-effective legal rules? A brilliant example of how economics can influence legal rules is resale price maintenance. Economists have derived a series of factors we can use to assess the competitive impact of resale price maintenance, and every legal system in the world should use these. But it’s not clear whether the new concepts in merger control – GUPPI and DIP and all the rest – have been as successful. So, are those concepts really ripe for use in legal investigations and litigation in the courtroom, that’s the question. For the second question, the historical figures are King Henry II and Becket. I sometimes hear representatives of competition authorities – no one here this morning, but in other settings – I’ve heard many competition authority personnel and economists expressing their frustration with ‘the courts’. This always reminds me of King Henry II, who says, ‘Will no one rid me of this meddlesome priest?’.38 And as far as the US is concerned (I limit my remarks to the US), the question is whether it is appropriate for a competition authority to take a position in merger control which is more aggressive, more prohibitive, than what the courts accept. One can argue about whether the US authorities really are adopting a more intrusive position than the courts, but let’s assume arguendo that they are in fact more aggressive than the courts. After all, we’re talking about shifting the burden of proof to the defendant on market definition. And think about what’s going on 37 See Dennis Carlton, ‘Revising the Horizontal Merger Guidelines’, 6 Journal of Competition Law and Economics 619 (2010). 38 The bloody deed was done in Canterbury Cathedral in December 1170, and Becket was made a saint.

Introduction and Panel 1 

23  

here – if we look specifically at the Department of Justice, the Department is part of the executive branch of government. And it’s saying, in effect, ‘we don’t like the law’ as interpreted by another branch of government, the judiciary. I’m troubled by the question of how long an authority can do that. If an authority has a new vision and wants to apply new rules and methodologies, it has got to persuade the courts that its vision fits within the law, and the courts may say we can’t do that under the statute we have here. Of course, on the European side, where you have an ongoing debate about whether effects matter in the enforcement of Article 102, it’s up the Commission to persuade the EU Courts that that’s what the law is. But if courts say no to innovative propositions about how to interpret the law, then what? I’ll stop there, thanks. Fingleton: Thank you, Barry. I’m impressed to see that the 12th Century Catholic Church is seen as the absolute standard in the rule of law that the courts should aspire to. (laughter) Well, thanks to everybody, we’ve heard some great speakers. We were also supposed to hear from Lars-Hendrik Röller, and Hendrik couldn’t be here with us but I’d like to mention a point that he makes in his paper which I think is relevant to our discussion. As he explains, efficiencies have not played a critical role in any mergers reviewed at the EU level. Hendrik raises the interesting point that parties may not want to claim efficiencies because it may be taken as a desperate signal, as he puts it. Then he mentions two cases where the Commission itself has introduced efficiency arguments even though the parties didn’t raise them. This has been our experience at the OFT as well. We have had to make the efficiencies arguments for the parties because the parties have failed to make them at Phase I. This raises the fascinating question of why the parties don’t trust us, because normally if merging parties rely on efficiency arguments we’ll try to take that seriously and we won’t try to read some desperate signal into it. * * * Due to flaws in the audio recording of the event, Dr. Fingleton’s remarks were followed on the tape by 40 minutes of dead silence.

Irwin Stelzer*

Merger Policy and Schumpeter’s Creative and Destructive Gale

The organizers have advised me to craft a paper that is designed to stimulate discussion; I have even been free to eschew footnotes and other scholarly paraphernalia, which I have tried, not entirely successfully, to do. What I propose to do here is avail myself of what I understand to be the virtues of the discussion format – to lay out thoughts and problems and then wait for the discussants to clear things up for me. I might borrow from Alison Oldale’s formulation and say that the views I express might not even be my own1 – at least not after hearing your comments. Since the audience includes many regulators, who in the end decide these things, I will pay careful attention to all such comments.2 The first matter worthy of consideration when we reappraise current merger policy is the fundamental question of market definition, a key to deciding whether a merger is or is not likely to have an adverse effect on competition.3 Or some might say, in times past it was such a key. I have long had some doubts about the wisdom of that exercise, ever since Oxford Professor Yarrow raised the issue with me. Yarrow now has an increasing band of supporters. Louis Kaplow, of Harvard University and the National Bureau of Economic Research, recently made what he characterizes as an “immodest claim” that: the market definition process is incoherent as a matter of basic economic principles and should be abandoned entirely. This conclusion is based on the inability to make meaningful inferences of market power in redefined markets; … the impossibility of determining what market definition is best in a sensible manner without first formulating a best estimate of market power, rendering further analysis pointless and possibly leading to erroneous outcomes….4

An intriguing thought, but in the real world it is not entirely useful, for several reasons. First, as several papers point out, different national jurisdictions emphasize different tests, and market definitions remains attractive to several of them. Second, as Barry Hawk points out, there is no unanimity among economists * Hudson Institute, Washington, D.C. 1 Alison Oldale, “Market Definition Is Dead. Long Live Market Definition?”, presentation, EUI Competition Workshop, 12 November 2010. 2 Indeed, this version of the paper replaces the original draft so that I might incorporate some of the remarks made at the Workshop. 3 In the following paragraphs I draw on a talk I recently gave at a conference sponsored by the Regulatory Policy Institute at Merton College, Oxford. 4 Louis Kaplow, “Why (Ever) Define Markets?”, The Harvard John M. Olin Discussion Paper Series, Discussion Paper No. 666, March 2010.

26  Merger Control in European and Global Perspective and lawyers on this point and, third, there is the small matter of the courts, which in America at least cannot ignore the “line of commerce” provision of the governing statute, which requires that the requisite lessening of competition be in a definable “line of commerce.”5 That sounds to this economist as demanding some sort of market definition, perhaps for the use suggested by Ms. Oldale – as “part of the organising framework for the competitive assessment.”6 In any event, it is clear from the 2010 Horizontal Merger Guidelines that market definition is still regarded by regulators as an important analytical tool. My guess is that we will end up with some combination of reference to market definition – perhaps renamed to obscure the fact that the tool so derided by so many is still in use – and the alphabet soup of newer measurement devices discussed by Amelia Fletcher who is careful to point out that “market shares and concentration measures… still have a valuable role to play when accurately assessed.”7 This brings me a worry and a suggestion – we have to pay less attention to static definitions of market share and more to the durability of existing shares in the face of a gale of creative destruction. In short, the question should not be, “Does a company have a dominant market share?” – or at least that should only be the starting point in an inquiry – but “Is that market share likely to prove ephemeral?”. I suppose this is merely another way of asking some of the questions we have always asked: Are switching costs high? Can the incumbent make entry difficult and costly? Does the incumbent have a history of anticompetitive behavior? Is the nature of likely entry such that new entrants will merely nibble around the edges of the currently dominant firm, or is the incumbent more likely to be swept away in the famous perennial gale of creative destruction? Perhaps these questions are nothing more than what Carl Shapiro has captured in his briefer formulation, that we have to recognize that “each industry has unique features” – an unfortunate fact for those lawyers who say they pine after certainty, which in the case of European merger control is declining, according to Professor Duso and his colleagues.8 If a seemingly dominant company in an industry of concern to the enforcement authorities does make an acquisition, and is subject to life-threatening assaults from new entrants sporting new technologies and better mousetraps, we might well need a more careful analysis of market-share or market-power durability than that in which we have traditionally engaged if we are to avoid Type 1 errors, which Duso et al. report to be on the increase.9 As Commissioner Almunia recently 5 Barry Hawk, “A Tale of Two Cities: Washington and Brussels Face the Courts”, this volume, pp. 53 et seq. 6 Oldale, cited above note 1, p. 8. Oldale added that “[m]arket definition plays a role here in identifying the products of interest, identifying horizontal overlaps and potential non-horizontal concerns, identifying those distinct markets requiring a separate analysis, focusing the process of evidence gathering and so on.” 7 Amelia Fletcher, “UPP: Up and Away? The Rising Prevalence of Pricing Pressure Indicators”, Presentation, EUI Competition Workshop, 12 November 2010. 8 Tomaso Duso, Klaus Gugler and Florian Szücs, “Merger Policy Evaluation: Where Do We Stand?”, this volume, pp 105 et seq. 9 Ibid.

Irwin Stelzer 

27  

remarked, we must “allow competition policy to look forward on the basis of innovation and not only on prices and output”.10 Please understand: I am not saying that these industries warrant special exemption, or inattention to any anticompetitive practices in which they might engage.11 In a world in which intellectual capital is rapidly replacing physical capital as the cornerstone of the wealth of nations, nothing can be more important than applying merger policy and, indeed, all aspects of the antitrust laws, with strict attention to the anticompetitive effect of certain trade practices, and the potential of any merger to affect competition adversely. But I am saying that we might want to adopt a merger policy that distinguishes between two cases, in both of which the acquiring firm has a dominant position, or is about to acquire one, as we traditionally measure such a position. In the first case, we are dealing with a dominant firm that might have its market share eroded a bit by new entrants: we can safely use its current position as some indication of dominance. In the second case, we might be dealing with an acquisition by a firm with a dominant position, as traditionally measured, but a position subject to destruction by a creative gale generated by a new entrant.12 Are we to treat these two cases as if there were no such difference? So I would suggest, for discussion, the question of whether what is required to accommodate the new, more rapid rate of devastating innovation that characterizes many markets is a shift in emphasis from current to prospective market shares, a consideration not only of the current state of play, but of what is looming on the horizon. This is especially important in industries that are now coming in for increased regulatory attention,13 and are included in the list that Commissioner Almunia recently cited as “high on our enforcement agenda”.14 For two reasons.

10 Joaquín Almunia, “New Transatlantic Trends in Competition Policy”, SPEECH/10/305, Brussels, 10 June 2010, p. 18. 11 In the next section I draw on a talk I delivered to The Federalist Society for Law & Public Policy in Washington, D.C. on 11 November 1998. 12 See, e.g., Amir Efrati, “Start-Up Aims at Google”, The Wall Street Journal, 1 November 2010, the tale of Blekko, Inc.’s challenge to Google. Or consider the fate of Myspace, which ceded so much market share to upstart Facebook that at this writing its fate is uncertain, or of Kindle, a venture of Amazon that almost immediately found its dominant market share eroded by iPad, which in turn faces new competition from a reconstructed and repriced Kindle and other devices. 13 “New leaders on both sides of the Atlantic are showing renewed interest in scrutinizing innovators. Competition agencies ideally help consumers by ensuring open, competitive markets while eschewing actions that impede innovation and competition. But the accelerating pace of technological change makes their task more difficult.” Timothy Muris, “Antitrust in a High-Tech World,” The Wall Street Journal, 12 August 2010. One such firm, Google, is a client of this writer, and others that might have some interest in this topic, once but no longer clients, include BSkyB, News Corp, News International, and AMD. 14 Joaquín Almunia, “State of Play and Future Outlook, Competition Day”, SPEECH/10/576, Brussels, 21 October 2010.

28  Merger Control in European and Global Perspective First, these industries are in a state of flux: previously unimagined and unknown businesses can suddenly attract a half-billion users; advertisers are continuously calculating and recalculating the marginal cost-to-marginal revenue relationships from various expenditures; in some aspects competitors are only a click away and eager to serve consumers. In these instances it seems to me that it behooves regulators to think carefully before concluding – from traditional measurements of market share, which are uncertain guides in any event – that a firm is dominant in any sense that should be considered when appraising the competitive effects of a proposed merger. Second, we find ourselves at a peculiar time. The firms that we might for these purposes call high-tech firms – the Googles, Microsofts, Apples, Twitters, Facebooks – are in most cases awash in cash. If they are to grow,15 they will have to rely at least in part, and perhaps in major part, on acquisitions, generally of firms that, added to their existing products, will enhance the package of services they offer to their customers. Often, these service and product enhancements make life easier for customers, as Intel argues will be the result of its acquisition of McAfee16 without, as was the case with Microsoft (a special case anyway because its history of anticompetitive practices casts a shadow over its current market position17) adversely affecting the ability of newcomers to enter any market. Note: any such enhancement of the consumer experience is different from claimed efficiencies. Professor Röller, in his paper for this workshop, reports that “efficiencies have not played much of a role in merger assessments”.18 Unfortunately, Professor Röller finds that static efficiencies have been at times “accepted by DG Comp” while “alleged dynamic efficiencies have never been accepted”.19 I say “unfortunately” for two reasons. First, most studies have found, and this is confirmed by real-world experience, that claimed reductions in “variable and marginal cost” – the static efficiencies most likely to be passed on to consumers – have very rarely been realized in practice. Second, it is dynamic efficiency that we are after – the driver of economic growth and improved living standards. The fact that they cannot meet the verifiability, merger-specific and consumer benefit tests is understandable: they are not immediately realizable, and it takes something of an act of faith to believe they will appear as a result of a merger. Which brings me back to my suggestion: in merger cases we can 15 Some investors would prefer share buy-backs or higher dividends to growth-by-acquisition, but that is rarely a popular strategy with managers who have so far succeeded in producing double-digit growth for shareholders, and are confident they can continue to do so, this time around by adding strategic acquisitions to continued organic growth. 16 See Don Clark, “Intel Looks to Protect More Than Computers,” The Wall Street Journal, August 20, 2010. Intel claims that the $7.7 billion acquisition (a 60% premium) will result in new chips that better protect PCs and other devices without slowing them down, and among other things allow consumers to remotely turn off stolen or lost cell phones. See also Eric Savitz, “Intel’s Brilliant – or Bonehead – Deal, Barron’s, 23 August 2010. 17 It seems obvious to me that Microsoft did not obtain its market position merely by the exercise of “superior skill, foresight and industry,” to borrow from Judge Learned Hand’s Alcoa decision. 18 Lars-Hendrik Röller, “Efficiencies in EU Merger Control: Do They Matter?”, this volume, pp. 61 et seq. 19 ibid.

Irwin Stelzer 

29  

incorporate (as a substitute for the well-nigh-impossible task of identifying dynamic efficiencies) the possibility of dynamic efficiency gains by giving weight to the destructive potential of new entry when assessing “dominance”. Of course, some would argue that the acquisitions I have been discussing represent vertical integration, and therefore cannot add to existing market power. I would disagree. Although “such non-horizontal mergers are, in principle, less likely to create competition concerns than horizontal mergers”,20 as Dr. Lübking points out, in some instances vertical mergers, in his words “may significantly impede effective competition”, perhaps by adding to the market power of the acquirer by making entry by rivals other than offerers of a similarly full line more difficult, or, put differently, by reducing the ease with which small, specialty firms can carve out a niche for themselves. On the other hand, regulators generally, or at least the European Commission, recognize that “vertical and conglomerate mergers provide substantial scope for efficiencies…”.21 My own guess is that the great majority of the vertical and conglomerate mergers we are likely to see in the industries that are attracting increasing attention – or will when they dip into their substantial cash piles – will pass muster as providing considerable customer benefits with little danger of creating barriers to entry or enduring and unacceptable market power. Still, it will take analysis of the specifics of each merger to separate the pro-competitive and competition-neutral mergers from those that are likely to have significant anticompetitive effects and, perhaps, to identify dynamic efficiencies.22 I raise these questions for discussion because I find myself uncertain about appropriate policies to deal with these difficult issues. And because if we get merger policy wrong, we will either prevent mergers that might benefit consumers, or permit those that confer exclusionary power on the merged entity, eventuating in an expensive effort to uncover and force abandonment of those practices – with even what seem the most draconian fines hardly making a dent in the ill-gotten gains accruing from such behavior, although if Commissioner Almunia succeeds in setting “the cost of non-respect of competition rules” higher than the projected gains from anticompetitive behavior, that situation will change.23 Do forgive me for the ambiguities in this paper. They are a testimonial to my faith in this workshop to provide answers to difficult questions, answers that have eluded me.

20 Johannes Lübking, “Some Reflections on the Application of the Substantive Test and of the Substantive Merger Guidelines”, presentation, EUI Competition Workshop, 12 November 2010. 21 Ibid. 22 For the role of static and dynamic efficiencies in merger regulation see Lars-Hendrik Röller, cited above note 18. 23 See Joaquín Almunia, “Compliance and Competition Policy”, SPEECH/10/586, Brussels, 25 October 2010.

Kirsten Edwards*

Estimating Diversion Ratios: Some Thoughts on Customer Survey Design

A. Introduction The recently updated merger guidelines in both the US and the UK downplay the role of market definition, as compared to earlier versions. These revisions follow recent cases and economic literature which highlight the difficulties of defining markets when products are differentiated, and throw doubt on the relevance of the subsequently calculated market shares.1 Both sets of guidance explicitly discuss alternative ways of assessing mergers involving differentiated products, and state that the authorities may seek to estimate diversion ratios to quantify the closeness of competition between two products.2 A number of screens for economic effects have emerged to replace market share screening in mergers. Since 2006, the UK OFT has calculated “Illustrative Price Rises” (“IPRs”) to screen first phase differentiated product mergers.3 In 2008, Joe Farrell and Carl Shapiro, who lead the economist teams at the FTC and DOJ respectively, proposed the Upward Pricing Pressure (“UPP”) model for merger screening. The IPR and UPP screens are appealing because they only * Director of Economics, Office of Fair Trading (At the time of writing: Principal, Compass Lexecon). I have benefited from discussions with participants at the 2010 EUI Workshop, and would like to thank, in particular, Amelia Fletcher and Chris Walters for comments. The usual disclaimers apply. 1 See Joseph Farrell and Carl Shapiro, “Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition”, 10(1) B.E. Journal of Theoretical Economics, Article 9 (2010). The authors discuss the Whole Foods/Wild Oats merger in which defining markets led to some odd results. Whole Foods and Wild Oats were national supermarket chains specializing in natural and organic goods. The Federal Trade Commission defined the market narrowly to include only premium natural and organic supermarkets. On this basis the parties were each other’s closest rivals, with 100% market share in many of the geographic markets. The FTC accordingly sought to block the merger. However, the District Court ruled that the evidence instead indicated a relevant market that included the product range of conventional supermarkets as well. On this market, the parties would have a very low market share, suggesting that the merger would not harm competition. The District Court’s decision on market definition was later overturned on appeal. See Federal Trade Commission v. Whole Foods Market, Inc., and Wild Oats Markets, Inc., 533 F.3d 869 (D.C. Cir. 2008). 2 US DOJ and FTC, Horizontal Merger Guidelines (August 2010), http://www.justice.gov/atr/ public/guidelines/hmg-2010.html, page 20. UK CC and OFT, “Merger Assessment Guidelines”, CC2 (revised) and OFT1254 (September 2010), http://www.oft.gov.uk/shared_oft/mergers/642749/ OFT1254.pdf, page 32. 3 See, e.g., CGL/Somerfield and LOVEFiLM/Amazon. The “IPR model” is based on a model by Shapiro. See Carl Shapiro, “Mergers with Differentiated Products”, 10 Antitrust 23 (Spring 1996).

32  Merger Control in European and Global Perspective require estimates of two parameters – diversion ratios and margins – and it is these two parameters which are the key drivers of price increases in mergers involving differentiated products. However, while the IPR and UPP screens are easy to implement, obtaining estimates of diversion ratios can be tricky. There are different approaches. Diversion ratios can be estimated from sales data, win/loss data or through event studies, if data are available. They can be approximated using market shares, but this defeats the object of avoiding the complexities of defining markets for differentiated products. They can also be estimated using customer survey data, and it is this method that I discuss in this paper. The design, implementation and interpretation of survey data is well understood by statisticians and market researchers. There is a vast body of academic literature which draws on lessons from research in the health, pensions and political polling areas in particular, where surveys have been implemented, repeated and refined over many years.4 In addition, following a consultation on the design and presentation of consumer survey evidence on merger guidelines, the OFT and CC jointly published a “Good practice” document on that subject,5 and the CC has also commissioned a study of the stated preference methods for analyzing consumer willingness to pay.6 This paper highlights some lessons which can be drawn on by competition policy practitioners when relying on survey evidence to estimate diversion ratios. Section B reviews a selection of past cases in which surveys have been used to estimate diversion ratios. I compare and contrast the different question formats used. Section C discusses how survey data should be interpreted, taking into account the various biases that can arise in using revealed preference data to estimate diversion ratios. Section D presents some lessons for future survey design.

B. Different ways of asking the “Diversion question” A diversion ratio is the fraction of unit sales lost on one product, due to an increase in its price, that would be diverted to a second product. Diversion ratios between the products sold by merging parties can be very informative for assessing unilateral price effects, with higher diversion ratios indicating a greater likelihood of such effects. Diversion ratios from one set of products to another can also inform market definition (and can be used to derive price elasticities). 4 For a useful survey of the literature, see Daniel McFadden et al, “Statistical Analysis of Choice Experiments and Surveys”, 16(3–4) Marketing Letters 183 (2005). 5 UK CC and OFT, Good practice in the design and presentation of consumer survey evidence in merger inquiries, CC2com1 and OFT1230 (March 2011), http://www.oft.gov.uk/shared_oft/ consultations/merger-inquiries/Good-practice-guide.pdf. 6 Accent and Rand Europe, “Review of Stated Preference and Willingness to Pay Methods” (April 2010), http://www.competition-commission.org.uk/our_role/analysis/summary_and_report_ combined.pdf.

Kirsten Edwards 

33  

When diversion ratios are measured using surveys, they rely on customers making judgements on how they would behave in a hypothetical situation. This is a non-trivial task for survey respondents, so thought needs to be given as to how to design questions to elicit the most accurate responses. There is no consensus on how this should be done when collecting data to estimate customer diversion. Competition authorities and investigated parties have tried a number of different techniques. The “diversion questions” are broadly in one of three formats: • What would you do if the price increased by 5–10%? • What would you do if this product was unavailable? • If the following options were available to you, which would you choose? The first two types of question are known as “contingent valuation” questions, which seek to directly extract the relevant information from the respondent. They may or may not provide a set of prompts from which the respondent can choose. The latter question is typical of a discrete choice methodology which involves providing the respondent with various scenarios that each involve choosing a product from a small selection. The “products” are described by their key characteristics (e.g. price, size, colour). In each scenario one or more product attributes change(s) from the previous scenario. This type of exercise allows the analyst to indirectly assess each customer’s preferences from the choices that he or she makes. There have been many cases in Europe which have used surveys to understand customer switching behaviour, in which the diversion question has cropped up in various guises.7

KLM/Martinair (European Commission, 2008, unconditional clearance) KLM/Martinair involved the merger of two airlines which both offered direct passenger flights from Amsterdam (Schiphol) to various destinations in the Caribbean. The European Commission commissioned a survey which was carried out at Schiphol airport. It surveyed over 1,000 passengers travelling to six destinations in the Caribbean, flying either with the parties or with two other carriers (Arkefly and Surinam Airways). The survey was primarily used to understand switching between destinations. Passengers were first asked whether they would switch to another destination in response to a 5–10% price increase. The subset of passengers who responded positively to this question (137 passengers) were then asked which other destination they would have chosen. Diversion ratios were calculated for each 7 The author was engaged by KLM during the KLM/Martinair transaction (M.5141, 17 December 2008) while at LECG, and also worked on the Somerfield/Morrison merger (http://www.competitioncommission.org.uk/our-work/directory-of-all-inquiries/somerfield-plc-wm-morrison-supermarketsplc) and the NEG/Thameslink merger (http://www.competition-commission.org.uk/our-work/ directory-of-all-inquiries/national-express-group-thameslink-great-northern-rail-franchise) at the Competition Commission.

34  Merger Control in European and Global Perspective route, showing which other destinations the vulnerable (marginal) passengers would switch to. These ranged between 13–34%. The survey also asked which other airlines passengers had considered travelling with, and concluded from the responses that KLM was a stronger competitive constraint on Martinair than vice versa.

Ryanair /Aer Lingus (European Commission, 2007, prohibition) Ryanair and Aer Lingus are both airlines with a base in Dublin, with overlapping routes to a number of European destinations. During its investigation, the Commission commissioned a survey on 12 of the 35 overlap routes. This survey also asked a two-part question to understand switching between the parties, but posed the questions in an order that differed from that used in KLM/Martinair. First, the survey asked “Which other airlines, if any, did you consider using for this route?”

and then followed up with “If the ticket price offered by an alternative airline had been 10% lower than the price you paid, would you have purchased that ticket instead”.

Over the 12 routes, on average 40.9% of Ryanair passengers said they had considered travelling with Aer Lingus, and 41.3% of Aer Lingus passengers said they had considered travelling with Ryanair. However, none of the responses for the second part of the above question – i.e., which of these passengers would switch in response to a price decrease – is published. The survey also asked passengers whether they would ever consider a flight to/from Belfast as an alternative to using Dublin airport. Seventeen percent of the passengers replied that they would. The Commission did not consider this sufficient to include the two airports in the same market. The Commission prohibited the merger. In its appeal to the General Court, Ryanair argued that the passenger survey was “‘seriously flawed’ both in terms of the design of the questions and the sampling techniques employed”.8 The court rejected this claim.

Somerfield/Morrison (UK CC, 2005, cleared with divestments) Somerfield/Morrison was a merger between two major supermarket chains in the UK. There were a number of local areas identified by the CC where the number of supermarket fascia was reduced from four to three or fewer as a result of the merger. To investigate competitive effects in these areas, the CC commissioned a survey of 5,235 shoppers, which was carried out at 54 Somerfield stores. 8

Case T‑342/07, Ryanair Holdings plc v. Commission [2010] ECR II-3457, para. 97.

Kirsten Edwards 

35  

The survey asked: “If this Somerfield store had not been available, which, if any, of these types of store would you have used instead?”9 The prompts provided included “a large out-of-town supermarket”, “a same size supermarket”, “a smaller convenience store” as well as other smaller shops. Those respondents who said they would use another supermarket or convenience store were then asked which particular store (fascia) they would have shopped at (with a list of fascia prompts provided). The CC required remedies in 12 local areas, each of which reported a diversion ratio from Somerfield to Morrison of 14% or more (translating into an “IPR” result of 5%). Subsequently, the OFT used this methodology in first phase retail mergers including CGL/Somerfield (2007) and Asda/Netto (2010). The OFT and CC endorse the above question format for diversion ratio estimates in their recent consultation on the presentation of survey evidence in merger inquiries.10

NEG /Greater Anglia (UK CC, 2004, unconditional clearance) The UK CC used a stated choice experiment to estimate customer preferences in the NEG/Greater Anglia merger. In this investigation of two rail franchises in the east of England, a survey was conducted on board the relevant train services. Passengers were provided with a number of scenarios from which they had to choose a mode of transport. An example of one of the scenarios is given below.11 SCENARIO Time on vehicle.................... Service frequency................. Time to and from stations.... Cost (Daily one-way)...........

Coach 3 hours 10 minutes Every 3 hours As now £12.00

Train 2 hours Every 30 minutes As now £13.00

Travel by coach

Travel by train

In this situation I would choose: To use neither

The data was analyzed using a multinomial logit model. It was used for two purposes: (i) to value different attributes of travel, in order to create a “generalized cost of travel” variable that encapsulates price, waiting time and an “interchange penalty”; and (ii) to estimate own and cross-price elasticities of demand. The results were used by the CC to decide this case, but to my knowledge the CC has not since carried out this type of experiment in a merger inquiry. NOP Consumer Research, Somerfield/Morrison Stores Merger Inquiry Final Report (July 2005). See note 5 above. 11 Reproduced from “Review of Stated Preference and Willingness to Pay Methods”, cited above note 6. The choices in the experiment were not published by the Competition Commission. 9

10

36  Merger Control in European and Global Perspective

C. Challenges of survey design The examples above show that there is no consensus on how questions should be asked in surveys to elicit likely consumer responses to a price increase. However, there is substantial evidence that suggests that the way a question is framed is critical to survey results. Furthermore, survey design should be well understood by the analysts who are using the results, to ensure that diversion ratios or elasticities are correctly interpreted. Survey design is concerned with minimizing survey error. Sources of survey error can broadly be categorized as: sampling error; coverage error; response error and measurement error. All of these have arisen in consumer surveys undertaken in view of merger reviews, and some are of greater concern than others.

Sampling error Sampling error occurs because only a subset of the population of interest is actually surveyed. The smaller the sample, the lower the degree of confidence we can have about an estimator. To illustrate, the width of the 95% confidence interval for an estimator taken from a sample of 100 would typically be ±10%. This falls to ±5% as the sample increases to 350–400 people in size.12 Sampling error has been a concern in a number of cases. It was a major problem in KLM/Martinair because the survey only asked the subset of passengers who said they would switch in response to a 5–10% price increase details on where they would switch. This meant that only 137 out of 1005 passengers (over six routes) gave their second choice preferences. On the Amsterdam-Havana route, where 20% of passengers said that they would switch in response to a price increase, the Commission reports that: “the indications which can be drawn from the customer survey on the Amsterdam – Havana route are rather limited. That is due to the small number of responses which were collected on this route (only 45 passengers travelling to Havana expressed a definite answer on the question relative to substitutability at destination). This is well reflected in the size of the 95% confidence interval for the loss of customers due to a price increase in the Amsterdam – Havana route: the confidence interval is, indeed, 8%–32%.”13

One way to increase the number of responses on diversion is to survey all customers on their switching preferences, not just those customers who say they would switch in response to a price increase. The OFT and CC joint guidelines on the presentation of survey evidence recommend asking all customers where they 12 Normal approximation, base percentage 50%. See Edith De Leeuw, Joop Hox and Don Dillman, “The Cornerstones of Survey Research”, in Edith De Leeuw, et al., eds., International Handbook of Survey Methodology, Psychology Press, 2008, Chapter 1. 13 M.5141, KLM/Martinair (17 December 2008), recital 294.

Kirsten Edwards 

37  

would switch if their first choice product was unavailable.14 This question format forces all customers to reveal their second choice product, and thus increases the size of the sample of responses for the diversion question. (Note that this gives rise to coverage issues – see below.) Sampling error is also a concern when cases involve many different markets. For example, in Somerfield/Morrison, 5,235 customers were surveyed in total, but there were on average only 97 responses per store. The CC carried out statistical tests to assess which of the store diversion ratios was statistically significantly above the chosen 14% threshold (although it still required divestment of the two stores which were not statistically above this threshold). The OFT does not appear to do this as a matter of course when looking at survey samples under 100 (though arguably as a first phase authority it errs on the size of caution). Sampling error can be reduced by increasing survey size. This is not always possible given the time and cost constraints of a merger investigation.15 However, the existence of sampling error should be well understood by analysts and decision makers, since it is relevant in understanding the robustness of a diversion ratio estimate.

Coverage error Coverage error is the percentage of the population of interest that is included in the sampling frame. Under-coverage occurs when some members of the population have no probability of being selected in the survey sample. Over-coverage occurs when people who are not in the population of interest are surveyed. Coverage error is only a concern if the people who are under-represented or over-represented in the sample have preferences that differ from those of the remainder of the population of interest. When a diversion ratio question is asked of all customers, there is over-coverage. A diversion ratio is usually interpreted as the proportion of customers who switch to product B in response to a price increase in product A. Here, the population of interest are the marginal consumers – those consumers who would switch in response to a price increase. If all consumers are asked where they would switch if their preferred products was not available, then the survey responses include those infra-marginal consumers who would not switch in response to a price increase.

See note 5 above. It is plausible that, in some merger investigations, the number of customers may be small and potentially all customers could be surveyed. 14 15

38  Merger Control in European and Global Perspective Infra-marginal consumers

Marginal Consumers

A diversion ratio calculated on the basis of those responses then takes into account the preferences of the average consumer (both marginal and infra-marginal) rather than just the preferences of the marginal consumers (in which we are really interested). In Somerfield/Morrison, the parties argued that marginal consumers are by definition more price-sensitive and that they have different preferences compared to those of the average consumer who responded to the survey. The CC was not persuaded. The problem of over-coverage should be dealt with on a case-by-case basis. When there is horizontal product differentiation (i.e., where products are differentiated in a way that cannot be ordered by, for example, style, taste or colour), then marginal consumers’ preferences might not differ systematically from those of average consumers. However, when there is vertical product differentiation (where products can be ranked according to differentiating factors such as quality), then it is plausible that marginal consumers – who are more price sensitive – are more likely to down-trade in response to price increases, whereas less price sensitive customers are more prepared to trade up. On this issue, the CC’s NEG/Thameslink inquiry provides an interesting case study. NEG owned the Gatwick Express rail franchise which serviced the London Victoria to Gatwick train route. It was bidding for another franchise which included the Thameslink train service between London Bridge and Gatwick Airport.16 Passengers were surveyed on both services. They were asked two questions (not sequentially) which sought to understand passenger switching behaviour: • They were asked what they would have done if the service they were travelling on was not available; and • They were asked what they would have done if their fare had been 5–10% higher on the service that they travelled on that day. The CC calculated diversion ratios on the basis of the first question (“average” passenger diversion ratios). The diversion ratios from the Gatwick Express to the Thameslink service were 14% for business passengers, 9% for commuters and 16

London Bridge and London Victoria are approximately 5 km apart.

Kirsten Edwards 

39  

17% for leisure passengers. The diversion ratios from the Thameslink service to the Gatwick Express service were 37% for business passengers, 73% for commuters and 46% for leisure passengers.17 The CC then compared these diversion ratios to the responses to the hypothetical price rise question. They presented a table showing the number of passengers who said they would switch between the parties following a 5–10% price increase (marginal passengers), with the number of passengers predicted to switch on the basis of the “average passenger” diversion ratios (shown in brackets below). Number of switching passengers

Business

Commuting

Leisure

Gatwick Express – Thameslink

8 (7)

0 (1)

11 (14)

Thameslink – Gatwick Express

4 (4)

0 (0)

11 (20)

For business passengers and commuters, there was no significant difference between the diversion reported by the average consumer and the diversion reported by the marginal consumer. However, the average leisure passenger was more likely to switch from the Gatwick Express to Thameslink (and vice versa) than the marginal leisure passenger. Marginal leisure passengers were more likely to switch to the (cheaper) Southern train services or coach services than the average leisure passenger. The differences are statistically significant.18 This is consistent with the argument that including infra-marginal passengers’ choices to calculate diversion ratios may overstate or understate the “true” diversion ratio (although there may be other reasons explaining the differences, such as question misinterpretation). In the above case, the CC did not further investigate the source of the difference.

Non-response error Non-response error occurs when some individuals do not, or are less likely to respond than others. This is a concern if those individuals differ from the survey respondents in a way that will bias the survey results. It may arise because people are more willing to answer surveys that are relevant to them. Non-responses should be identified, and statistical tests can be carried out to see whether, for example, a particular demographic is less likely to respond to the survey. Ex post weighting of survey results can also address low response rates.

17 The results are not surprising given that there was a third train service run by Southern between London Victoria and Gatwick. Passengers headed to or from Victoria on the Gatwick Express could therefore take a service from the same station. Passengers using the Thameslink would have to travel to or from Victoria to take an alternative train service or coach. 18 At the 1% level for Thameslink passengers and at the 10% level for Gatwick Express passengers.

40  Merger Control in European and Global Perspective

Measurement error Measurement error occurs when an individual’s response to a question is inaccurate. This error is particularly difficult to foresee and to measure ex post because it can arise for many reasons. There are a few types of measurement error that are of particular concern when estimating diversion ratios. Specifically: •





Framing effects arise when the way that a question is posed affects the answer. For example, if response prompts are provided, people have been shown to be more likely to choose the top few responses in written surveys and the bottom few responses in telephone/spoken surveys. In Ryanair/Aer Lingus, passengers were asked which other airlines they had considered flying on and were given the response prompts: (i) Ryanair, (ii) Aer Lingus, (iii) Other. This may have biased the results towards these airlines as opposed to others on the route.19 Hypothetical biases arise when respondents are being asked how they would behave in a particular scenario. Respondents may not properly construct the relevant scenario before answering, so they may not consider, for example, switching costs or transactions costs. Studies of consumer willingness to pay have found that participants don’t always take questions seriously and can be overconfident in their responses. Studies have also found that people are not very good at correctly interpreting percentages, and that translating percentages into natural numbers leads to better understanding of survey questions.20

All three of these problems could arise simultaneously in a hypothetical question which asks how someone would respond to a 5–10% price increase and then provides answer prompts. Measurement error is difficult to quantify ex post, but there are ways to minimize it ex ante. Some authors have also suggested that, when a question is particularly important and cognitively demanding, the responses can be improved by telling the respondent in advance that you would like them to think hard about that particular question (“cheap talk” scripts).21 There are also many other recommendations on how generally to minimize misunderstanding in questionnaires, such as: avoiding unfamiliar, ambiguous or technical words; clarifying time frames; refraining from asking multiple questions at the same time; specifying the format of response; and avoiding embedded assumptions about a respondent’s viewpoint. Furthermore, all of these potential problems might be identified in pilot surveys. Pilots are 19 Note that, as the survey was carried out over 12 routes, it may not have been practical for the market research company to identify all possible airlines on those 12 routes, or to tailor 12 separate surveys. 20 See, e.g., Ulrich Hoffrage et al., “Communicating Statistical Information”, 290 Science, No. 5500, 22 December 2000, pp. 2261–2262. 21 Jayson Lusk, “Effects of Cheap Talk on Consumer Willingness-to-Pay for Golden Rice”, 85 American Journal of Agricultural Economics 840 (2003).

Kirsten Edwards 

41  

identified as one of the most important stages of survey design in most of the survey literature.22

D. Concluding remarks Poorly designed surveys can affect customer responses, and the diversion ratios calculated by those responses, and, ultimately – if misinterpreted – the outcome of merger investigations. The elimination of error in survey design is next to impossible, in particular in the demanding time frame of a merger inquiry. However, thought can be given as to how one may: (i) minimize the potential error ex ante; (ii) understand any biases that have arisen ex post; (iii) adjust, where possible, the estimator (of, for example, a diversion ratio) for those biases; and (iv) take them into account in interpretation. Noted above are some ways of minimizing biases in questions about hypothetical price increases or lack of product availability. There are many other ways of improving survey quality. Pilot surveys and cheap talk scripts help both to identify potentially problematic questions and to engage the respondents in providing thoughtful answers. Rotating the response prompts that are shown to survey participants can reduce framing biases. Training and quality checks can be carried out by market research companies to check that interviewers understand the purpose of the questionnaire and that they are getting appropriate responses from participants. More research could be carried out to understand the best format for “diversion ratio” questions. Stated choice models, such as the one carried out by the CC in NEG/Greater Anglia, are thought to avoid many survey biases because they mimic the choice situation that a customer would face, rather than asking the customer to imagine it. For this reason, stated choice models are often recommended by market researchers over contingent valuation methods.23 However, they are more complex, more time consuming and more costly to administer. Experiments have been carried out in other fields to examine the best way to elicit information from people.24 In the field of competition policy, controlled experiments could be carried out to understand whether there are systematic biases in the way that we ask consumers about their switching behaviour. In the meantime, survey data can still provide useful estimates of diversion ratios, so long as the robustness of those estimates is properly understood. Analysts should investigate potential sources of bias to assess whether an estimate might be above or below the “true” diversion ratio. Furthermore, survey evidence See De Leeuw, Hox and Dillman, cited above note 12, Chapter 1. “Review of Stated Preference and Willingness to Pay Methods”, cited above note 6. 24 See, e.g., F. Thomas Juster, “Prediction and Consumer Buying Intentions”, 50 American Economic Review 604 (1960); Lusk, “Effects of cheap talk”, cited above note 21. 22 23

42  Merger Control in European and Global Perspective should be analyzed alongside other evidence gathered in the course of a merger investigation. Qualitative evidence might provide some rough guidance as to the expected value of diversion ratios (for example, in UK retail markets they have been explained by, inter alia, the number and proximity of stores in local areas).25 Checking the consistency of qualitative evidence and survey evidence may help to identify any unusual results which can then be further investigated.

25 Chris Walters, “Diversion Ahead! Approximating Diversion Ratios for Retail Chain Mergers”, Competition Commission Occasional Paper (November 2007), http://www.competition-commission. org.uk/our_role/analysis/diversion_ratios.pdf.

C. Scott Hemphill*

Higher Profits as a Merger Defense: Innovation, Appropriability, and the Horizontal Merger Guidelines

Parties to a horizontal merger may avoid antitrust liability by showing that the transaction will have a pro-competitive effect. The major focus of such an “efficiencies” defense is usually price — for example, that the merged entity will enjoy economies of scale, thereby lowering costs and putting downward pressure on price, ultimately to the benefit of consumers. A second, less frequent defense is to argue that the transaction is likely to increase innovation, again to the benefit of consumers. The revised Horizontal Merger Guidelines, issued by the U.S. Department of Justice and Federal Trade Commission (FTC) in 2010, expanded the treatment of innovation as a basis for an efficiencies defense.1 This comment examines an important but neglected aspect of the Guidelines’ discussion of innovationbased efficiencies. New language in the Guidelines suggests, as one way among several in which innovation matters, that a transaction might be approved if it permits a firm to “appropriate a greater fraction of the benefits resulting from its innovations.”2 In other words, higher profit is sometimes a merger defense. The Guidelines neither explain why increased appropriation should support clearance of a merger nor identify the limits of such a principle. This comment is an effort to fill that gap. The analysis proceeds in three parts. First, I introduce the “increased appropriation defense” as a type of innovation-based efficiencies defense contemplated in the revised Guidelines. Second, I present a transaction to which the increased appropriation defense was not applied, a transaction challenged by the FTC in federal court. In this deal, a drug maker bought the rights to NeoProfen, a new drug that treated a disease for which the drug maker already owned the only alternative drug therapy. My account emphasizes nonprice competition between the two drugs, in contrast to the court’s focus on price competition. I conclude that substantial price and non-price competition existed between the two drugs, making the availability of an efficiencies defense relevant to the outcome of the case. * Chief, Antitrust Bureau, Office of the New York State Attorney General (At the time of writing: Professor of Law, Columbia Law School). Participants in the 2010 ABA Antitrust and Innovation symposium and at the EUI’s 2010 Workshop on merger control provided helpful comments. I have consulted with the Federal Trade Commission on matters related to competition in the pharmaceutical industry, but the views expressed here are my own. 1 United States Department of Justice & Federal Trade Commission, Horizontal Merger Guidelines (2010). 2 Ibid. § 10.

44  Merger Control in European and Global Perspective Third, I consider how an increased appropriation defense might have applied to the NeoProfen acquisition. At its most expansive, the increased appropriation defense might support any acquisition that allows an innovator to increase the profits from its innovation. This interpretation is doubtful, I argue, in light of other provisions of the Guidelines. At a minimum, the increased appropriation defense permits a combination of complementary assets, the bringing together of which makes possible or enhances the marketing or distribution of an innovation. A difficult intermediate case arises when a merging party is poised to make socially beneficial expenditures—for example, investment in the development of an improved version of a product—that are profitable only if its ability to appropriate the benefits is preserved through merger.

I. The increased appropriation defense The revised Guidelines devote significant attention to innovation—typically though not exclusively conceived as the development and distribution of new, improved products and services—as a relevant consideration in horizontal merger analysis. The previous iteration of the Guidelines, issued in 1992 with modest revisions in 1997, said little about innovation. When agency leadership decided in 2009 that a renovation was overdue, the Agencies jointly held a series of workshops to consider possible changes. From the first workshop, it became clear that a new approach to innovation was likely: the very first question posed to panelists asked how the Guidelines might incorporate innovation more effectively.3 The revised Guidelines incorporate innovation in three significant respects. First, a merger’s effect on innovation is granted equal billing with price and output effects. For example, the Guidelines’ introductory section states, “A merger enhances market power if it is likely to encourage one or more firms to raise price, reduce output, diminish innovation, or otherwise harm customers as a result of diminished competitive constraints or incentives.”4 Second, the Guidelines consider several scenarios in which a merger is likely to dampen the incentive to innovate, for example, if the result of one merging firm’s successful research is a product that largely captures sales from the other merging firm.5 Third, innovation is more firmly established as one possible basis for an efficiencies defense.6 The Guidelines’ discussion of innovation-based efficiencies is not lengthy: it is just one paragraph of five sentences, at the end of the section about efficiencies. Such brevity is not unusual in the Guidelines, even for important 3 Transcript, Horizontal Merger Guidelines Review Project 50 (3 December 2009), http://www.ftc. gov/bc/workshops/hmg/transcripts/091203transcript.pdf (question of Carl Shapiro) (“So, let’s stipulate that innovation is really important. . . . There’s virtually nothing in the [then-current] guidelines on innovation effects. How might we do that [i.e., incorporate innovation] [...]?”). 4 Horizontal Merger Guidelines, cited above note 1, at § 1 (emphasis added). 5 Ibid. § 6.4. 6 Ibid. § 10.

C. Scott Hemphill 

45  

topics. Here, however, concision has an unfortunate consequence, because the import of the paragraph is not entirely clear. The first two sentences are straightforward: “When evaluating the effects of a merger on innovation, the Agencies consider the ability of the merged firm to conduct research or development more effectively. Such innovations may spur innovation but not affect short-term pricing.”7 This sensible statement is a more specific version of an approach discussed in general terms elsewhere in the efficiencies section.8 For example, a merger might permit two firms, each with research and development efforts that are too small to be effective, to combine and thereby achieve adequate scale. A related possibility is that the merger combines complementary research and development assets and approaches, as in the Genzyme/Novazyme transaction analyzed by the FTC in 2004.9 The third and fourth sentences set out the increased appropriation defense: “The Agencies also consider the ability of the merged firm to appropriate a greater fraction of the benefits resulting from its innovations. Licensing and intellectual property conditions may be important to this enquiry, as they affect the ability of a firm to appropriate the benefits of its innovation.”10 This statement is somewhat cryptic. The basic idea is that increased appropriation — the ability to extract the consumer surplus created by an innovation — is a good thing. Certain mergers increase appropriability and hence efficiency. That raises an obvious question: under what circumstances is increased appropriation efficient, and are all such situations recognized as an efficiencies defense? Unfortunately, the Guidelines do not spell this out. There is just one more sentence in the paragraph, but it is on a different subject, the treatment of reduced research and development expenditures.11 We are on our own. To consider possible interpretations, it is helpful to have a concrete set of facts to work with, to which we now turn.

II. The NeoProfen acquisition In 2006, Lundbeck, Inc. acquired the rights to NeoProfen, a drug used to treat premature infants with a rare heart condition called patent ductus arteriosus (PDA).12 At the time of the acquisition, Lundbeck already had a PDA drug on the market called Indocin IV. NeoProfen was not yet approved for use by the Food Ibid. Ibid. (“[M]erger-generated efficiencies may enhance competition by permitting two ineffective competitors to form a more effective competitor, e.g., by combining complementary assets.”). 9 See Statement of Chairman Timothy J. Muris in the Matter of Genzyme Corporation/Novazyme Pharmaceuticals, Inc. 17 (2004), http://www.ftc.gov/os/2004/01/murisgenzymestmt.pdf. 10 Horizontal Merger Guidelines, cited above note 1, at § 10. 11 The final sentence states: “Research and development cost savings may be substantial and yet not be cognizable efficiencies because they are difficult to verify or result from anticompetitive reductions in innovative activities.” Ibid. 12 Actually, Lundbeck is the successor in interest to Ovation Pharmaceuticals, which made the initial purchase, a detail ignored in the main text. 7 8

46  Merger Control in European and Global Perspective and Drug Administration (FDA), but was approved later that year. Indocin IV and NeoProfen were the only two drugs available to treat PDA; the major alternative treatment is surgery. In 2008, the FDA approved a generic version of Indocin IV. The FTC, joined by the state of Minnesota, challenged the acquisition after its consummation as a violation of the FTC Act and section 7 of the Clayton Act. State law and Section 2 of the Sherman Act were invoked as well. After a trial, the District Court rejected the challenge,13 and that judgment was affirmed on appeal.14 The District Court concluded that there was little price competition between the two drugs, and hence that the FTC had failed to prove that Indocin IV and NeoProfen were in the same market. The court relied heavily on the testimony of neonatologists, who testified as to their insensitivity to price in deciding which drug to prescribe. The court apparently did not consider direct evidence of an actual price rise resulting from the consummated merger as an alternative to the market definition approach. The revised Guidelines strongly promote a variety of alternatives to traditional market definition in identifying anticompetitive effects of a merger.15 The omission is particularly striking because the court found that Lundbeck increased the price of Indocin IV by 1300%, and delayed the increase until shortly after the NeoProfen acquisition closed.16 The price rise and its timing suggest an anticompetitive effect resulting from the merger. The District Court, however, concluded that Lundbeck would have increased its price by the same amount even if it had not purchased the rights to NeoProfen.17 Even under a traditional market definition analysis, the court’s result is surprising. The District Court’s own findings of fact suggest significant price competition between the two drugs. The court’s finding that Lundbeck delayed its Indocin IV price increase until after the NeoProfen acquisition closed suggests that Lundbeck perceived significant substitution between the two products. The court also concluded that an independent owner of NeoProfen, in setting the price of NeoProfen, would not have disregarded Indocin IV’s price.18 In addition, the court found that, when Lundbeck launched NeoProfen, it executed a plan to offer it at a discount compared to Indocin IV,19 consistent with academic studies that show that competing, later marketed drugs are often introduced at a discount relative to the pioneer drug.20 FTC v. Lundbeck, No. 08-6379, 2010 WL 3810015 (D. Minn. 2010). FTC v. Lundbeck, Nos. 10-3458, 10-3459, 2011 WL 3629347 (8th Cir. 2011). See, e.g., Horizontal Merger Guidelines, cited above note 1, at §§ 2.1.1 (“Actual Effects Observed in Consummated Mergers”) and 2.1.2 (“Direct Comparisons Based on Experience”). The FTC, which filed the Lundbeck case prior to the issuance of the revised Guidelines, focused its efforts on a market definition approach. For an analysis, see Richard Parker, Michael Antalics and Bilal Sayyed, “Shrinking from the ‘Third Rail’: Avoiding Direct Effects Analysis in Lundbeck”, 25 Antitrust 14 (Spring 2011), at 14 (noting that the complaint alleged a market, and that the post-trial briefing emphasized the sufficiency of the market definition approach). 16 FTC v. Lundbeck, cited above note 13, at ¶ 58. 17 Ibid. 18 Ibid. ¶ 63. 19 Ibid. ¶ 82. 20 See, e.g., Z. John Lu and William Comanor, “Strategic Pricing of New Pharmaceuticals”, 80 Review of Economics and Statistics 108 (1998). 13 14 15

C. Scott Hemphill 

47  

The District Court apparently thought this evidence was irrelevant to the competition question because doctors were the decision makers to such a degree that price was an insignificant factor in the purchase decision. But even if this were so, there was non-price competition between the drugs that was eliminated as a consequence of the acquisition.21 The court’s findings themselves demonstrate significant non-price competition between the drugs. As the court noted, doctors are split in their usage of these two drugs.22 Doctors vary in their perception of key non-price attributes of safety and effectiveness. Some think more highly of Indocin IV, others prefer NeoProfen.23 Those physician perceptions are affected by the nature and quantity of information that drug makers provide, particularly through pharmaceutical sales representatives. If the drugs were independently owned, each company would aggressively emphasize the advantages of its drug. Lundbeck, as the owner of Indocin IV but not NeoProfen, would have continued to emphasize Indocin IV’s advantages over NeoProfen, such as a longer safety record. The competition between two independent drugs would have proceeded normally, and the success of each drug would depend on its ability to meet the needs of patients. Instead, upon purchasing the rights to NeoProfen, Lundbeck changed its strategy. Lundbeck gave its sales representatives a financial incentive to sell NeoProfen, but not Indocin IV.24 The representatives were told to emphasize the benefits of NeoProfen, but only the weaknesses of Indocin IV.25 This new strategy, prompted by common ownership of the two drugs, deprived doctors of competing, independent information about which drug is best for their patients. This suppression of non-price competition was designed to change minds in one direction, from Indocin IV to NeoProfen. The apparent goal was to shift customers to NeoProfen before generic substitutes for Indocin IV were approved, at which point Indocin IV profits would decline. In short, the District Court failed to recognize the non-price dimension of competition. The failure is apparent in this statement: “Were NeoProfen and Indocin IV in the same product market, Lundbeck’s attempt to persuade neonatologists to switch from Indocin IV to NeoProfen would not make sense.”26 In fact, it is precisely because the two drugs are in the same product market, and compete on a non-price basis, that the attempts at persuasion do make sense. And because of Lundbeck’s strategy to reduce the effect of anticipated generic competition in Indocin IV, these efforts at persuasion were aimed in only direction. The NeoProfen acquisition thus ought to have been recognized as a substantial threat to both price and non-price competition. Had the District Court reached that 21 For an analysis emphasizing non-price competition, see Brief of Amicus Curiae American Antitrust Institute at 7–16, FTC v. Lundbeck, Nos. 10-3458, 10-3459 (8th Cir. 2011), http://www. antitrustinstitute.org/sites/default/files/AAI%20Amicus%20Brief.pdf. 22 FTC v. Lundbeck, cited above note 13, at ¶ 94. 23 Ibid. ¶¶ 95–108. 24 Ibid. ¶ 81. 25 Ibid. (“Shortly after acquiring the rights to NeoProfen, Lundbeck instructed its sales representatives to focus on Indocin IV’s weaknesses relative to NeoProfen’s anticipated benefits.”). 26 Ibid. ¶ 116.

48  Merger Control in European and Global Perspective conclusion, it might then have proceeded to consider whether any procompetitive efficiencies could nevertheless justify the transaction. An increased appropriation defense would have been one argument to examine. The next section considers what such a defense might look like.

III. The scope of the increased appropriation defense The most aggressive version of the increased appropriation defense is also the simplest. Lundbeck, as owner of Indocin IV, faces more intense competition with NeoProfen as a rival, and hence lower profits on Indocin IV. Blunting the competitive force of its rival, whether by removing NeoProfen from the market or altering its pricing strategy to maximize joint profits from the two products, on the other hand, increases appropriability. The revised Guidelines might be read as recognizing a defense whenever the acquisition increases appropriability. The defense might be particularly appropriate where “intellectual property conditions” fail to confer adequate appropriability. Indocin IV, which lacks patent protection, might appear to be such a product. A similar point could be made from NeoProfen’s perspective — that reducing competitive pressure from Indocin IV increases appropriability for NeoProfen. The “intellectual property conditions” argument would be weaker, since NeoProfen enjoys both patent protection and “orphan drug” protection, which provides some insulation from competition for seven years after FDA approval.27 For either drug, the justification that underpins the broadest version of the increased appropriation defense is the same. The prospect of increased profits from higher prices in turn supports an investment in innovation in the first instance. This justification for entering into a merger, rooted in concern for dynamic efficiency, has been noted by commentators.28 It is doubtful that enforcement agencies would fully embrace this strong version of the defense, which would be available in any horizontal merger of innovators. On this theory, price increases are desirable, and mergers that help secure price increases — for example, ridding the market of a maverick, even a merger to monopoly — are actually favored. However, such results are directly contrary to a central goal of the Guidelines, which is to prevent mergers that yield higher prices.29 It is unlikely that 21 U.S.C. § 360ee. See, e.g., Michael Katz and Howard Shelanski, “Mergers and Innovation”, 74 Antitrust Law Journal 1, 19 (2007) (“Suppliers with many product-market rivals may have less ability to appropriate the returns from innovation that make the investment in innovation worthwhile, [...] because their innovations are readily copied or invented around by rivals . . . .”); cf. Transcript, cited above note 3, at 61 (comments of Tim Muris) (noting “there are lots of models that say it’s better to have fewer firms because you can capture the benefits to innovation,” and endorsing Katz and Shelanski’s view that nevertheless it does not follow that all mergers must be approved on innovation grounds). 29 Horizontal Merger Guidelines, cited above note 1, at § 1 (preventing price increases is a central concern). See also ibid. §§ 2.1.5, and 7.1 (merger less likely to be permitted where it removes a maverick from the market). 27 28

C. Scott Hemphill 

49  

the increased appropriation defense is properly to be understood as creating such a stark conflict within the Guidelines without any clear statement to that effect. Statements within the efficiencies section of the Guidelines further undercut this interpretation. For example, in that section it is explained that efficiencies are relatively less likely to be recognized where they relate to research and development: although “potentially substantial”, such efficiencies are “generally less susceptible to verification and may be the result of anticompetitive output reductions.”30 Here, the claimed dynamic efficiency is directly the result of just such an output reduction. Furthermore, even if the tradeoff between dynamic and static efficiency sometimes supports a valid antitrust defense, there is a significant question about how to incorporate it in a particular merger analysis in which the relevant innovation has already been elicited. In other words, even if it is the case that limits on competition are necessary to elicit certain innovations, that cannot easily be said of Indocin IV or NeoProfen. Instead, the benefits from the proposed rule would accrue in other, future transactions, in which the parties could anticipate beforehand that their profits would be greater thanks to the availability of the defense. Neither the Guidelines nor the Clayton Act provides a clear basis for treating those future benefits for other innovators as a point in favor of a particular transaction. Let us instead consider a more limited interpretation of the increased appropriation defense. Suppose that Lundbeck is highly suited to the commercialization of PDA drugs, thanks to its long experience with Indocin IV. It has excellent access to neonatal physicians, enjoys their confidence, and possesses a strong brand reputation. An independent owner of NeoProfen is likely to lack these advantages, particularly during the period prior to drug approval. That will impede its relative ability to market the drug. In the limit, the independent owner might be unable to market its drug at all. (This assumption does not fit the actual NeoProfen facts; Abbott, the former owner, is a capable drug marketer.) On these facts, Lundbeck’s ownership of NeoProfen would improve appropriability on account of the complementarity between the NeoProfen technology and Lundbeck’s marketing and distribution assets. Exploiting this complementarity is socially desirable and, provided that a similar gain could not be accomplished without a merger, it is a cognizable benefit under the Guidelines. It is worth noting, however, that not every exploitation of a marketing advantage is unambiguously procompetitive. Suppose, for example, that Lundbeck had superior knowledge of which arguments would sway a particular doctor to choose one drug over another. If the bases for physician choices were not fully aligned with social welfare, then exploitation of the information might not be procompetitive. The case of complementarity is key to the increased appropriation defense. It fits a reference, elsewhere in the Guidelines, to a type of “merger [that] is likely to enable innovation that would not otherwise take place, by bringing 30

Ibid. § 10.

50  Merger Control in European and Global Perspective together complementary capabilities that cannot be otherwise combined [...]”31 The example of combining technology with distribution also helps explain the reference to “licensing and intellectual property conditions” quoted above. The point appears to be that a lack of intellectual property protection for the innovation can make licensing difficult. That statement is less likely to apply where, as with NeoProfen, the relevant asset is a patent-protected drug. A clearer case in which consolidation might be necessary to fully exploit an innovation is a cost-reducing or quality-improving innovation that is not subject to effective intellectual property protection and not easily made the subject of an enforceable contract. One possible example is PepsiCo’s investment in California Pizza Kitchen in the early 1990s. The investment was made so that PepsiCo could absorb CPK’s expertise in managing full-service restaurants and apply that expertise to PepsiCo’s other operations.32 Achieving the knowledge transfer in exchange for compensation might have been difficult to arrange through contract. In some cases, a transaction that achieves this efficiency might also create an opportunity for a price increase. For example, consider a case in which the newly acquired product more effectively reaches a new set of customers who would otherwise not be served, but at the same time common ownership raises prices for other customers. The general efficiencies analysis in the Guidelines suggests that enforcement agencies may insist that the combination not result in any price increase compared to independent ownership.33 If that statement is controlling, and thus if the traditional Guidelines focus on prices continues to apply, then the new statement about increased appropriation makes little real difference in practice. The conflict also appears in a further difficult intermediate case. Suppose that, as seems likely, Lundbeck’s greatest worry was not competition from NeoProfen, but rather that Indocin IV would soon be subjected to generic competition. NeoProfen would be attractive as a patent-protected haven to which customers could be shifted prior to generic entry. Such shifts are a frequent strategy for brand-name firms facing generic competition.34 The FTC considered a similar transaction in 1999, when Lilly proposed to buy the rights to a successor to its then-blockbuster antidepressant Prozac, which was nearing the end of market exclusivity.35 Ibid. § 6.4. Russ Britt, “California Pizza Kitchen Sells Slice—PepsiCo Takes 50% Stake in L.A. Chain”, L.A. Daily News (21 May 1992) (“This is a way for us to learn from proven operators about something we know nothing about. We think the best way to do that is to work alongside them.”). I thank Doug Ginsburg for suggesting this example. 33 Horizontal Merger Guidelines, cited above note 1, at § 10 (“To make the requisite determination, the Agencies consider whether cognizable efficiencies likely would be sufficient to reverse the merger’s potential to harm customers in the relevant market, e.g., by preventing price increases in that market.”). 34 For discussion, see C. Scott Hemphill and Mark Lemley, “Earning Exclusivity: Generic Drug Incentives and the Hatch-Waxman Act”, 77 Antitrust Law Journal 947 (2011). 35 Sheila Anthony, “Riddles and Lessons from the Prescription Drug Wars: Antitrust Implications of Certain Types of Agreements Involving Intellectual Property”, speech, San Francisco, 1 June 2000, http://www.ftc.gov/speeches/anthony/sfip000601.shtm. The FTC closed its investigation without challenging the transaction. 31 32

C. Scott Hemphill 

51  

To make this example interesting, let us posit in addition that Lundbeck would make socially beneficial expenditures in PDA therapy, but only if its PDA franchise is preserved. For example, the drug maker might find it profitable, as well as beneficial to patients, to provide free samples as a way to build physician interest in the franchise.36 This strategy would not be profitable, and hence would not be pursued, once generic entry occurs. Or — to more strongly implicate the incentive to innovate — suppose Lundbeck has a significant incentive to make further improvements to PDA drugs, but again, only so long as its franchise is protected from generic entry. Here, in contrast to the broadest version of the defense, improved appropriability would increase R&D incentives as to the product concerned by the merger. But the other arguments against the strong form of increased appropriability remain intact. That is, increased appropriability might plausibly promote innovation, but as before, in a way that runs counter to other statements in the Guidelines. While preserved incentives to make product improvements are consistent with the essence of the increased appropriation defense, at least on these posited facts they sharply conflict with the Guidelines’ continuing primary focus on price. * * * In the end, then, we are left with a puzzle. At its narrowest, the increased appropriation defense applies only in cases where the transaction will produce no price increase. That would permit certain mergers where a new marketing technique or cost-reducing innovation finds wider application through merger, or where a patented product reaches the market that otherwise would never reach customers. In such cases, however, we hardly need the new defense. Without the benefit of this statement about increased appropriation, we would likely conclude that such transactions are acceptable because they pose no traditional threat to competition in any event. On the other hand, in those situations where the defense might be meaningful — where there might be a price increase, but the associated increase in appropriation accompanies or even makes possible a welfare improvement — it seems to run afoul of the Guidelines’ continuing focus on price competition. How the Agencies will balance these contending effects remains to be seen.

36 Cf. Cornerstone Research, “Court Acknowledges Novel Sampling Defense in Pharmaceutical Case” (26 February 2008), http://www.cornerstone.com/novel_sampling.

Barry Hawk*

A Tale of Two Cities: Washington and Brussels Face the Courts

I. Introduction In August of 2010 the Federal Trade Commission and the Department of Justice released the 2010 Horizontal Merger Guidelines (revised Merger Guidelines, or “RMG”).1 The final version of the RMG followed an extensive comment period.2 This release marks the first major changes to the Guidelines in over 18 years, replacing the 1992 Guidelines. In announcing the release of the RMG, Christine Varney, Assistant Attorney General in charge of the DOJ’s Antitrust Division, explained that the revisions are meant to “provide more clarity and transparency” as to how the federal antitrust agencies evaluate the likely competitive impact of mergers and are intended to “provide businesses with an even greater understanding of how [the Agencies] review transactions.”3 For example, the RMG are said to update the treatment of unilateral effects to reflect the substantial changes in economic learning and agency practice since 1992.4 There is a fundamental tension in many areas of competition law, including merger control, between sophisticated fact-specific economic analysis and clear, predictable legal rules. The decades-old debate about legal structural presumptions (and safe harbors) reflects this tension.5 Advances in economic theory in the past 25 years have aggravated that tension. It is fair to conclude that the philosophy underlying the RMG falls on the side of fact-specific merger analysis: “These Guidelines should be read with the awareness that merger analysis does not consist of uniform application of a single methodology. Rather, it is a factspecific process through which the Agencies, guided by their extensive experience, apply a range of analytical tools to the reasonably available and reliable evidence to evaluate competitive concerns in a limited period of time.”6 * Fordham Law School. 1 A copy of the 2010 Horizontal Merger Guidelines is available on the FTC’s website: http://www. ftc.gov/os/2010/08/100819hmg.pdf. 2 See public comments at http://www.ftc.gov/os/comments/hmgrevisedguides/index.shtm. 3 See Press Release of 19 August 2010, http://www.ftc.gov/opa/2010/08/hmg.shtm. 4 See Carl Shapiro, “The 2010 Horizontal Merger Guidelines: From Hedgehog to Fox in Forty Years”, 77 Antitrust Law Journal 49 (2010). 5 See, e.g., Derek Bok, “Section 7 of the Clayton Act and the Merging of Law and Economics”, 74 Harvard Law Review 226 (1960). 6 See section 1 (Overview), para. 4.

54  Merger Control in European and Global Perspective This can be described as the eclectic approach to merger analysis. For example, Carl Shapiro (then-Chief Economist at the DOJ) writes: “merger enforcement in recent years has become increasingly eclectic, reflecting the enormous diversity of industries in which the Agencies review mergers and the improved economic toolkit available. The Agencies and the courts look at a wide variety of evidence and use a wide variety of methods to determine whether mergers may substantially lessen competition. Based on decades of experience examining mergers, the Agencies recognize that each industry has unique features and each merger presents unique circumstances.”7

II. Three issues raised by critics of the RMG The RMG have been well received as providing a needed updating of the 1992 Guidelines, notably because of the revised Guidelines’ better alignment with present agency practice, among other things. But the generally favorable reaction has not been free of criticism: the RMG have raised concerns with regard, in particular, to the way they approach unilateral effects analysis. In this chapter I briefly discuss three criticisms: (1) the claim that RMGs’ eclecticism is litigationdriven and leaves too much discretion to the Agencies (e.g., because there are no presumptions or limiting principles respecting the UPP theory and unilateral effects); (2) the ambiguous role of market definition; and (3) the asserted lack of reliable data to undertake the UPP analysis.

A. Discretion and legal uncertainty First, perhaps the major criticism voiced by (mostly) legal commentators is that the RMG increase the mystery of U.S. merger control despite the declared intention to increase transparency. According to this line of criticism, the RMG provide a list of advocacy tools from which the Agencies can select – without sufficient constraints – in order to give themselves greater flexibility, discretion and maneuvering room when litigation before the courts becomes necessary.8 The Shapiro, cited above note 4, at pp. 51–52. See, e.g., Kevin Arquit, “Unilateral Effects Analysis in the 2010 Horizontal Merger Guidelines: Beginning of the End of the Age of Restraint?”, in Barry Hawk, ed., International Antitrust Law and Policy: Fordham Competition Law 2010, Juris Publishing, 2011, chapter 6: “In a variety of ways, the changes to the unilateral effects section confer significant discretion to the agencies to investigate and potentially block mergers under novel economic theories, but in material respects do so with insufficient guidance to practitioners, courts, or even the agencies. This added discretion is consistent with the desire for greater flexibility at the agencies, which is likely the result of a number of high profile losses where the courts effectively relied upon the previous versions of the Guidelines to conclude that the agencies were overreaching.” See also Michael Sohn, “Pricing of Differentiated Products in the 2010 Horizontal Merger Guidelines: Some Unanswered Questions”, in Hawk, ed., Fordham Competition Law 2010, cited above, chapter 8. 7 8

Barry Hawk 

55  

RMG have been criticized as listing various theories to challenge mergers without (sufficiently) delimiting the application of those theories. One example of the critics’ charge of unfettered discretion (and uncertainty) is the elimination of any concentration threshold or safe harbor for application of the Upward Pricing Pressure (UPP) theory, unlike the 1992 Guidelines, which referenced (or at least were interpreted by lawyers and firms as indicating) a 35% threshold presumption or 35% safe harbor. Putting aside the historical question of whether the 1992 Guidelines did or did not articulate a 35% safe harbor, the RMG reject such a threshold on the reasoning that “a merger combining two products that are close substitutes can lead to substantial unilateral price increases for those products even if their combined market share is less than thirty-five percent […], the presumption could only properly be invoked if market shares are a reasonable proxy for diversion ratios … [and] modern economic theory relates unilateral price effects with differentiated products more directly to diversion ratios and margins than to the combined market share of the merging firms.”9

Another example of asserted uncertainty concerns the UPP theory. Critics contend that, unless some benchmark is used, the U.S. Agencies will have (unfettered) discretion where a merger results in UPP, and that this can be the case in all horizontal merger scenarios because the RMG fail to calibrate or quantify the unacceptable UPP.10 The Agencies acknowledge that UPP will not be used to challenge all horizontal mergers, but critics argue that the RMG fail to say when a merger will be challenged, and there is no (public) limitation or benchmark, such as a 10% efficiency offset.11 The RMG do not state whether a “standard deduction” Shapiro, cited above note 4, at p. 70. This concern was expressed about the proposed RMG. For example, Professor Willig said, in response to the proposed draft, that “[t]he draft does not articulate at all how the agencies calibrate this quantified measure. What constitutes a large enough value of diverted sales to raise substantial concerns? What is a small enough value of diverted sales to indicate that concerns about these competitive effects are unwarranted? To what should a measured value of diverted sales be compared?” Willig urged that the Guidelines should “provide guidance on how to use this new central tool for assessing unilateral competitive effects.” The final RMG remain less than pellucid on the point: “If the value of diverted sales is proportionately small, significant unilateral price effects are unlikely” and “For this purpose, the value of diverted sales is measured in proportion to the lost revenues attributable to the reduction in unit sales resulting from the price increase. Those lost revenues equal the reduction in number of units sold of that product multiplied by that product’s price.” 11 For example, the chief economists at the DOJ and FTC have suggested that the predicted price effects (UPP) might be offset by merger specific efficiencies or by repositioning of their products by rivals. See Joseph Farrell and Carl Shapiro “Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition,” 10(1) B.E. Journal of Theoretical Economics, Article 9 (2010). SSRN version at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1313782. The American Bar Association voiced a similar concern about the proposed Guidelines in the following terms: “[T]he Proposed Guidelines could be read to suggest that the Agencies’ policy is to presume unilateral effects from a considerably wider range of mergers than currently is the case, based on an evaluation of largely static factors (pre-merger data suggesting diversion ratios, and pre-merger margins), and then to treat repositioning arguments with the kind of skepticism that typically is given to de novo entry arguments. It is widely understood by practitioners, and acknowledged by some in the Agencies, that entry arguments are rarely successful in defending a merger that otherwise has been deemed potentially anticompetitive. […] [R]epositioning will often be much easier and more likely than new entry. […] Requiring the parties to demonstrate that such repositioning meets the requirements of de novo entry under section 10 appears unjustified.” American Bar Association (6/4/2010) #54805000026, page 16. http://www.ftc.gov/os/comments/hmgrevisedguides/548050-00026.pdf. 9

10

56  Merger Control in European and Global Perspective will be used, or what any such “standard deduction” might be.12 In response, Carl Shapiro contends that “the claim that unilateral effects models ‘always predict a price increase’ and thus are unsuitable for merger enforcement […] is incorrect. […] [T]he criticism ignores efficiencies, repositioning and entry.”13 The adequacy of this response depends on whether predictable and robust analyses of efficiencies, entry and repositioning can be confidently made, which has itself been subject to debate.14

B. The role of market definition There has been considerable debate about the role of market definition under the RMG. The RMG explain that “[d]iagnosing unilateral price effects […] need not rely on market definition” or concentration.15 A number of prominent economists and lawyers have decried the reduced reliance on market definition (and market share/concentration data) and the RMGs’ emphasis on direct assessment of unilateral effects and UPP.16 Some practitioners have voiced the concern that the de-emphasis on market definition is a tactical litigation response to the difficulty of persuading courts to accept narrowly defined markets.17 Dennis Carlton has 12 The RMGs’ general discussion of efficiencies provides little further insight, and they envision efficiencies defined with reference to a relevant market — the very sort of analysis that the revised Guidelines deem unnecessary when assessing UPP: “The Agencies will not challenge a merger if cognizable efficiencies are of a character and magnitude such that the merger is not likely to be anticompetitive in any relevant market. To make the requisite determination, the Agencies consider whether cognizable efficiencies likely would be sufficient to reverse the merger’s potential to harm customers in the relevant market, e.g., by preventing price increases in that market.” 13 Shapiro, cited above note 4, at p. 80. Shapiro also states, at pp. 76–77, that: “[t]he value of diverted sales, taken alone, does not purport to quantify the magnitude of any postmerger price increase. […] The value of diverted sales is an excellent simple measure for diagnosing or scoring unilateral price effects, but it cannot capture the full richness of competition in real-world industries […] all of the quantitative methods … must be used in conjunction with the broader set of qualitative evidence that the Agencies assemble during a merger investigation. […] Further information about demand is needed, and additional analysis is required, to translate these incentives into predictions of post-merger price increases. To accomplish this, DOJ economists and economists working for merging parties often undertake merger simulation exercises.” (footnote omitted; emphasis in original) 14 Some economists also criticize the RMG’s adoption of the UPP theory on several grounds, including theoretical weaknesses and practical limitations. See, e.g., Dennis Carlton, “Revising the Horizontal Merger Guidelines”, 6 (3) Journal of Competition Law and Economics 619 (2010). 15 See Merger Guidelines 2010 (for link, see above note 1), at p. 21. 16 See, e.g.. Carlton, cited above note 14; Sohn, cited above note 8, and commentators cited therein. See also Dennis Carlton and Mark Israel, “Will the New Guidelines Clarify or Obscure Antitrust Policy?”, 10 Antitrust Source (October 2010). 17 As Sohn (cited above note 8) points out: “In recent years, the government’s biggest court losses in horizontal merger cases have turned on market definition issues. See Federal Trade Commission v. Arch Coal, Inc., 329 F. Supp. 2d 109 (D.D.C. 2004), appeal dismissed, No. 04-5291, 2004 U.S. App. LEXIS 19405 (D.C. Cir. Sept. 15, 2004); United States v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004); Federal Trade Commission v. Whole Foods Market, Inc., 502 F. Supp. 2d 1 (D.D.C. 2007), reversed and remanded, 533 F.3d 869 (D.C. Cir. 2008).” One court in 2010 rejected the contention that market definition in unilateral price effects cases can be ignored. The Court rejected the testimony of the plaintiff’s economic expert who proposed that in lieu of the traditional market definition approach, it would be appropriate to rely upon a unilateral effects analysis based on the Upward Pricing Pressure model alluded to in the 2010 Guidelines: “The

Barry Hawk 

57  

suggested that the Guidelines should clarify that narrow market definitions can be economically appropriate and reasonable, especially in evaluating the unilateral effects of mergers in industries with differentiated products.18 Agency officials have responded that the claim of abandoned market definitions is exaggerated. For example, some U.S. officials concede that market definition will always be alleged and proved in litigation, although market definition is not always the starting point in the Agencies’ daily merger analysis.19 As Carl Shapiro has put it: “[O]ur Networks and Technology Section […] said [with respect to digital sector mergers], ‘we can’t start with market definition in most of our merger cases. It doesn’t work. […] We try to figure out how the industry is evolving, where these companies are going, are they knocking up against each other, are they complementary in some ways — just figuring out effects.’ Nobody there was saying ‘Throw market definition overboard,’ but it wasn’t the framing question out of the box.”

In other words, merger analysis does not start with market definition but rather with the formulation of candidate theories of competitive harm.20 Another agency response is that market definition will always be required, but that market shares and concentration will be deemphasized compared with other evidentiary factors. In summary, it is fair to note that the exact conditions in which market definition will and will not be used may be less clear than ideal, and the jury is out on whether (or to what extent) merger analysis can avoid market definition and whether courts would accept this shift in the Agencies’ approach. It is true that the courts have relied heavily on prior versions of merger guidelines to adopt new analytic tools. Examples include the courts’ treatment of HHI before and after the 1982 Merger Guidelines and the current entry test before and after the 1992 Horizontal Merger Guidelines. Courts gradually became more sympathetic to these ideas, which gradually gained widespread acceptance. History may repeat itself with the 2010 (re)iteration of the Guidelines but it is far less clear whether courts will follow the RMG’s deemphasis on market definition, particularly given the statutory language requiring a lessening of competition in a “line of commerce.”21 It is reasonable to expect courts to be strongly indisposed to eliminating market definition from merger analysis. Court notes that its research has not revealed a single decision of a federal court adopting this UPP test. In light of the case law’s clear requirement that a Plaintiff allege a particular product market in which competition will be impaired, this absence of authority is hardly surprising.” City of New York v. Group Health Inc., et al., 06-cv-13122 (S.D.N.Y. 2010). 18 See Dennis Carlton, “Comment on Proposed Horizontal Merger Guidelines” (4 June 2010), http://www.ftc.gov/os/comments/hmgrevisedguides/548050-00034.pdf, at p. 7. See also Carlton, cited above note 14. 19 See, e.g.. the remarks of Carl Shapiro, in Barry Hawk, ed., International Antitrust Law and Policy: Fordham Competition Law 2010, Juris Publishing, 2011, chapter 18. 20 Shapiro, cited above note 4, writes (at page 708) that: “investigations usually start with the formulation of candidate theories of harm to competition and the exploration of evidence to support or reject those theories. In most cases, especially where market boundaries are unclear, DOJ staff will analyze evidence of possible harm before it has determined the scope of the relevant market. Indeed, the same piece of evidence may be relevant to competitive effects and to market definition,” 21 See, e.g., Leah Brannon and Kathleen Bradish, “The Revised Horizontal Merger Guidelines: Can the Courts be Persuaded?”, 10 Antitrust Source (October 2010).

58  Merger Control in European and Global Perspective In sum, my guess is that market definition can be expected to remain important, but reliance on market shares and concentration data may decline, with the Agencies defining markets more toward the end than the beginning of the inquiry. Moreover, in litigation the U.S. Agencies may be forced by the courts to define (and defend) narrow markets in unilateral effects cases. The U.S. RMG and the European Commission’s merger analysis are largely similar, and convergences outweigh the differences in importance. But the role of market definition appears more solidly rooted under the EU Merger Regulation than the less certain status under the RMG. The RMG drop the presumption contained in the 1992 Guidelines that a merger is likely to result in unilateral effects where the parties’ combined market share is 35 percent or more. It is fair to say that the RMG deemphasize market concentration data and information – certainly as compared with the 1992 Guidelines. The Commission appears more reluctant than the U.S. Agencies to declare or suggest that markets need not be defined (at least in certain cases). However, this is not entirely clear. In cases where market definition is most likely to be ignored under the RMG, the merger analyses in the U.S. and the EU are largely the same in the sense that the U.S. Agencies and the European Commission consider the same factors, such as, for example, diversion ratios. Concentration-based presumptions (that rest on market definition) do appear more important in the Commission’s practice compared to the RMG, although concentration presumptions have been less emphasized in Europe since the adoption of the SIEC test and the publication of the EU Horizontal Merger Guidelines in 2004.

C. Reliability of data Many commentators question whether sufficient reliable data (notably, profit margin information and diversion ratio information) can be collected in many cases so as to support a predictable and robust application of UPP theory, particularly outside the retail merger area where scanner data are used.22 The Agencies are confident that the data and information are available23 and that optimism is shared by some observers from outside the U.S. such as Alison Oldale, then-Chief Economist at the UK Competition Commission.24 See, e.g. the remarks of Michael Sohn, cited above note 8. Shapiro writes: “the documents of merging parties can be informative regarding diversion ratios and margins. Firms often are keenly interested in identifying the rivals to whom they lose business, or from whom they can gain business. Businesses are far more likely to address these questions in their day-to-day operations than they are to address questions relating to the hypothetical monopolist test. Margins are also central to business decisions. Margins are an essential element of pricing decisions, and the return on a marketing campaign that attracts new customers depends directly on the price/cost margins that will be earned on those customers.” Shapiro, cited above note 4, at p. 80. 24 See Remarks of Alison Oldale in Barry Hawk, ed., International Antitrust Law and Policy: Fordham Competition Law 2010, Juris Publishing, 2011, chapter 9. 22 23

Barry Hawk 

59  

III. Conclusion In order to provoke some controversy, let me accept, arguendo, the critics’ view that the U.S. RMG are (in some measure) agency litigation strategy responses to judicial review (i.e., to losses before federal courts). Let me then suggest the following proposition for further discussion: the U.S. RMG, viewed as an agency response to judicial review, contrast poorly with the European Commission’s response to judicial review several years ago. The Court of First Instance dealt the Commission three important losses in the latter’s implementation of the pre-2004 EC Merger Regulation (i.e., Regulation 4064/89). The Commission reacted to these defeats by significantly changing (or “reforming”) its procedures, substantive test and institutional mechanisms in order to address the criticisms of the court. This reaction arguably contrasts with the U.S. Agencies’ reaction to a number of litigation losses involving mergers in the U.S. courts, assuming that the RMG reflect an attempt to strengthen the Agencies’ litigation position in future cases. What will be the effect of the RMG in the courts? Any answer to that question will depend partially on the extent to which the Agencies rely on the RMG before the courts when challenging mergers, particularly where opposing parties cite contrary case law in support of their own positions. It is not inconceivable that the Agencies will then resort to traditional market share/market concentration analysis as a litigation strategy. Unlike the Commission, the U.S. Agencies are unconstrained by a prior reasoned decision and thus there is more room in the U.S. for the unpalatable divorce of regulatory standards and judicial standards, which could hardly have been the intent of the legislators and Section 7 of the Clayton Act. I once described the pre-2004 EU (non-merger) notification system as “Catholic”, with its focus on overbreadth of unlawful restrictive agreements (“sins”), “confession” (notification) and “forgiveness” (exemption) by the Commission.25 This system fostered a priesthood of Brussels fonctionnaires and lawyers who knew the intricacies and nuances of actual enforcement. Regulation 1/2003 and the elimination of the notification system constituted a “Protestant Reformation”, with a new focus on self-analysis by firms and their counsel. How might the RMG be seen in a theological world? Despite the effort to increase transparency, the embrace of eclectic fact-specific economic analysis might strike some as a return to the Delphic oracle. The RMG appear to increase the importance of the priesthood of agency officials, regulatory lawyers and economic consultants who, with experience in (largely unpublished) cases, will learn when and where the U.S. Agencies will draw the lines when selecting from the list of theories in the RMG. Given that the U.S. Agencies do not publish the reasons for clearing a merger, at best only the lawyers and economists who have 25 Barry Hawk, “EU ‘modernisation’: a latter-day reformation”, Global Competition Review (August/September 1999), p. 12.

60  Merger Control in European and Global Perspective substantial experience with the (Delphic) U.S. merger control process will have a well formed idea of where the Agencies are drawing the (inevitable) lines. The RMG do reduce mystery by making transparent present agency practice. But critics might argue that what is being made more transparent is an eclectic menu of analytical tools without sufficient limiting principles. Delphi replaces Rome and the Reformation, and all the priestesses have Ph.Ds in economics. As noted at the outset, there is a fundamental tension between sophisticated fact-specific economic analysis and legal structural presumptions (and safe harbors). Generic guidelines that describe an agency’s practice of sophisticated fact-specific economic analysis are useful for their transparency. The difficult task is formulation (and transparency) of legal presumptions that do not unduly sacrifice the sophisticated economics. Of course, this tension between factspecific economic analysis and predictable, certain legal rules is not unique to merger analysis. It underlies other antitrust issues as well, such as the debates about rules of reason versus per se rules for examining restrictive agreements, and about form-based versus effects-based rules for dominant firm conduct. Publications of clearance decisions (or detailed summaries) would help reduce some of the legal uncertainty of fact-specific merger control, but this might well require a reduction in the (large) number of merger notifications under the HartScott-Rodino Act. However, there are too many (government and private) interests benefiting from the present H-S-R notification system to expect any (significant) reduction of the number of U.S. notifications. (Whether any public interest is served by hundreds to thousands of annual notifications is a different question.) The U.S. system of notifications, with only rarely published agency decisions, compares poorly with the system established under the EU Merger Regulation, and with many merger control systems outside the U.S. Of course, this places even more importance on the RMG to provide clear rules, including thresholds and safe harbors. While U.S. substantive merger analysis is a useful model for merger control around the world, the U.S. procedural system has become an outlier and probably serves as a model of how not to engage in merger control.

Lars-Hendrik Röller*

Efficiencies in EU Merger Control: Do They Matter?

1. Introduction One of the key issues in competition policy is the objective that guides when and how decisions are taken. This is one of the most vigorously debated issues in recent years. Amongst the proposed standards are the consumer surplus (or consumer welfare) standard, the total welfare standard (i.e. consumer welfare plus firms’ profits), and the goal of protecting the “competitive process” itself. Without going into a detailed discussion of the appropriate standard, it is fair to say that the European Commission is applying a consumer surplus standard. Moreover, this standard is accepted by many other jurisdictions, including the US and other major antitrust agencies around the world. Besides contributing towards international convergence, a consumer surplus standard has certain advantages over the other policy standards. By focusing on consumers, it is clear that competition policy does not protect competitors but competition. It also allows for a proper analysis of the effects of competition on markets and ultimately consumers. Using a consumer surplus standard brings in the use of economics when assessing how market structures and competitive positions of firms affect market outcomes. One of the more significant trends in EU competition policy has been the increased use of economics. Under the so-called “more economic approach”, an effect-based analysis has been emphasized, that is, decisions are taken in light of their effects on markets and, in particular, on consumers. Although it is helpful that Europe has converged on a consumer surplus standard in recent years, there are nevertheless some shortcomings that need to be kept in mind in order to make improvements.1 Focusing explicitly on the consumer raises the legitimate concern that issues surrounding “competitiveness” are not sufficiently taken into account. After all, most economists agree that investment, innovation and more generally dynamic efficiencies are of vital importance for Europe’s economy, and ultimately for Europe’s welfare. * Economic Advisor to Chancellor Merkel, on leave from the European School of Management and Technology, ESMT. (At the time of writing: President, ESMT). This paper is based on a study commissioned by the European Round Table of Industrialists (ERT). A version of this paper was published by the 39th Ottobeuren Seminar. I thank Bas Dessens for excellent research support. 1 There are some areas of competition policy in which the consumer surplus standard runs into theoretical difficulties and thus is harder to implement, such as price discrimination, buyer power, and to some extent state aid. The main reason for these difficulties is the fact that economic efficiency and consumer surplus do not necessarily coincide. Price discrimination, for example, tends to lower consumer surplus, but increases output and thereby economic efficiency.

62  Merger Control in European and Global Perspective Although a consumer surplus standard does not automatically exclude the evaluation of benefits from increased investment and innovation, these benefits are only considered to be of relevance for competition policy agencies if they are ultimately passed on to consumers. It is therefore important for the link between efficiencies and consumer benefits to be taken into account. Whether this has been accomplished in EU competition cases over the last few years is an empirical question. This paper will look at the evidence available to date.

2. Efficiencies in EU merger control In May 2004 a new merger regime took effect.2 Besides the introduction of a new merger test (the so-called “SIEC” test), an explicit efficiency analysis was introduced in the accompanying Horizontal Merger Guidelines (“Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings”). As the new merger regime was fully in line with a consumer surplus standard, it is especially important to look at the conditions under which efficiencies would be considered relevant for merger decisions. The Horizontal Merger Guidelines explain that efficiencies are relevant for merger control in the following terms: “For the Commission to take account of efficiency claims in its assessment of the merger and be in a position to reach the conclusion that as a consequence of efficiencies, there are no grounds for declaring the merger to be incompatible with the common market, the efficiencies have to benefit consumers, be merger-specific and be verifiable. These conditions are cumulative.” (Paragraph 78 of the Guidelines; emphasis added.)

In other words, there are three conditions that need to be met in order for efficiencies to count: (i) consumers must benefit; (ii) the efficiencies must be merger-specific; and (iii) they must be verifiable. The Guidelines further stipulate how these conditions are to be assessed. With regard to consumer benefits: “The relevant benchmark in assessing efficiency claims is that consumers (105) will not be worse off as a result of the merger. For that purpose, efficiencies should be substantial and timely, and should, in principle, benefit consumers in those relevant markets where it is otherwise likely that competition concerns would occur.” (Paragraph 79 of the Guidelines.)

With regard to benefit to merger specificity: “Efficiencies are relevant to the competitive assessment when they are a direct consequence of the notified merger and cannot be achieved to a similar extent by less 2 With effect from 1 May 2004, Regulation (EC) No 4064/89 was repealed and replaced by Regulation (EC) No 139/2004 on the control of concentrations between undertakings.

Lars-Hendrik Röller 

63  

anticompetitive alternatives. In these circumstances, the efficiencies are deemed to be caused by the merger and thus, merger-specific (108). It is for the merging parties to provide in due time all the relevant information necessary to demonstrate that there are no less anticompetitive, realistic and attainable alternatives of a non-concentrative nature (e.g. a licensing agreement, or a cooperative joint venture) or of a concentrative nature (e.g. a concentrative joint venture, or a differently structured merger) than the notified merger which preserve the claimed efficiencies. The Commission only considers alternatives that are reasonably practical in the business situation faced by the merging parties having regard to established business practices in the industry concerned.” (Paragraph 85 of the Guidelines.)

With regard to verifiability: “Efficiencies have to be verifiable such that the Commission can be reasonably certain that the efficiencies are likely to materialise, and be substantial enough to counteract a merger’s potential harm to consumers. The more precise and convincing the efficiency claims are, the better the Commission can evaluate the claims. Where reasonably possible, efficiencies and the resulting benefit to consumers should therefore be quantified. When the necessary data are not available to allow for a precise quantitative analysis, it must be possible to foresee a clearly identifiable positive impact on consumers, not a marginal one. 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.” (Paragraph 86 of the Guidelines.)

All three conditions need to be met in order for the efficiencies to be accepted by the Commission and to count in the decision. The first condition, consumer benefit, follows directly from the consumer surplus standard. An issue we will address below relates to the difference between static and dynamic efficiencies. Static efficiencies are associated with variable and marginal cost savings, while dynamic efficiencies are associated with fixed cost savings, increased investment, and innovation.3 The distinction between static and dynamic efficiencies is important, as it is related to the consumer surplus standard and thus to condition (i), i.e., the requirement that the efficiencies must directly benefit consumers. In the Merger Guidelines the Commission states: “In line with the need to ascertain whether efficiencies will lead to a net benefit to consumers, cost efficiencies that lead to reductions in variable or marginal costs (106) are more likely to be relevant to the assessment of efficiencies than reductions in fixed costs; the former are, in principle, more likely to result in lower prices for consumers (107).” (Paragraph 80 of the Guidelines.)

At footnote 107, the Commission adds that: “Generally, fixed cost savings are not given such weight as the relationship between fixed costs and consumer prices is normally less direct, at least in the short run.” 3 See paragraph 81 of the Merger Guidelines: “Consumers may also benefit from new or improved products or services, for instance resulting from efficiency gains in the sphere of R & D and innovation. A joint venture company set up in order to develop a new product may bring about the type of efficiencies that the Commission can take into account.”

64  Merger Control in European and Global Perspective In other words, one may get the impression that the Guidelines stipulate that static efficiencies are more likely to benefit consumers, and as such are more important in the assessment of merger control. Given the above discussion on the consumer surplus standard, this may lead to an overly restrictive approach. The introduction of an explicit efficiency analysis in merger control in May 2004 constitutes an ideal institutional scenario where one can test whether and to what extent efficiencies are taken into account within the framework of a consumer surplus standard. The next subsection will take a look at the evidence from the case law.

3. Evidence on the use of efficiencies in EU merger control In order to understand the role of efficiencies, we have considered all phase II merger cases decided through early 2009 under the EU Merger Regulation (i.e., Regulation 139/2004). Table 1 summarizes these cases. As can be seen, there were a total of 49 cases, which led to 37 published decisions. Of the published decisions, 19 cases were cleared with no remedies imposed, while 17 were cleared with remedies. Finally, one case was prohibited (Ryanair’s proposed takeover of Aer Lingus). We thus use the 37 published decisions as our sample for this study. For a complete list of all cases, see the Appendix. Table 1. Sample for analysis: phase II merger cases under Regulation 139/2004 139/2004 case decisions Art. 8(1)

Art. 8(2)

Art. 8(3)

Aborted/ withdrawn

Total

Published

19

17

1

0

37

Not published *

0

1

0

11

12

Total

19

18

1

11

49

Source: http://ec.europa.eu/competition/mergers/cases/. *As of 29 May 2009.

Recall that a consumer surplus standard does not imply that efficiencies or profits to firms do not matter in the decision. It only implies that efficiencies and profits to firms matter when they benefit consumers, are merger specific, and are verifiable. In addition, they must be alleged by the merging parties, as the Commission will only consider efficiency arguments if the parties put them forward. Accordingly, using the published decisions, we categorize the 37 cases in our sample according to whether efficiencies have been “alleged” by the parties, are deemed by DG COMP as “verifiable”, “merger specific”, or having “consumer benefit”. Finally, we report the number of cases where all three conditions are present.

Lars-Hendrik Röller 

65  

We also separate the above categorization into static and dynamic efficiencies, in order to investigate whether static efficiencies are more important in the assessment of merger control. Recall that static efficiencies are those related to a decrease in variable costs, which in a competitive setting are likely to be passed on to consumers in the short run. Dynamic efficiencies are those related to fixed cost savings, increased investment, and innovation, which in a competitive setting are likely to be passed on to consumers in the long run. Table 2 reports our findings for the 37 cases. As can be seen in Table 2, static efficiencies have not played a major role in phase II merger evaluations since 2004. Static efficiencies were alleged by the parties in only 5 of the 37 cases. Of those 5 cases, there were 3 cases where those efficiencies were deemed either verifiable, merger specific, or beneficial for consumers. In 2 of those cases, all three conditions were deemed to be satisfied, i.e. the efficiencies were accepted as being relevant for the merger assessment. In other words, alleged static efficiencies were not infrequently accepted by DG Comp: in 2 out of 5 cases, alleged static efficiencies were accepted, which is in about 40% of the cases. This suggests that parties were rather successful in claiming static efficiencies, even though the static efficiency claims proved not to be decisive in either of the 2 cases. Table 2: Efficiencies in EU phase II merger decisions: 2004–2009 Static efficiencies

Dynamic efficiencies

Total number of cases

Alleged *

5

6

37

Verifiable

3

2

37

Merger-specific

3

3

37

Consumer Benefit

3

3

37

Accepted

2

0

37

Decisive

0

0

37

Source: http://ec.europa.eu/competition/mergers/cases/. *In Case M.5046 – Friesland/ Campina, the Commission indicated that the parties had put forward an efficiency defence, but did not provide details. Therefore, those efficiency claims could not be included in the above overview.

Table 2 also addresses the role of dynamic efficiencies. Dynamic efficiencies have not played a major role in phase II merger evaluations since 2004 either, having been alleged in 6 out of 37 cases. Furthermore, alleged dynamic efficiencies have never been accepted by DG Comp, failing either the “verifiability”, mergerspecificity”, or the “consumer benefit” criteria. Interestingly, there is no evidence that DG Comp follows a per se approach with regard to dynamic efficiencies. Indeed, the three criteria are deemed to hold individually in at least one case in our sample, although they never hold simultaneously in one case.

66  Merger Control in European and Global Perspective The following Box summarizes the cases where efficiencies have played a role our sample of EU merger decisions.4 Remarkably, in 2 cases DG COMP indicated pro-competitive (static) efficiencies, yet the parties did not claim them. As a result they were not taken into account. In Itema Holding/Barcovision Division (M. 4874, 4 August 2008), which was a vertical acquisition of a sensors and other input manufacturer for the textiles industry by a producer of textile machinery, DG Comp suggested it would have accepted a static efficiency claim with regard to the elimination of double mark-ups, but this could not be verified as the parties had not claimed such efficiency. Similarly, in DONG/Elsam/Energi E2 (M.3868, 14 March 2006), which was an acquisition of regional Danish electricity generation incumbents and electricity suppliers by the state-owned gas incumbent, DG Comp indicated that any efficiencies due to dual fuel supplies can be pro-competitive but as DONG had not set out an efficiency defence, they could not verify whether the relevant criteria were met. The role of efficiencies in EU merger cases • T-Mobile Austria/Telering (2005): Mobile phone operators. Dynamic efficiencies claimed, consumer benefit not accepted. • Inco/Falconbridge (2006): Companies active in the mining, processing and refining of nickel and other metals. Both static and dynamic efficiencies were claimed, but deemed not to be merger specific or to benefit consumers. • Ryanair/Aer Lingus (2006): Two leading airlines operating from Ireland. Both static and dynamic efficiencies where claimed, but not deemed verifiable or merger specific. However, the efficiency assessment was indicated as not being decisive: “even if these criteria were met, the claimed efficiency gains would in all likelihood be insufficient in magnitude to reverse the anti-competitive effects identified […]” (recital 1127 of the decision) • TomTom/Tele Atlas (2007): Vertical acquisition of a navigable digital map provider by a portable navigation devices producer. The claimed static efficiencies were accepted (elimination of double mark-ups), but the dynamic ones were not because they were deemed not verifiable. • Nokia/NAVTEQ (2008): Vertical acquisition of a navigable digital map database provider by a mobile telephone producer. Static efficiencies were accepted (elimination of double mark-ups), but dynamic efficiencies were deemed not verifiable or merger specific. • KLM/Martinair (2008): Dutch airlines active in the transport of passengers and cargo. Both static and dynamic efficiencies were claimed but rejected due to a lack of verifiability.

4 In M.5046, Friesland/Campina (17 December 2008), a horizontal merger between two Dutch companies active in a range of dairy products, the Commission indicated that the parties put forward efficiencies, but did not provide details. It was therefore unnecessary to consider the efficiency defence invoked by the parties. It is not clear whether the claimed efficiencies were static or dynamic, or whether they would have been accepted, but they were definitely not decisive.

Lars-Hendrik Röller 

67  

4. Remarks on efficiencies in EU merger control The main finding of this section is that efficiencies have not played a major role in phase II EU merger evaluations since 2004. Only in 6 of the 37 published cases (i.e., 16%) did efficiencies play any role at all, that is to say, merging parties claimed efficiencies only in those 6 cases. Once static efficiencies were claimed, they were accepted in about 40% of the cases, but dynamic efficiencies were never accepted as being relevant for the merger assessment. In the end, efficiencies have never been decisive in EU merger decisions. Since the economic rationale of mergers is to create efficiencies, this finding is rather disappointing. As empirical evidence shows, there are mergers with significant efficiencies, while others produce virtually no efficiencies at all. As a result, a merger assessment that does not consider efficiencies more explicitly is likely to come to wrong conclusions. A related question is why efficiencies are not more often claimed by the parties. There are two possible explanations. The first possibility is that, in 31 of the 37 cases, there were no efficiencies to be claimed. This is unlikely. Second, efficiencies are not being claimed because it is perceived that they are unlikely to contribute to a positive assessment of the proposed merger. There is a further refinement of the second explanation: by claiming efficiencies, the parties signal that they have a “weak” case, i.e. they signal that the proposed merger is likely to be anticompetitive. In other words, claiming efficiencies may be held against the parties as a signal that the merger would lead to the creation or reinforcement of market power. This signaling dilemma is further supported by the Commission’s approach to not providing information on efficiencies. “It should be noted that submitting information in response to Section 9.3 is voluntary. Parties are not required to offer any justification for not completing this section. Failure to provide information on efficiencies will not be taken to imply that the proposed concentration does not create efficiencies or that the rationale for the concentration is to increase market power.” (Commission Regulation 802/2004 of 7 April 2004, footnote 1.)

In other words, submitting information on efficiencies is both voluntary and failure to provide this information does not imply that the merger is anticompetitive. Both of these conditions dampen the incentives for firms to come forward with evidence on efficiencies: if parties claim efficiencies, it may be held against them; if they do not, it will not be held against them. This approach to information on efficiencies is reminiscent of the so-called efficiency offence, where evidence of efficiencies is turned against the merging parties because they may lead to (or signal) market power. Efficiency offence arguments have been heavily criticized. In the current context, an informational efficiency offence will often not lead to firms providing the much needed information. If efficiency analysis is to play a more systematic role in merger control, the right incentives have to be in place to encourage parties

68  Merger Control in European and Global Perspective to provide the necessary information. This is currently not the case. For instance, asking parties to provide a justification for the failure to provide information on efficiencies would create better incentives for an appropriate efficiency analysis. Despite this favourable treatment of not providing information on efficiencies, the fear that efficiencies will not be accepted (or will be turned against the parties) is not borne out by our case evidence as far as static efficiencies are concerned. The Commission has looked upon static efficiencies in a positive way. After all, in about 40% of the cases (2 out of 5), static efficiencies have been accepted. It appears that a more pro-active stance by the merging parties – at least vis-à-vis static efficiencies – is warranted. The situation is rather different with regard to dynamic efficiencies. As stated in the Merger Guidelines, the current approach by the Commission is to emphasize static efficiencies, not dynamic efficiencies. This approach is confirmed by our empirical findings. In none of the 6 cases where dynamic efficiencies were claimed did DG Comp accept them as being relevant for the assessment. Nevertheless, there does not appear to be a per se approach to dynamic efficiencies: all three conditions have been met in different cases – but they have not been satisfied cumulatively in the same case.

5. Conclusion The empirical evidence presented in this paper does not suggest that efficiencies have much of a role in merger assessments. Neither static nor dynamic efficiencies have been decisive in EU merger cases. The reason for this irrelevance of efficiencies in merger control is not that DG Comp frequently does not accept efficiency claims; it is primarily because the parties do not put forward efficiency claims in the first place, since the current legal approach does not provide any incentive to do so. The current context can be characterized as one of an informational efficiency offence, which will not lead to firms providing the much needed information. Given the importance of efficiencies as the rationale for mergers, the role of efficiencies in merger control should be strengthened by creating better incentives for parties to submit the relevant information. The role of dynamic efficiencies is even more limited than static efficiencies. This approach has been adopted deliberately by the Commission, and the evidence of the last few years confirms this. The evidence also shows that there is no per se approach to dynamic efficiencies, which implies that the claimed dynamic efficiencies are assessed thoroughly by DG Comp. Given the economic importance of dynamic efficiencies, it may be time to be more explicit about what kind of dynamic efficiencies are acceptable.

Lars-Hendrik Röller 

69  

Appendix Merger decisions used (all phase II cases under Regulation 139/2004 through 29 May 2009) Year

Case

Decision

Published*

2008

M.5153 – ARSENAL / DSP

Art. 8(2)

No

2008

M.5141 – KLM / MARTINAIR

Art. 8(1)

Yes

2008

M.5046 – FRIESLAND / CAMPINA

Art. 8(2)

Yes

2008

M.4989 – ALO / MX

Aborted/withdrawn No

2008

M.4985 – BHP BILLITON / RIO TINTO

Aborted/withdrawn No

2008

M.4980 – ABF / GBI BUSINESS

Art. 8(2)

Yes

2007

M.4956 – STX / AKER YARDS

Art. 8(1)

Yes

2008

M.4942 – NOKIA / NAVTEQ

Art. 8(1)

Yes

2008

M.4919 – STATOILHYDRO / CONOCOPHILLIPS

Art. 8(2)

Yes

2008

Art. 8(1)

Yes

2007

M.4874 – ITEMA HOLDING / BARCOVISION DIVISION M.4854 – TOMTOM / TELE ATLAS

Art. 8(1)

Yes

2008

M.4799 – OMV / MOL

Aborted/withdrawn No

2007

M.4781 – NORDDEUTSCHE AFFINERIE / CUMERIO Art. 8(1)

Yes

2007

M.4747 – IBM / TELELOGIC

Art. 8(1)

Yes

2007

M.4734 – INEOS / KERLING

Art. 8(1)

Yes

2007

M.4731 – GOOGLE / DOUBLECLICK

Art. 8(1)

Yes

2007

Art. 8(2)

Yes

2007

M.4726 – THOMSON CORPORATION / REUTERS GROUP M.4662 – SYNIVERSE / BSG (wireless business)

Art. 8(1)

Yes

2007

M.4647 – AEE / LENTJES

Art. 8(1)

Yes

2007

M.4525 – KRONOSPAN / CONSTANTIA

Art. 8(2)

Yes

2007

M.4523 – TRAVELPORT / WORLDSPAN

Art. 8(1)

Yes

2007

Aborted/withdrawn No

2006

M.4513 – ARJOWIGGINS / M-REAL ZANDERS REFLEX M.4504 – SFR / TELE 2

2006

M.4498 – HgCAPITAL / DENTON

Aborted/withdrawn No

2006

M.4439 – RYANAIR / AER LINGUS

Art. 8(3)

Yes

2006

Art. 8(2)

Yes

Art. 8(1)

Yes

2006

M.4404 – UNIVERSAL MUSIC GROUP / BMG MUSIC PUBLISHING M.4403 – THALES / FINMECCANICA / AAS / TELESPAZIO M.4397 – CVC / FERD / SIG

2006

M.4381 – JCI / FIAMM

Art. 8(2)

Yes

2006

M.4215 – GLATFELTER / CROMPTON ASSETS

Art. 8(1)

Yes

2006

M.4209 – THULE / SCHNEEKETTEN

Aborted/withdrawn No

2006

M.4187 – METSO / AKER KVAERNER

Art. 8(2)

Yes

2006

M.4180 – GAZ DE FRANCE / SUEZ

Art. 8(2)

Yes

2006

Art. 8(2)

Yes

Aborted/withdrawn No

70  Merger Control in European and Global Perspective 2006

M.4094 – INEOS / BP DORMAGEN

Art. 8(1)

Yes

2006

M.4000 – INCO / FALCONBRIDGE

Art. 8(2)

Yes

2005

Art. 8(1)

Yes

2005

M.3975 – CARGILL / DEGUSSA FOOD INGREDIENTS M.3923 – AMI / EUROTECNICA

2005

M.3916 – T-MOBILE AUSTRIA / TELE.RING

Art. 8(2)

Yes

2006

M.3868 – DONG / ELSAM / ENERGI E2

Art. 8(2)

Yes

2006

M.3848 – SEA-INVEST / EMO-EKOM

Art. 8(1)

Yes

2005

M.3796 – OMYA / J.M. HUBER PCC

Art. 8(2)

Yes

2005

M.3696 – E.ON / MOL

Art. 8(2)

Yes

2005

M.3687 – JOHNSON & JOHNSON / GUIDANT

Art. 8(2)

Yes

2005

M.3653 – SIEMENS / VA TECH

Art. 8(2)

Yes

2005

M.3637 – TOTAL / SASOL / JV

Aborted/withdrawn No

2005

M.3625 – BLACKSTONE / ACETEX

Art. 8(1)

2004

M.3423 – RWA / AMI / INTER-FERT

Aborted/withdrawn No

2004

M.3178 – BERTELSMANN / SPRINGER / JV

Art. 8(1)

Source: http://ec.europa.eu/competition/mergers/cases/.

Aborted/withdrawn No

Yes Yes

V. Merger control and best practices

5 PANEL DISCUSSION

Chair: Philip Lowe

Speakers: John Boyce Rachel Brandenburger Jochen Burrichter Thomas Deisenhofer Kirsten Edwards

Cal Goldman William Kovacic Howard Shelanski Mario Siragusa Sven Völcker



328  Merger Control in European and Global Perspective

PANEL 5: Merger control and best practices Philip Lowe: This final session is about best practices. What is best, and whose practices they are, is going to be revealed to us by our three speakers. First of all, John. John Boyce: Well, I must confess that when I was asked to focus on best practices in merger control I was a little bit uncertain about what I was meant to be addressing. Was I supposed to be commenting on those guidelines that some competition agencies issue, which are labelled “best practices”? Or was I being given a bit more flexibility, to talk about what I think is the best thing competition authorities should be doing and how we, as practitioners, should do our jobs? The first thing I did, actually, was to go to the dictionary to find out what “best practices” means. I was hoping to find some sort of Shakespearian quote or something showing that it was first used by Henry II (laughter). But to my surprise, the oldest reference in the Oxford English dictionary to the use of the word “best practices” was only in 1984. That was perhaps in the days when accountants were more innovative than they now are, because it pre-dated the concentration in that industry (more laughter). According to the OED, in understanding accounting in 1984, that was the first time that the concept was used. It goes on to talk about another use of it in 1993 by John Kay, who some of you will know as an economist who wrote about corporate success.1 He explained that, if you want to run a successful corporation, one thing you might do, obviously, is to look at what other firms are doing and copy it. This strategy, he said, is more felicitously expressed as “adopting best practice”. So that’s not a bad idea. Just see what other people are doing, copy it, and that’s best practice. Well, things have moved on a little bit since then. One dictionary definition which I did quite like is from businessdictionary.com, which describes “best practices” as “methods and techniques that have consistently shown results superior than those achieved with other means, and which are used as benchmarks to strive for”. It did, however, go on to say that there is “no best practice that is best for everyone (laughter), or in every situation, and no best practice remains best for very long, as people keep on finding better ways of doing things.” Then, if you look at one of my favourite dictionary sources, Wikipedia, it’s very modern, very democratic. It comes up with similar concepts and it actually picks up some of the terminology that Bill was using earlier on. Wikipedia suggests that best practice is to be measured by its effectiveness: getting the best results. Best practice should also be about getting things done in the most efficient way, with the least amount of effort. Now, that all sounds very sensible to me, but Wikipedia actually goes on to say that there is a danger of best practices becoming a bit of a business buzz word. It says that, “as the term has become more popular, some more 1 See John Kay, Foundations of Corporate Success: How Business Strategies Add Value, OUP, 1993/1995. For additional information on Kay, see http://www.johnkay.com/about.

Panel 5 

329  

organisations have begun using the term ‘best practices’ to refer to what are in fact merely rules, causing a linguistic drift in which a new term such as ‘good ideas’ is needed to refer to what would previously have been called ‘best practices’”. I think that really sets the scene for what I think best practices should really be about. It’s all about learning from experience. That’s from your own experience, but also by looking at what other people are doing and by learning from their experiences. They might have done something wrong, so you say that’s what we want to avoid doing. It’s all about learning by doing and learning by watching. If you look at the whole area of international merger control that we’ve been talking about for the last couple of days, it’s a prime area where best practice makes sense. Increasingly, regulators are applying similar rules, and increasingly they clear similar deals, or even the same deal, so we can try to learn from what others are doing. In my paper, I try to focus on three areas where best practices make sense. The first area is best practices by agencies and by lawyers, facilitating the process of advising clients about how to get deals through. We’ve been talking about alliteration, well some time ago as a junior lawyer I learned the 6 P’s: proper preparation prevents piss poor performance (laughter). Now, increasingly that is a problem when you’re advising on international deals; where should you be notifying? I think firms are doing this regularly; you’ve got your databases, you’ve got the guides that people go to. Even on the most mundane deals, you may have to notify in a number of jurisdictions. I was just talking to Nick during the break about how, when I was a junior lawyer, the bane of our lives was being asked: “Can you review this agreement to see whether it is caught by the Restrictive Trade Practices Act in the UK?” Now, I’d have to do this and I’d think that this isn’t what I wanted to do as a competition lawyer. I wanted to deal with effects and problems. So, I think that as a legal community we were all quite pleased to see that piece of arcane legislation repealed, but what’s come along to replace that for our junior lawyers now and for the junior case handlers of the European Commission is dealing with cases that raise no competition issues whatsoever. Now, this is where I think there are some best practices out there. The ICN has been fantastic at trying to point people in the right direction by coming up with common techniques for cases, that is, both cases that shouldn’t raise problems and those that do. One of the key buzzwords for me in this area is transparency. There can be clarity as to what rules should be about, and if agencies are transparent about how their rules operate in practice that makes life much easier for us in practice. If I’ve got a couple of suggestions, my first one picks up on the point that the ICN is not a group of legislators, it’s just agencies so they don’t make any binding rules. However, I think there is still some scope for advocacy by the agencies, just making it clearer to their governments that the rules don’t make sense. It sounds as if John is having the problem the other way around at the moment, in that he’s being pushed in the direction of rules that could be more bureaucratic. I’ve talked a bit about the issue of rules, which are triggered purely by jurisdictional thresholds. I just wonder, raising the same point again, whether

330  Merger Control in European and Global Perspective there is scope for regulators – I used the word “regulator” there actually! – to exercise some self-restraint on whether to enforce the rules. Thomas, you said that you don’t have the power to do that, but to some extent you do. Let’s say a Korean company and a Japanese company set up a coffee shop in Tanzania. Say they happened to meet the thresholds and didn’t think of notifying the European Union, but someone spotted it when they were reading the newspaper in Tanzania while on holiday from the European Commission. I don’t think you would be using the Commission’s resources effectively by requiring them to file a notification. Fold up the paper and move on! (laughter) Another idea that I’ve got for the ICN is that if you hunt around the ICN’s website, you’ll find some summaries that some jurisdictions have put in, showing what their rules are, what their thresholds are. They’re pretty long, and quite difficult to understand, so I wonder whether there might be some scope for cleaning them up a bit and encouraging members to do that. You can do that behind closed doors without NGAs having to hear about it. Possibly agencies could even self-assess and spell out where the rules are ICN best practice-compliant and where they’re not. There’s nothing wrong with having rules that don’t comply with the ICN’s best practices, but it is useful for us as practitioners to know when rules don’t comply. So those are the areas where I think more can be done. A lot has been done to make life easier for us advising clients at the outset on deals. The other area of best practice is for the conduct of investigations, and we talked a lot about that in the previous session. Best practice is all about using the 3 Cs: convergence, cooperation and comity. Thomas talked about challenges, which is another C word. I like C words when it comes to competition! You were talking about the cultural challenges, the confidentiality issues – another C word that you didn’t even mention was a C word – and the challenges of communication that you have dealing with the other regulators. We heard Rachel’s more complicated MRTD acronym, but I took those letters and just threw them into the Scrabble bag and they came back as 4 Cs: mindfulness is consciousness of what the other regulators are doing; respect, well Nick said it – consideration, or courtesy or care. There you are, that’s three words (laughter). Trust is confidence – have confidence in what your other ICN members are doing; and dialogue is, of course, communication. So, there are plenty of C words out there. The last area I wanted to touch on briefly is remedies, which is an important area… Audience: Commitments! (laughter) Boyce: Yes, commitments, sorry! This is an area where, as a practitioner, it is very important to get the regulators working together because the risk of having divergent commitments can actually increase the antitrust “price” and make deals unattractive. Philip was talking right at the end of the last panel about the airlines sector, and I spent some of the past two years dealing with a transatlantic alliance case.2 I found that quite bizarre in some ways, because you have the 2

M.5747, Iberia/British Airways (14 July 2010).

Panel 5 

331  

sort of odd US practices dealing with cooperation in the airline sector, where the DOJ has got no jurisdiction at all and the Department of Transportation deals with it in an incredibly so-called “transparent” way where everything is on the docket. Also, how can you have cooperation between the Commission and the DOT under those circumstances? But you get around that somehow. The normal US approach, where orders to show cause come up quite quickly, as happened with Star Alliance and Sky Team, didn’t seem to happen in our case for some reason. That was because the Commission was cooperating so well, and not just with the Department of Transportation. I’m complimenting the Commission here, actually! (laughter) There was very close cooperation between the Commission and the DOT and there was also very close cooperation between the Commission and the DOJ. Where there didn’t seem to be close cooperation was between the DOT and the DOJ, which I found quite intriguing. I’ve stayed away from talking about best practices in the sense of guidelines because I think Jochen’s going to say a bit about that. In my paper3 I’ve tried to summarize in an annex the available best practice documents. I also commend to you, if you’re not familiar with them, the ICN’s Recommended Practices;4 I’ve summarized them in a separate annex in my paper. They’re very good and they’ve withstood the test of time, being written some five years ago. I don’t think they need to be updated much, even though one of the recommendations is that practices should be updated from time to time. The European Commission is a bit delinquent in that respect, because the Commission’s best practices for mergers pre-date the 2004 merger regulations,5 so all the references are wrong. I’ve also in an annex put down some recommended practices for lawyers, which I think we generally all try to comply with. We have guidelines like that, and I’m sure you do too, that are there to help our junior lawyers work out how you should be dealing with relatively simple cases, which may need to be notified all over the world and where you need to coordinate with local lawyers as required. To conclude, I think best practice is a great thing to strive for. If we keep on developing best practices and coming up with new, good ideas, then ultimately we will get to Utopia. But you’ve got to keep on trying to get there, because even when you think you are there, you’re not. It’s going to change, so keep on moving forward. Lowe: As school children in the North of England, we were actually taught to recite this little rhyme: “good, better, best. Never let it rest, until the good is better and the better best.” (laughter). Next we have Jochen. Jochen Burrichter: Well, it’s very difficult for me to add to what John has already said. I would like to categorize, or to analyze, the existing best practice rules which are in place. The best practices which you can observe provide guidelines 3 Boyce, “Best Practice in Merger Control: It Ain’t What You Do It’s the Way That You Do It … And That’s What Gets Results”, this volume, pp. 343 et seq. 4 http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 5 http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf.

332  Merger Control in European and Global Perspective on how merger reviews can be conducted, or should be conducted, if issued by international institutions in terms of transparency, case handling, communication, confidentiality and inter-agency coordination. If you try to categorize the existing best-practice rules, you can distinguish between the external best practices, which are issued for the benefit of merger control agencies by international institutions like the ICN, or between agencies. The second category is internal best practices, as I call them. That is, the best practices issued by the agencies themselves for the merging parties and for their own practice. The third category, which John covered at the end, is best practices issued by companies or by law firms. Let me concentrate on the two first categories. The external best practices for agencies are an important means to influence convergence. We have two categories, as I already said: firstly, the international organizations issue best practices. By far the most significant driver with regard to these best practices is the ICN, which seeks to facilitate cooperation and the exchange of know-how between competition authorities globally. The Merger Working Group published a comprehensive set of materials on best practices, which John already recommended that we should study. The most important contributions are the Guiding Principles for Merger Notification and Review6 and the Recommended Practices for Merger Notification and Review Procedures.7 Further important guidance for merger control agencies comes from the OECD. In 2005, the Council of the OECD adopted a Recommendation that touched fairness and cooperation.8 And then another document was issued by the Business and Industry Advisory Committee of the OECD; that committee published their Recommended Framework for Best Practices in International Merger Control Procedures in October 2001.9 The other important category of best practices consists of bilateral best practices, and by far the most important one is the Best Practices on Cooperation in Merger Investigations of the US-EU Merger Working Group.10 If you try to analyze the existing best practices, what is the scope? If you look at the external best practices, you see that, consistent with the rationale of external best practices, i.e., providing guidance for merger control authorities on how to shape their procedures, external best practices only establish general and broad principles for an efficient and transparent set up. These principles are usually verbalized as recommendations to agencies, therefore external best practices provide a model rather than a formal code. So you can conform to them but you are not obliged to do that. Procedural issues that are usually dealt with include recommendations on the structural alignment of notification thresholds with, for the sake of legal certainty and reduction of unnecessary transaction costs, a clear preference for revenue-based thresholds rather than market share http://www.internationalcompetitionnetwork.org/uploads/library/doc591.pdf. http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 8 OECD, Recommendation of the Council on Merger Review (2005), http://www.oecd.org/ dataoecd/3/41/40537528.pdf. In April 2012, the OECD’s Competition Committee published a report (based on the proceedings of Working Party No. 3) entitled Procedural Fairness and Transparency: Key Points. See http://www.oecd.org/dataoecd/16/34/50235955.pdf. 9 http://www.biac.org/statements/comp/BIAC-ICCMergerPaper.pdf. 10 http://www.ftc.gov/os/2011/10/111014eumerger.pdf. 6 7

Panel 5 

333  

thresholds. Moreover, the recommendations are pronounced regarding the agencies’ requirements on the timing of a notification, a clear determination of the triggering event and preferably no deadline, and the content of the notification. If you analyze the internal best practices which exist, frequently they deal with the details regarding the case team set up, the decision-making process, pre-filing guidance, communication issues – for example the timing of the Statement of Objections – guidance on the conduct of meetings, and access of third parties to such meetings. There is also relevant guidance on the right of access to the file, timing, accessible documents, entitled persons, procedures on the requirement for the initial notification, and safeguarding confidentiality with regard to the business secrets provided by the merging parties. The aims and benefits of merger best practices can be summarized very quickly. First of all, with regard to agencies, it is clear that merger control agencies benefit from the external best practices because they assist antitrust agencies with respect to a structuring of the relevant merger review procedures, therefore acting as a catalyst for better convergence. They also help to streamline the intra-agency procedures and therefore boost efficiency with respect to the merger review. The promotion of external best practices is desirable, but I think, as John said, there is still something to do. The increasing proliferation of merger control regimes and the variety of jurisdictions which affect globally active undertakings heighten the need for national agencies to publish more internal best practices on the conduct of their procedure. With regard to external best practices we have the ICN’s Recommended Practices,11 which is an excellent document that still can be improved, as John already explained. If one would try to define the minimum requirements of the internal best practices, I think it is clear that each internal best practice should deal at least with a minimum standardized set of issues which cover the most important aspects of merger control procedures. I think these are: pre-filing consultations, the scope of requests for information, the conduct of meetings, involvement of third parties, access to the file and confidentiality issues. To what extent are these best practices binding? An authority that issues such best practices normally should be bound to what they have pronounced; if there is a good reason, they may deviate from it. Another aspect that I want to mention is that in favour of the parties, or in favour of a better procedure, the binding effect should be limited. That especially applies, for example to access to file, if in appropriate cases access to file can be given earlier than you normally find in the best practices. So I would like to conclude with that and let me add that most of the proposals that John made, I would also have made. Thank you. Mario Siragusa: I will concentrate on what Jochen characterized as internal best practices, and in particular the Commission’s best practices. The Commission has made a substantial effort trying to codify and standardize its internal procedures, and there are in fact two groups of best practices: one specific to mergers,12 and 11 12

http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. Link above note 5.

334  Merger Control in European and Global Perspective a second, broader best practice for antitrust proceedings.13 On mergers, the 2004 best practices include, of course, a standard model for commitments and the trustee mandate. Then there are the more recent remedy notices,14 so that’s the block of best practices which pertain to mergers. We recently saw more best practices for antitrust proceedings published.15 There, we have of course the best practices for the submission of economic evidence16 and the guidance on the role of the Hearing Officer.17 So these are the two groups that I was talking about. It seems to me that all this documentation plays a very important and excellent role, and is inspired basically by enhancing the cooperation spirit of the relationship between the Commission and the companies involved in those various processes. That is clear, for instance, from the efforts that have been made to clarify the pre-notification process. What I find particularly important is the agreement that has been reached for the Commission to pursue fact-finding also in that process and even to talk to third parties, with the agreement of the merging entities. I think this is a very interesting development, which has clarified a very important portion of what happens during the pre-notification process. Also, I think these best practices have been inspired generally by an effort to be more transparent. The state of play meetings and the access to some of the key documents before the SO is sent all seems to me to be a step in the direction of more transparency. There have also been important decisions in the more general antitrust best practices, published more recently. In particular, I found this idea that there should be an earlier formal opening of the proceedings helpful, so that everybody knows that the proceeding is starting. Also, publications on the development of the proceeding, the issuing of the SO and the final decision, so that the opening of the procedure will be publicly announced so that everybody is informed about the progress of the proceedings. Now I would simply like to make a couple of remarks on ways in which to improve this, because I think everything can always be improved. First of all, a reflection on what the nature of all these documents is. I suppose that, particularly to the extent that they standardize proceedings that are applied by the Commission in individual cases, the Court will interpret those in the same way it has done with the Commission’s other guidelines. So we may see a development of case law 13 See now Commission Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU, 2011 OJ C308/6. 14 See Commission Notice on remedies acceptable under the Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, 2008 OJ C267/1. 15 See above note 13. 16 For the submission of economic evidence (in antitrust as well as merger cases), see http:// ec.europa.eu/competition/antitrust/legislation/best_practices_submission_en.pdf. 17  See http://ec.europa.eu/competition/consultations/2010_best_practices/hearing_officers.pdf. At the time of the 2010 Workshop proceedings, the terms of reference of the Hearing Officer were stipulated in Commission Decision on the terms of reference of hearing officers in certain competition proceedings, 2001 OJ L162/21. Those terms of reference were later expanded to some degree (e.g., during the investigative phase, or in the context of commitment proceedings under Article 9, etc.) by Decision 2011/695/EU of the President of the European Commission of 13 October 2011 on the function and terms of reference of the hearing officer in certain competition proceedings, 2011 OJ L275/29. For details, see Wouter Wils, “The Role of the Hearing Officer in Competition Proceedings before the European Commission”, 35 World Competition 431 (2012).

Panel 5 

335  

which says that a state of play meeting can take place and, if requested, must take place. So we can imagine this at least for certain aspects of these guidelines. I can imagine that there are other areas of the best practices which probably are more general and which may not be subject to judicial intervention. But this is probably something that we may see developing in the next few years. As to the possible improvements, I found that there are four areas in which I think some improvements can be made. More generally, on access to the file, I think that it should be possible to allow access from the beginning of the proceedings, both in merger cases and in antitrust cases. Once the procedure is open, and of course in the case of mergers with notification and in the case of antitrust proceedings with the formal opening of the proceedings, the file is being built. In some of our national procedures, parties have the right to access the file periodically to be informed as to the exact status of the file and the documents contained therein. This allows the party to be on a level playing field with the Commission, because when the Commission comes to know, for instance, what third parties are saying on the proposed merger, I do not see why the merging parties should not be put in exactly the same position as the Commission. This is even more important in antitrust proceedings, but it certainly could apply also to mergers. So I think that giving access to key documents before the SO is a good step in the right direction, but I propose as a next step to see whether it is possible to give full access to the file periodically during the course of the proceedings. A second suggestion relates to contacts with third parties. The Commission will of course meet and talk to third parties in the process of merger review, as in the process of antitrust proceedings. But here there is this distinction between formal and informal contacts, which I find troublesome. More generally, I think that all meetings that Commission officials have with parties – both parties to the proceedings and third parties – should somehow be recorded. There should be minutes of the meetings, summary minutes of what has been said, and this should be put in the file, with access for all interested parties. Again, this is done by some national authorities and I don’t see why this should not also be done at the Commission level. This is particularly if the results of those meetings somehow affect the Commission’s thinking and can be used by the Commission when it makes its findings. There should be some record of it in the file; otherwise there is no way it could be subject to review by the Court. The third area where improvements could be made concerns the role of the Hearing Officer. Now, I’m not sure if this can be done by way of best practices. Maybe this needs an ad hoc communication of the Commission. But certainly there were some expectations when the Commission announced the recent publication of the best practices that this could have been the occasion to somehow modify and enhance the role of the Hearing Officer. I don’t think that this has been done. The Hearing Officer’s role has been codified and explained, but I did not find anything new in the document. I’d be interested to hear whether you agree with this analysis or whether you think that there has been a change in that respect. I

336  Merger Control in European and Global Perspective certainly think that the Hearing Officer could play a more comprehensive role, not limited to procedure and access to file issues.18 My last comment is on an improvement which has been made and which probably relates more to general antitrust proceedings and not so much to merger review. It has been announced that there will be an early, official opening of the antitrust proceedings. Generally, what we have seen is that this opening is done with a very short communication, sometimes only one page or half a page as an indication of what the area is that the Commission is investigating. Not much information is given as to the scope and the real possible allegation or preliminary theory on which the opening of the case is based. To the extent that the Guidelines say that the opening of the procedure will come after an initial assessment of the case, maybe it could be useful to have a more clear indication of the nature and scope of the matter when the Commission announces the opening of the proceedings. So these are my comments on the specifics. Otherwise, I think that it is a very good initiative and I am sure that the Commission will continue to update its best practices because of the need for continual improvement. This is particularly because some of those issues covered by the best practices the Commission really have to do with due process and with the way in which the specific proceeding is going to be followed by the Commission. Thank you. Lowe: Thanks very much. John said it right at the beginning: you can’t have best practice unless you have promised to improve it regularly. For those of you who have not been to the Competition Workshop before, it has been a major driver of improvement on substantive issues and on procedural issues. The 2009 Workshop featured a barrage of criticism from some of the people around this table, but others as well, saying that the Commission didn’t do enough in this area.19 I think you can actually attribute the appearance of a document on best practice in antitrust proceedings and the submission of economic evidence and the organization of hearings precisely to the pressure you put on us on that occasion. This is just as, three or four years ago, I would say that thanks to you and to the Workshop, I was able to argue – with great difficulty – for a Guidance Paper on the application of Article 102 to unilateral conduct.20 So we have a clear indication of a cycle of debate and reaction. It’s not surprising that you don’t see a change in the precise role of the Hearing Officer, because it would have been rather odd for Neelie Kroes or myself to have left to Commissioner Almunia the test of having to do something radical. That’s up to him and his team to consider. Now, would anyone else like to comment on these best practices? 18 As seen the previous footnote, the Hearing Officer’s terms of reference were enhanced to some extent pursuant to Decision 2011/695/EU of the President of the Commission. 19 See Ehlermann and Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011. 20 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.

Panel 5 

337  

Sven Völcker: Just briefly on a couple of points. First, on Mario’s suggestion that best practices might become justiciable: I personally have my doubts, speaking under Judge Forwood’s control, of course, that the Court would ever get to that point. Looking at the past litigated merger cases, perhaps the procedural deficiencies have not been well pleaded, but they certainly have not succeeded – with the notable exception of Schneider,21 which may have been a special case. The Courts’ substantive review has been much more searching than perhaps its willingness to accept that there were certain flaws in the procedure here and there. I’m also not sure we should wish for that kind of review, because the moment it is clear that the Commission will be held to the best practices in every case by the Court, we will see a freezing of those practices and they will not evolve to the stage of being the very best. The other point is generally transparency, which I think is something the DOJ has been very outspoken on. In terms of access to file, there have been a lot of practical improvements that are very noticeable over the last year or two, down to things like an electronic index with hyperlinks to the document which saves you, in the very short time frame that you have to respond to a Statement of Objections, an enormous amount of time. I also think things like supplementary access to the file, really up until the point that the Commission needs to get close to taking its decision, are very helpful. I’m not sure we need to go to the process that the Bundeskartellamt and perhaps other national authorities have of really giving rolling access to the file. I think we all know, after having looked at these files, that 95% of it is pretty irrelevant and we would prefer in the first phase that the case team be fully focused on resolving the key issues of the case. Nevertheless, I think there is one thing that I’ve never really understood, which is that, in theory, if there are key documents in a case, you get access to them under the best practices the day after the opening of Phase II. But why do you get access to them only after the Commission has taken that decision to go into Phase II, which we all know has very significant consequences for the parties and the timing of the deal? If there is really a key document, whether it is a separate complaint or a request for information which has a 30-page response to the last question – “Do you have any concerns about this transaction?” – why not give access to that to the parties immediately? Certainly, in one case that I’ve had recently, I think that getting that information outside the framework of the best practices was extremely helpful for resolving a lot of technical issues in that case. In that vein, Rachel, I guess that the DOJ doesn’t have any best practices in terms of access to the file, because there is no file in that sense. However, I wonder whether giving access to key documents at a stage before the DOJ actually goes to court must also help the cause of transparency. Nicholas Forwood: So far as the legal status of best practice statements is concerned, there are obviously two ways that that could come before us. First of all, in the context of a plea of legitimate expectations, perhaps its limits are not quite 21

Case T-310/01, Schneider Electric SA v Commission [2002] ECR II-4071.

338  Merger Control in European and Global Perspective as broad as you might think they were, at least in relation to the sort of things that can be covered by it. What is becoming more important of course is the principle of good administration, which has now got support in the Charter of Fundamental Rights.22 Now, a person is entitled to have the administrative authorities deal with his case properly and effectively and so on. Clearly, statements as to best practice, which we have to remember is not necessarily exactly the same as good practice – best may be ideal, but good may be good enough – are clearly going to be relevant in that connection. One particular area where it may be relevant is in relation to the control of complex economic assessments in areas where there is, in some form or another, a marginal control. There, of course, as the Court reminds us in Tetra Laval,23 you verify the basic facts and so on. You also look to see, in the words of the General Confession of Sins, whether you have left undone those things that you ought to have done, or whether you have done those things you ought not to have done. In that connection, I can well see that if there is a statement in a best practice document that one ought to do X, Y, or Z, but that hasn’t been done in a particular case, then that may be the sort of thing that might be thrown in to undermine a complex assessment, to the extent that we would not otherwise be looking at it closely and forming our own view if indeed that be the case. Kirsten Edwards: In terms of submission of economic evidence, I really welcomed the best practice guidelines, but I think a lot of it is actually down to communication between the individual case team and the parties and their advisors. I’ve had very different experiences even since the best practice guidelines have been published in this respect. My best experience, which I think really demonstrated best practice on the side of the Commission and, not to be too modest, on our side too, was in Kraft/Cadbury,24 where we did a vast amount of work in Phase I. This was one of these cases where there was a merger simulation in Phase I, but the only reason we could do that much in Phase I was partly because we spent so much effort trying to pre-empt all of the questions that the case team and the Chief Economist’s Team would come up with, because we didn’t have time to go through a lengthy process of questioning. The case team was also very up front with us in that case, at every stage, on exactly what they had seen so far, what their views were at that time, and what more they needed to see to enable them to decide one way or another about it. Because of that, we did manage to get to a Phase I clearance with divestments in a way that everybody was happy with. 22 The right to good administration is provided for in Article 41 of the Charter, 2010, OJ C83/02. Article 41 provides, in its first three paragraphs, that “1. Every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions, bodies, offices and agencies of the Union. 2. This right includes: (a) the right of every person to be heard, before any individual measure which would affect him or her adversely is taken; (b) the right of every person to have access to his or her file, while respecting the legitimate interests of confidentiality and of professional and business secrecy; (c) the obligation of the administration to give reasons for its decisions. 3. Every person has the right to have the Union make good any damage caused by its institutions or by its servants in the performance of their duties, in accordance with the general principles common to the laws of the Member States.” 23 Case 13/03 P, Commission v Tetra Laval [2005] ECR I-1113. 24 M.5644, Kraft Foods/Cadbury (6 January 2010).

Panel 5 

339  

I think everyone was really proactive in that case, but I’ve been on cases where on our side we’ve just had no idea what the Commission was thinking at a particular point and I’m sure the Commission was thinking: what are these crazy economists going to put in front of them next? (laughter) So I think a large part of it is about communication, but I also think one of the challenges is actually talking to the parties themselves about best practice, because I think most people around this table and in the profession are in a lot of agreement about what best practice should be, particularly as regards economic evidence. However, when it comes to talking to your client and saying that they need to be a bit more here or there, until it comes out of the Commission’s mouth they are not willing to do anything. They are always willing to put their money where their mouth is and do as you advise, but often they think: “Why should we do this, why isn’t the Commission doing this, why is this on us?” So I think there is still a challenge in trying to get the companies who are being investigated to understand where we are coming from with the best practice. Lowe: A pre-condition for that is that the Commission has to know where it is coming from when it actually receives the original pre-notification, and starts to develop a theory or potential theories of harm and has an investigative technique which isn’t a fishing expedition. The opposite to that is when it has absolutely no theory of harm, thinks there is no problem, then it gets a key document which looks brilliant. Then it needs a non-confidential version to be able to share it with everyone else, and it’s two days before the deadline, and we can’t meet what Mario said is the right approach, which is to share key documents which are fundamental to opening Phase II. But Thomas is going to tell us about his own practical experiences. Thomas Deisenhofer: First of all, one general comment on due process, which is taken extremely seriously by Commissioner Almunia and Alexander Italianer.25 One counter-intuitive thing is that you think you can’t get enough of due process. The reality is that, depending on where you stand, there are trade-offs. There is first of all due process and speed, and effectiveness. Then there are tradeoffs between the due process of the one and the due process of the other, which is the most important thing. Confidentiality involves tradeoffs on all sides too. All these points that you have made are extremely delicate points where the difficulty is to find the right balance and to calibrate it correctly, so that you provide as much due process as you can while maintaining speed and while maintaining the rights of third parties. Depending on where you stand as a lawyer, obviously you will defend more this side than the other side, but due process is not an unlimited resource. The second thing is how seriously we take it in practice. It’s not just what is written in the rule that’s important, but what we do. I think Sven mentioned 25 See, e.g., Joaquín Almunia, “Fair process in EU competition enforcement”, SPEECH/11/396 of 30 May 2011; Alexander Italianer, “Safeguarding due process in antitrust proceedings”, Fordham speech, 23 September 2010.

340  Merger Control in European and Global Perspective key documents; it’s true that there is currently no rule that key documents before Phase I can be given, but as Philip said, on the one hand the case team has in reality 15 days from notification during which a lot of time is spent waiting for the replies to questions we try to send out the day the notification arrives. We send them out so that the companies have a bit of time to reply, then we try to coordinate inside. So there might be a document coming up, and then the internal coordination, and then on day 15 we have to speak to the parties. I think it’s right to tell the parties where we’re going, but if on top of that you then have to produce non-confidential versions of key documents, with the difficulties of getting them from third parties, and then manage possible remedies or whatever, that would be a challenge if it were to become mandatory. What is happening in practice, as you’ve rightly described, is that if it’s possible, and if the complainant has no problem with it, then directly or indirectly you can promptly identify relevant technical issues. Routinely, at least at my level, we try to give non-confidential versions of documents even in Phase I in order to sort out things because the ultimate objective is to get to the issues as soon as possible. My last point is on attitude. It’s all about attitude: attitude of the enforcer, and of the parties and third parties. If there is a give and take, and if people are reasonable, then solutions are normally found beyond the written best practices. Bill Kovacic: I have a response to Sven’s question posed to Rachel, and she can certainly address it herself as well. I think there’s a basic tradeoff inherent in the US system, that is that parties like the fact that we can’t do anything that hurts until we’re in front of a judge and a judge says so. That inhibits access to the file, and I think if we had the capacity to take the decision in the first instance it would be much more forthcoming. The tradeoff in the system design is that as long as we have an adversarial contest to resolve the matter, there is always going to be an inhibition to putting all of the evidentiary cards face up. If we had a process that resembled that of the Commission, if the original Hart-Scott vision had been in place, or if the FTC for example could take the steps that the European Commission could take, I imagine that we would have a similar bundle of disclosure safeguards ex ante, but the possibility of litigation makes it hard to be as revealing as we might be. Thank you. Rachel Brandenburger: All I would add is that, from my experience in private practice counselling transatlantically, although the procedures were different and one spent quite a lot of time explaining to clients why it worked this way and that way, I’m not aware of cases in front of either US agency where I felt that I did not understand what the theory of harm was, or what the evidence was in front of the US agencies, even if I did not have access to the file. If you have other experiences, then we need to hear about them. Howard Shelanski: You may remember the proposals that Charles James made in 2001–2002, where he said that the DOJ was willing to put things on a different

Panel 5 

341  

path if it had an earlier common revelation of information, especially in instances where a consensus might be achievable. I don’t know exactly what the experience with that turned out to be, but it was a commitment to put more information forward earlier. I hear continuing complaints as well, Sven, about whether, even without access to the file, we are forthcoming enough. I don’t know whether that’s because we don’t say enough on our side at the FTC, or whether we say it quite clearly but the parties don’t want to believe it. Cal Goldman: Just one or two quick additions. When I was running the Canadian Competition Bureau I consulted with my US colleagues about issuing news releases even when we took no-challenge decisions. I was cautioned that if we did that, then we would end up being sued. Well, we’ve come a long way. The Canadian and the US agencies now issue all kinds of information. That’s a best practice. Secondly, the US authorities, especially the FTC, should be commended for stakeholder consultations on an international scale. I suggest that what the FTC has done on global hearings, consulting the global community and not just stakeholders in the US, is a very good example of trying to achieve as much input through a form of best practice as is possible. It’s a model for other jurisdictions. Thirdly, I’d like to make a small point concerning the OECD. Peer review at the OECD does not generally include BIAC. I think it should. It would be constructive to consider including BIAC at least in the country reviews up until the point where there is any discussion of confidential enforcement actions. I don’t see any harm, I can see only good coming of it. Thank you. Lowe: Thank you very much. I think we’ll draw a line under that discussion of best practices. I just want to say one or two things in conclusion of this day and a half. In June of last year, I took an executive decision that we should discuss mergers. Not because I thought that there would be a huge amount of disagreement on many issues, but because it was an opportunity to discuss the new guidelines envisaged in the US and the UK. I think we did have a very stimulating discussion on the use of methodologies, including market definition issues and other investigative techniques. I didn’t notice a great deal of alarm, but certainly some caution about using those techniques in the right context, and recognizing too that virtually all our techniques need to complement each other to arrive at a robust result. On the standards of assessment and evidential requirements before the EU Courts, I think we’re still going to debate that quite a lot. At the moment I don’t see at least the Commission in Europe, or the EU Courts, actually moving on any particular issue of structure or procedure. Maybe that will become necessary if indeed we have another spate of Court activity with respect to certain cases, and there are some in the pipeline. On the procedural issues of convergence particularly, I thought that we would not expect to have a great deal of advance compared with the work already done in the ICN context and elsewhere, but I did see, in both yesterday’s session and this morning, continuing emphasis on evaluating performance and

342  Merger Control in European and Global Perspective indeed confronting ourselves with either the criticism of over-enforcement or under-enforcement. What that would mean in terms of ongoing work inside our agencies almost certainly needs clarification. We emphasized how important it was to build up a relationship of cooperation and confidence between those people who are actually going to tackle the same issues in the same markets. I did raise a small alarm bell, which I think will become larger, that the more our markets become global, the more pure competition will be mixed up, at least in the public perception and in the government’s perception, with other issues such resource scarcity, security of supply, and general concerns about concentration. These may take any agency, if put under pressure, beyond what it would regard as the normal standards of proof necessary to prohibit a merger. I’m just saying that this is an alarm bell, and it may be less loud if we are able to look more closely at the longer-term effects of further concentration in global markets. It comes back to the issue of what the strength of potential competition is outside any particular jurisdiction, and what the serious assumptions being made by one jurisdiction are in taking a decision in relation to the effects in its own jurisdiction. There, thank goodness that there is the Bundeskartellamt to look at BHP Billiton. With the Commission, obviously. But, without some constructive cooperation between the agencies in relation to transactions like that, I think that there would be a serious issue of government failure of addressing major transactions, which have maximum impact in practically all jurisdictions. On the last issues of best practice, we can only get better. Looking ahead, Mel and I will be thinking about next year’s subject, and we’ll undoubtedly choose a more controversial one where we can scratch each other’s eyes out on procedural issues or inter-institutional issues. I thank you very much, particularly those who have come a long way to be with us to share in this discussion. As I say, the Workshop hasn’t been as bloody as some of the previous ones. Last year, when we discussed due process there were far greater, far more powerful epithets used than anything in the last 36 hours, but those lively discussions did lead us to interesting conclusions.26 We can say that there’s a lot of progress being made, but there are some alarm bells ringing and we shouldn’t be complacent. I thank you again for your participation, and as always we will try to set an agenda next year to match the standards set by Claus Ehlermann and all of you in times gone by. Thank you (applause).

26

See Ehlermann and Marquis, Evaluation of Evidence and its Judicial Review, cited above note 19.

Tomaso Duso, Klaus Gugler and Florian Szücs1

Merger Policy Evaluation: Where Do We Stand?

1. Introduction The empirical evaluation of merger policy has received much attention from economists and practitioners during the past decade. Even though the literature on the topic is substantial and continues to grow, no canonical framework of evaluation has emerged and no incontestable policy recommendations have been produced. The former point is in part responsible for the latter: while a plurality of aspects is desirable when looking for the “correct” framework of evaluation for merger policy, it also reduces comparability between studies. In this survey of the existing literature on the empirical evaluation of merger policy, we classify studies according to four different dimensions, depending on the main focus of their investigation. These four categories correspond to different aspects of merger policy, in which different notions of efficiency are relevant. We hope that this classification will help to develop a more general framework for future studies and to delimit the areas of competence of the existing literature. The four dimensions of merger policy effectiveness we propose are: (i) legal certainty and determinants of intervention; (ii) decision errors; (iii) effectiveness of remedies; and (iv) deterrence. We believe that a comprehensive evaluation of merger policy must not neglect any of these areas. In the following, we try to classify the existing literature on merger policy evaluation along those lines.

2. Dimensions of the effectiveness of merger policy We propose a comprehensive and intuitive framework in which the efficiency of merger control is evaluated along four dimensions linked to the chronological order of a merger investigation: prior to a merger notification, legal certainty and predictability of jurisdiction allow the firms to anticipate the authority’s likely reaction and thereby reduce the uncertainty involved; during the investigation, the frequency of decision errors indicates whether the competition authority intervenes in anticompetitive cases and clears pro-competitive mergers, while the 1 Respectively: Humboldt University, Berlin; Vienna University of Economics and Business; University of Vienna.

106  Merger Control in European and Global Perspective degree of effectiveness of remedies captures their impact on markets and is meant to measure their success in preventing anticompetitive rents that would otherwise be generated by an undesirable merger. Finally, we investigate whether merger decisions have a deterrent effect on the notification of anticompetitive mergers.

2.1. Legal certainty and determinants of intervention Legal certainty and predictability of the proceedings are desirable qualities in merger control (e.g., Smith (1957) and Elman (1965)). In particular, a predictable merger con­trol process reduces the high costs and reputation losses entailed by a lengthy procedure (Neven, Nuttall, and Seabright (1993)) and it increases the credibility of the authorities by encouraging them to base their decisions on accurate facts and sound eco­ nomic reasoning. Since transparency of the proceedings reduces the welfare-detrimental risk of political influence, legal certainty is intrinsically connected to the issue of the determi­nants of intervention: if merger policy is predictable, we should be able to find significant determinants of the competition authority’s decisions. There are numerous articles conducting such analyses. For example, Coate and McChesney (1992) find that the FTC responds to political pressure from Congress, while Khemani and Shapiro (1993) find no evidence of political influence in Canadian merger control. In a more recent article, Bergman, Jakobsson, and Razo (2005) investigate merger cases scrutinized by the European Commission and report that there is no evidence of political motives in its decisions. Aktas, De Bodt, and Roll (2007), on the other hand, find that the Commission fosters European firms by protecting them from the competition of non-EU rivals. In Duso, Gugler, and Szücs (2010), the determinants of the Commission’s interventions are as­sessed with regard to legal certainty. While reporting both economic and political factors influencing the Commission’s decision, we also find elements of path dependency in the decision process: a high ratio of interventions in past periods increases the probability of an intervention in the current period. With regard to legal certainty we find that, in re­cent years, the prediction of unconditional clearance has become more difficult. Parties are thus less able to foresee whether a combination will be approved unchallenged. This suggests that legal certainty has been diminished, e.g., by the reform of merger policy in 2004.

2.2. Decision errors A merger control authority with the express goal of maximizing consumer welfare can commit two kinds of errors in its jurisdiction: it can prohibit a pro-competitive merger (which is called a type I error) or unconditionally clear an anticompetitive

Tomaso Duso, Klaus Gugler and Florian Szücs 

107  

one (a type II error).2 In the first case, consumers do not enjoy the price reductions an efficiency ­increasing merger would have brought about. In the second case, consumer welfare is reduced by the increase in prices or decrease in quality following the increase in market power. Under standard assumptions of oligopoly theory, rivals (i.e., firms in the same product market) will benefit from an anticompetitive merger and, vice versa, they will suffer from a pro-competitive one. This identifying assumption along with event studies3 of the reaction of rivals’ stock prices in a time window around the announcement of a merger case, which is interpreted as the market’s assessment of the transaction, can be used to classify subsamples of pro- and anticompetitive mergers. Duso, Neven, and Röller (2007) report evidence on type I and type II errors in the Commission’s decisions and identify institutional and political factors that drive them. While market definitions and industry effects do play a role, the Commission does not seem to be influenced by lobbying on the part of the firms. We employ the same methodology in Duso, Gugler, and Szücs (2010) and find that, in recent years, the Commission committed more type I errors but less type II errors. Additionally, we identify structural (market definitions and industries) as well as political factors (the political influence of big EU countries and the US) that influence the decision. We also find some evidence of successful lobbying by the merging firms.

2.3. Effectiveness of remedies If the policy tools employed by the merger authority are efficient, we would expect them to have a significant impact on the market, which is reflected in the stock price of the concerned parties.4 The expected profits from an anticompetitive merger will be priced in the stock prices of the merging parties and their rivals well before the merger is actually consummated. An efficient antitrust intervention should thus – as long as it is not perfectly anticipated – reduce these expected gains and thus induce a drop in stock prices. Furthermore, we would expect a relation between the size of the abnormal returns around the merger announcement and the abnormal returns due to the merger remedy. The standard tools for analysis of this kind are event studies. 2 Since data on blockings are sparse, weak type I errors (merger remedies in pro-competitive mergers) and strong type I errors (prohibition of a pro-competitive merger) are usually distinguished. 3 The relevant methodology is introduced in Fama, Fisher, Jensen and Roll (1969); Ellert (1976); Eckbo (1983); Stillman (1983). McAfee (1988) offers a critique of the method. 4 Given efficient markets, the actions of the authority will – even under uncertainty – be anticipated by market participants. An approach to deal with these expectations is described in Duso, Gugler, and Yurtoglu (2010). In Duso, Gugler, and Szücs (2010) we confirm the result concerning prohibitions and additionally find that unconditional clearances are a positive signal for the competitors of the merging firms, indicating possible type II errors.

108  Merger Control in European and Global Perspective Brady et al. (2000) look at the effect of the regime change due to the introduction of the European Merger Regulation in late 1989. While the observed regime effects are ambiguous and mostly insignificant, they find significant effects for firms that underwent a merger review in that period. In particular, where a merger raised “serious doubts” on the part of the Commission, this reflected negatively on the stock value of the firms. Similarly, Aktas, De Bodt, and Roll (2004) analyze the stock-market reactions of merging firms under scrutiny by the Commission to various events in the merger review procedure. They find that, while a clearance in the first stage of investigation is good news even if it involves conditions and obligations, the initiation of an in-depth investigation has a significantly negative effect. At the end of 2005 the Commission’s Directorate General for Competition published an in-house study on merger remedies. It reviews the design and implementation of 85 different remedies adopted in 40 decisions adopted by the Commission between 1996 and 2000. The analysis is done by means of inter­views with the committing parties or sellers, licensors and grantors, the purchasers or buyers, licensees and grantees and the trustees. The main results are that only 57% of the analyzed remedies were classified as clearly effective, 24% as partially effective, 12% as unclear and 7% as clearly ineffective. In another study based on interviews, Baer et al. (1999) conducts an ex post review of 37 divestitures ordered by the FTC between 1990 and 1994. They find that in 28 out of the 37 cases viable competitors were created, but also that merger remedies were not particularly effective in solving the anticompetitive problems. A framework to relate abnormal returns at the announcement of a merger to those around the competition authority’s decision is developed in Duso, Gugler, and Yurtoglu (2010). They find that only prohibitions achieve a full reversal of the anticompetitive rents generated at the announcement of a merger. Approvals subject to remedies are only partially effective in achieving rent reversion. Furthermore, remedies seem to be most efficient after the first investigation phase, while their impact is diminished after an in-depth investigation.

2.4. Deterrence Sørgard (2009) points out that an optimal merger control regime involves the deterrence of anticompetitive mergers and acquisitions: a tough and effective merger control policy should inhibit firms from proposing anticompetitive mergers in the first place, and should encourage them to look for alternative strategies that are beneficial or neutral from a competition perspective. Only a few studies have analyzed the impact of “tough” decisions on notifications in suc­cessive periods. Seldeslachts, Clougherty, and Barros (2009), by using a panel of antitrust jurisdictions over the period 1992–2003, find that prohibitions deter future merger activ­ity while remedies do not. They focus on the frequency

Tomaso Duso, Klaus Gugler and Florian Szücs 

109  

aspect but cannot say whether merger policy over-deterred pro-competitive mergers. The question of the composition of mergers which are deterred by the policy is addressed in a follow-up paper, Clougherty and Seldeslachts (2010), analyzing US merger policy in the period 1986–1999. They find that secondrequest investigations and, even more so, agency interventions yield significant deterrence effects for horizontal mergers, i.e., they negatively affect the ratio be­ tween horizontal to total notified mergers. They conclude that a tougher merger policy makes firms move away from potentially problematic horizontal mergers toward vertical mergers that are more likely to be efficiency-increasing. In Duso, Gugler, and Szücs (2010), we use the stock-market reaction of the merg­ ing firms’ rivals to distinguish pro- and anticompetitive mergers. Using data on EU merger cases between 1990 and 2007, we show that prohibitions significantly reduce the number of anticompetitive merger notifications in subsequent periods. Abandoned transactions and withdrawals of previously notified mergers (in particular after an in-depth investigation) entail significant deterrence effects which remedies do not have.

3. Some Statistics on EU Merger Cases The table below summarizes the main characteristics of EU merger cases since the beginning of EU merger control until the end of 2007. We divide the cases into a pre-reform period (1990–April 2004) and a post-reform period (May 2004–2007). Pre-reform Mean

S.D.

Post-reform Mean

S.D.

Remedies in Phase 1

0.069

(0.05)

0.055

(0.04)

in Phase 2

0.041

(0.04)

0.044

(0.03)

0.028

(0.03)

0.011

(0.01)

Cleared Prohibited

0.931

(0.05)

0.945

(0.04)

Phased 2 cases

0.008

(0.01)

0.001

(0.00)

Abort/Withdraw

0.055

(0.23)

0.026

(0.16)

0.033

(0.03)

0.024

(0.01)

“Remedies” reports the fraction of cases that were cleared subject to conditions and obligations. “Cleared” is the fraction of cases that were permitted without conditions or obligations, Phase 2 cases were subjected to an in-depth investigation. “Abort/Withdraw” denotes transactions that were notified but withdrawn during the review procedure.

The evolution of merger notifications and the Commission’s actions from 1990 to 2007 is graphically represented in the figure below. Three observations stand out. First, and not surprisingly, there is an ongoing increase in merger notifications

110  Merger Control in European and Global Perspective over time. Second, outright prohibitions are an extremely rare event prior to merger policy reform in 2004 (0.8%) and almost non-existent afterwards (there were only two mergers prohibited since the reform, or 0.1%). Third, while the relative activity of the Commission (ratio of remedies, ratio of cases going to phase 2) diminishes, remedies in phase 1 become more prominent (from 4.1% to 4.4%). This implies that the use of remedies in phase 2 diminishes considerably (from 2.8% to 1.1% of all mergers).

We report the merger notifications per year as well as the ratio of different outcomes (remedies in phase 1 and phase 2, prohibitions, aborted transactions) to the notifications in the population of all EU merger cases.

4. Conclusions and recommendations Surveying the literature on the empirical evaluation of merger policy, one discovers a multitude of different approaches and findings. These differences are in large part due to constraints on the availability of data, which in turn dictate different empirical strategies. Since confidential parts of economic analyses by competition authorities are in general unavailable to researchers, proxies have to be used for key decision variables. Furthermore, in­dividual studies are usually limited to a certain timeframe of observations within a single jurisdiction. As a result of this heterogeneity of method and focus, the comparability and consistency of findings in this area are modest. This situation could be improved by establishing a more general framework of evalua­tion of competition policy, along with “canonical” empirical strategies to

Tomaso Duso, Klaus Gugler and Florian Szücs  111   address partic­ular questions within that framework. We believe that most studies in the field of the empirical evaluation of merger policy can be categorized as one or more of four important aspects of this task – legal certainty and determinants of intervention, decision errors, effectiveness of remedies and deterrence – and that such a categorization might help to increase the comparability of results. In this chapter we have attempted such a categorization, and our main findings can be sum­marized as follows. First, in Duso, Gugler, and Szücs (2010) we find that legal certainty in European merger control has decreased in recent years. In particular, it has become more difficult to predict whether the Commission will clear a merger unconditionally. Second, with regard to decision errors, Duso, Neven, and Röller (2007) find that institutional factors – in particular, market definitions and whether an in-depth investi­gation was conducted or not – play a role in determining type I and type II errors by the Commission, but that its decisions are not influenced by lobbying on the part of the firms. While we confirm those findings in Duso, Gugler, and Szücs (2010), we also find some evidence of successful lobbying. Third, Duso, Gugler, and Yurtoglu (2010) find that only outright prohibitions achieve a full reversal of anticompetitive rents. We confirm this finding in Duso, Gugler, and Szücs (2010) and find only weak support for the effectiveness of other remedies. However, in recent years the Commission has almost discontinued the use of prohibitions. It appears that the Commission blocks too few mergers. Finally, significant deterrence effects can be observed. While Seldeslachts, Clougherty, and Barros (2009) and Duso, Gugler, and Szücs (2010) find that prohibitions decrease the number of total and anticompetitive notifications respectively, Clougherty and Seldeslachts (2010) find that second requests and interventions by the FTC deter horizontal mergers.

References Aktas, Nihat, Eric De Bodt and Richard Roll (2004): “Market response to European regula­tion of business combinations”, 39 Journal of Financial and Quantitative Analysis 731–757. Aktas, Nihat, Eric De Bodt and Richard Roll (2007): “Is European M&A Regulation Protec­tionist?”, 117 The Economic Journal 1096–1121. Baer, William et al. (1999): “A Study of the Commission’s Divestiture Process”, http://www.ftc.gov/os/1999/08/divestiture.pdf. Bergman, Mats, Maria Jakobsson and Carlos Razo (2005): “An econometric analysis of the European Commission’s merger decisions”, 23 International Journal of Industrial Organization 717–737. Brady, Una and Robert Feinberg (2000): “An examination of stock-price effects of EU merger control policy”, 18 International Journal of Industrial Organization 885–900.

112  Merger Control in European and Global Perspective Clougherty, Joseph and Jo Seldeslachts (2010): “Deterrence of Horizontal Mergers: Empirical Evidence from U.S. Industries”, mimeo. Coate, Malcolm and Fred McChesney (1992): “Empirical evidence on FTC enforcement of the merger guidelines”, 30 Economic Inquiry 277–293. Duso, Tomaso, Klaus Gugler and Florian Szücs (2010): “An Empirical Assessment of the 2004 EU Merger Policy Reform”, working paper. Duso, Tomaso, Klaus Gugler and Burcin Yurtoglu (2010): “How Effective is European Merger Control?”, mimeo. Duso, Tomaso, Damien Neven and Lars-Hendrik Röller (2007): “The political economy of European merger control: evidence using stock market data”, 50 The Journal of Law and Economics 455–781. Eckbo, B. Espen (1983): “Horizontal mergers, collusion, and stockholder wealth”, 11 Journal of Financial Economics 241–273. Ellert, James (1976): “Mergers, antitrust law enforcement and stockholder returns”, 31 Jour­nal of Finance 715–732. Elman, Philip (1965): “The Need for Certainty and Predictability in the Application of the Merger Law”, 40 New York University Law Review 613–627. Fama, Eugene, Lawrence Fisher, Michael Jensen and Richard Roll (1969): “The adjustment of stock and prices to new information”, 10 International Economic Review 1–21. Khemani, Shyam and Daniel Shapiro (1993): “An empirical analysis of Canadian merger policy”, 41 The Journal of Industrial Economics 161–177. McAfee, Preston (1988): “Can event studies detect anticompetitive mergers?”, 28 Economics Letters 199–203. Neven, Damien, Robin Nuttall and Paul Seabright (1993): Merger in Daylight: The Eco­nomics and Politics of European Merger Control, Centre for Economic Policy Research. Seldeslachts, Jo, Joseph Clougherty and Pedro Pita Barros (2009): “Settle for Now but Block for Tomorrow: The Deterrence Effects of Merger Policy Tools”, 52 The Journal of Law and Economics 607–634. Smith, Blackwell (1957): “Precedent, Public Policy and Predictability”, 46 Georgetown Law Jour­nal 633–645. Sørgard, Lars (2009): “Optimal Merger Policy: Enforcement vs. Deterrence”, 57 Journal of Industrial Economics 438–456. Stillman, Robert (1983): “Examining antitrust policy towards horizontal mergers”, 11 Journal of Financial Economics 225–240.

James S. Venit*

The Scope of EU Judicial Review of Commission Merger Decisions

Overview This chapter’s conclusions may be summarized as follows: 1. The scope of the EU Courts' judicial review of Commission merger (and other antitrust) decisions is specified (and limited) by the Treaty. The EU Courts have interpreted this limitation as restricting their review to the legality of the Commission's decision. In effect this gives the EU Courts the ability to review: (i) substantive law; (ii) procedural issues; (iii) the accuracy and completeness of the facts;1 and (iv) the soundness of the conclusions the Commission has drawn from the facts. If the Commission has erred in any of these areas, its decision may be annulled. By way of shorthand, ensuring the legality of a Commission decision amounts to ensuring that there has been no manifest error in the decision. as concerns the four areas identified above. 2. Where complex economic assessments are involved, the Commission is, in addition, entitled to a certain deference, provided its conclusions are plausible and there has been no manifest error. In practice this means that the EU Courts cannot substitute an alternative and equally plausible conclusion for the one the Commission has reached as long as the facts relied on by the Commission are accurate and complete, the conclusions the Commission has drawn from them are plausible, the law has been correctly applied and no material .procedural errors committed. 3. Points 1 and 2 do not, however, mean that the Courts cannot exercise their full powers of review in determining: (i) the accuracy and completeness of facts relied on by the Commission; (ii) whether substantive law has been correctly applied; and (iii) whether there have been material procedural errors. The ability to review fully the Commission's factual findings – which in effect involves an assessment as to whether the Commission has met its burden of proof – creates the potential scope for far reaching review, and at least in the merger area, the General Court has at times conducted such extensive reviews. * Skadden, Arps, Slate, Meager, & Flom, Brussels. 1 Since review by the Court of Justice is limited to questions of law, the requirement that the Court not substitute its assessment of the facts for that of the Commission applies in practice to the General Court.

114  Merger Control in European and Global Perspective 4. The requirement that the Court must defer to the Commission in cases involving complex economic assessments can raise delicate issues as concerns the Court's review of whether the Commission has omitted any key facts and, to an even greater degree, whether its conclusions are sufficiently supported by the facts. In contrast to ascertaining whether the facts relied on by the Commission are correct, such assessments, by their nature, require the Court to make judgments that may call into question and/or depart from the Commission's conclusions. 5. In such cases, the principle according to which the Court may not substitute its judgment for that of the Commission suggests that the Court can only challenge the completeness of the factual record where there has been an obvious and material omission and that it can reject the conclusions the Commission has drawn from the facts (assuming they are accurate and complete) only where those conclusions are not plausible. Conversely, if the Commission's conclusions are plausible, the Court is not entitled to reject them merely because there are one or more contrary and equally plausible alternative conclusions. 6. In practice, it is not always easy to tell whether the General Court2 has adhered to these constraints in its review of merger decisions. The General Court’s judgment in Tetra Laval was particularly controversial in that regard but was ultimately upheld by the Court of Justice. This suggests that judicial review of mergers can be quite robust, notwithstanding the deference owed to the Commission in respect of complex economic assessments. 7. Sony/BMG has established the counter-intuitive proposition that there is no judicial bias that favours decisions to permit, as opposed to prohibit, a merger. That is, that the Commission will be held to the same standard of review in both positive and negative decisions. On its face, Sony/BMG would at first appear difficult to reconcile with the Court’s approach to mergers allegedly giving rise to conglomerate (and arguably coordinated) effects. In such cases, the Court has indicated that the Commission’s prohibition decisions must be particularly convincing. Distinguishing between the standard of proof or judicial review and the Commission’s burden of proof provides a coherent way of reconciling the tension between Sony/BMG, on the one hand, and cases like Airtours and Tetra Laval, on the other. Under this analysis, whereas the judicial standard of review remains the same for positive and negative merger decisions, in certain cases – such as conglomerate mergers or mergers allegedly giving rise to coordinated effects – the quality of the evidence relied on by the Commission (and thus in effect its burden of proof) will be particularly high when it seeks to prohibit such a merger.

2 In the discussion that follows, what was then the Court of First Instance is referred to by its current name – the General Court.

James S. Venit 

115  

Discussion I. The Scope of Judicial Review in Merger Cases The Treaty on the Functioning of the European Union (TFEU) provides for a different scope of judicial review depending on whether the EU Courts are reviewing the substance and procedure of a Commission decision or whether they are reviewing the amount of a fine. According to Article 263(2) TFEU (ex-Article 230(2) EC), the Court "shall for this purpose have jurisdiction in actions brought by a Member State, the European Parliament, the Council or the Commission on grounds of lack of competence, infringement of an essential procedural requirement, infringement of the Treaties or of any rule of law relating to their application, or misuse of powers." Thus, as concerns Commission decisions other than the part relating to fines, the Court's jurisdiction is limited by the Treaty to jurisdictional and procedural issues and infringement of the Treaty or any related rule of law and misuse of powers. Or to put it another way, the Court can only assess the legality of the decision under what has come to be known as the “manifest error” test.3 The scope of review pursuant to Article 263(2) TFEU contrasts significantly with the unfettered, original jurisdiction that the Court enjoys in respect of the amount of fines imposed by the Commission. Here, according to Article 261 TFEU (ex Art. 229 EC), “Regulations adopted jointly by the European Parliament and the Council, and by the Council, pursuant to the provisions of the Treaties, may give the Court of Justice of the European Union unlimited jurisdiction with regard to the penalties provided for in such regulations.”4 As a result, the Court has the power to decrease or increase any fine or periodic penalty payment imposed by the Commission and thus, in effect, to substitute its determination of the appropriate fine for that of the Commission. Fines or periodic penalty payments are relatively rare in merger cases, and are never linked to the substantive assessment of the merger. Thus, the contrast between the unlimited scope of the Courts’ jurisdiction over fines and its more limited jurisdiction when reviewing the substance and procedural correctness of a Commission merger decision is primarily relevant because it highlights the limits imposed by the Treaty on the Court’s ability to review such decisions.

3 See Bo Vesterdorf; “Judicial Review in EC Commission Law: Reflections on the Role of the Community Courts in the EC System of Competition Law Enforcement,” 1 Global Competition Policy 1, 9 (2005). 4 In the area of merger control, the Court’s unlimited jurisdiction with respect to fines has been implemented in Article 16 of Regulation 139/2004 (the “EU Merger Regulation”).

116  Merger Control in European and Global Perspective The limits imposed on the Court’s review of merger cases5 was comprehensively articulated in the Ryanair case,6 in which the Court stated that: “the basic provisions of the [Merger] regulation […] confer on the Commission a certain discretion especially with respect to assessments of an economic nature, and […] consequently, review by the [EU] Courts […] must take account of the margin of discretion implicit in the provisions of an economic nature which form part of the rules on concentrations […] Whilst the Courts of the European Union recognise that the Commission has a margin of discretion with regard to economic matters, that does not mean that they must refrain from reviewing the Commission’s interpretation of information of an economic nature. Not only must they establish, in particular, whether the evidence relied on is factually accurate, reliable and consistent but also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it […] In addition…where the institutions have a power of appraisal, respect for the rights guaranteed by the legal order of the European Union in administrative procedures is of even more fundamental importance. Those guarantees include, in particular, the duty of the Commission to examine carefully and impartially all the relevant aspects of the individual case, the right of the person concerned to make his views known and also his right to have an adequately reasoned decision […]”7

Subject to an important qualification that will be discussed in Section III, Ryanair endorses a fairly robust level of judicial review as concerns the correct application of the law, respect for procedural requirements, the accuracy and completeness of the facts (including facts of an economic nature) and the soundness of the conclusions drawn by the Commission from the facts. Examples of the practical application of this review will be discussed in Section II. That discussion will show that, insofar as errors of law, fact and procedure are concerned, the Court’s review of the legality of a decision is not limited by any need to show deference to the Commission’s assessment. Moreover, as will be seen from that discussion and the discussion in Section III, the deference the Courts are deemed to owe the Commission in cases involving complex economic assessments does not mean that the Courts will accept conclusions that are not adequately supported by accurate, reliable, complete and consistent evidence. 5 The question of whether the scope of the EU Court’s review is broader in non-merger cases is beyond the remit of this chapter. In the AstraZeneca judgment, involving the application of Article 102, the General Court stated that “it follows from settled case-law that, although as a general rule the Community judicature undertakes a comprehensive review of the questions as to whether or nor the conditions for the application of the competition rules have been met, the review of complex economic appraisals made by the Commission is necessarily limited to checking whether the relevant rules on procedure and on stating reasons have been complied with, whether the facts have been accurately stated and whether there has been any manifest error of assessment or a misuse of powers. Case T-321/05 Astra Zeneca v. Commission [2010] ECR II-2805, para. 32. AstraZeneca has asked the Court of Justice to set aside the judgment of the General Court (which confirmed the Commission’s infringement decision) or to reduce the amount of the fine levied by the Commission. See Case C-457/10 P, not yet decided. 6 Case T-342/07, Ryanair v. Commission [2010] ECR II-3457. 7 Ibid., paras. 29–31.

James S. Venit 

117  

II. Judicial Review With Respect to Procedural Irregularities and Questions of Law It appears beyond dispute that, subject to review by the Court of Justice, the General Court is entitled to undertake a full judicial review with respect to questions of law, as concerns both procedural irregularities and errors in the application of the substantive law. As will be discussed below in Section III, full review also extends to issues concerning the accuracy and completeness of the facts relied on by the Commission.8 The scope of the Court’s review of questions of substantive and procedural law has been robustly summarized by Judge Vesterdorf: “As regards matters of law, the Community courts exercise full jurisdictional control. Indeed it is for the Community courts to provide the definitive interpretation of Community law, be it in Treaty provisions or secondary legal provisions such as those contained in the Merger Regulation, and this goes for both procedural and substantive legal provisions. The Community courts interpret the law and then check whether the Commission has applied the correct legal principles in the case under examination. There is no margin of appreciation left to the Commission in this respect as to what are the legal criteria to apply.”9

Errors of substantive law With respect to errors of substantive law, the General Court has carried out a full judicial review in various instances. Examples include: (a) MCI v. Commission,10 in which the General Court annulled a Commission decision which prohibited a notified transaction despite the fact that the parties indicated to the Commission that they were abandoning the merger and had withdrawn their notification. The court concluded that the Commission “exceeded the limits of its powers” under the EU Merger Regulation, and “infringed the legitimate expectation of the notifying parties.”11 (b) Tetra Laval v. Commission,12 in which the General Court considered that the Commission was wrong to categorically dismiss behavioral commitments offered by the parties to remedy conglomerate effects issues.13 8 The scope of the Court’s review of the accuracy and completeness of the facts relied on by the Commission has been most clearly articulated in the cases dealing with complex economic assessments and, for that reason, these issues will be discussed in Section III. 9 Bo Vesterdorf, “Standard of Proof in Merger Cases: Reflections in the Light of Recent Case Law of the Community Courts”, 1 European Competition Journal 3, 12 (2005). 10 Case T-310/00, MCI v. Commission [2004] ECR II-3253. 11 Ibid., paras. 107 and 113. 12 Case T-5/02, Tetra Laval v. Commission [2002] ECR II-4381. 13 Ibid., paragraph 161. See also Case C-12/03 P Commission v. Tetra Laval [2005] ECR I-987, paras. 85 et seq., in which the Court of Justice upheld the General Court’s finding in this respect. Based on the Court’s criticism in Tetra Laval, the Commission’s revised Remedies Notice of 2008 now states that “[c]ommitments relating to the future behaviour of the merged entity may be acceptable only exceptionally in very specific circumstances”. Notice, 2008 OJ C267/1, para. 17.

118  Merger Control in European and Global Perspective (c) Airtours v. Commission,14 in which the General Court established what amounts to a legal standard setting out the necessary conditions for a finding of collective dominance (in today’s parlance, “coordinated effects”), which the Commission subsequently incorporated into its 2004 Guidelines on the assessment of non-horizontal mergers (at paragraph 41).

Procedural irregularities In the area of procedural irregularities, certain issues – such as whether the defendant was given full access to the file or an opportunity to be heard at an oral hearing – would appear to raise relatively straightforward questions of law. However, other areas can be less clear in the sense that the ruling on the point of law may depend on an assessment of the facts which, in theory, can lead the General Court to substitute its own assessment for that of the Commission. An instructive example of the latter occurred in Schneider Electric v. Commission.15 Here, the General Court annulled the Commission’s decision for breach of the applicant’s rights of defense resulting from what the court determined to be the impreciseness of the Commission’s allegations in the Statement of Objections. Subjecting the content of the Statement of Objections to what was effectively a de novo review, the court held that “[i]t is not apparent on reading the statement of objections that it dealt with sufficient clarity or precision with the strengthening of Schneider’s position vis-à-vis French distributors of low-voltage electrical equipment as a result not only of the addition of Legrand’s sales on the markets for switchboard components and panel-board components but also of Legrand’s leading position in the segments for ultraterminal electrical equipment. The Court observes in particular that the general conclusion in the statement of objections lists the various national sectoral markets affected by the concentration, without demonstrating that the position of one of the notifying parties on a given product market would in any way buttress the position of the other party on another sectoral market.”16

The General Court concluded that, as a consequence, the Statement of Objections did not permit Schneider Electric to assess the full extent of the competition problems to which the Commission claimed the concentration would give rise at the distribution level, and thus the applicant was not put in a position to submit in good time proposals for divestiture capable of resolving all the competition problems identified by the Commission.17 In the follow-up damages case brought by Schneider against the Commission, the General Court acknowledged its full power of judicial review with respect to procedural irregularities, noting that the Commission’s margin of discretion arising from the difficulty inherent in undertaking a complex market analysis Case T-342/99, Airtours v. Commission [2002] ECR II-2585. Case T-310/01, Schneider Electric v. Commission [2002] ECR II-4071. 16 Ibid., para. 445. 17 Ibid., paras. 453 et seq. 14 15

James S. Venit 

119  

under a very rigid time constraint is “irrelevant” to the court’s review of procedural errors committed by the Commission.18 However, while the EU Courts enjoy full judicial review with respect to procedural irregularities, a procedural defect will only lead to the annulment of the decision if the violation is sufficiently substantial and could have caused a change in the final decision.19

III.

Judicial review of complex economic assessments

It is well established that, within the parameters of Article 263(2) TFEU, the EU Courts are required to show deference to the Commission where complex economic assessments of the type normally made in merger decisions are involved.20 What Constitutes a Complex Economic Assessment? There is room for discussion as to what constitutes a complex economic assessment. As will be seen from the cases below, decisions to approve or prohibit a merger at some level involve a complex (and as we shall see, speculative) economic assessment, which is why the language articulating the Court’s need to defer to such assessments regularly appears in EU Court judgments reviewing merger decisions. What is less clear in the abstract is whether the review of the various inputs during the merger review process, such as, e.g., studies relating to market definition, the forecasting of market trends, or econometric analyses of the merger’s projected impact, that inform the ultimate decision to approve or prohibit, also involve complex economic assessments or whether these inputs constitute merely economic facts to be assessed like any other factual evidence. At least one Judge of the General Court has taken the view that economic ‘facts’ and evidence, however complex, should not be accorded deference simply because of their economic nature and technical difficulty. According to Judge Nicholas Forwood, there is a need to distinguish between “complex” and “non-complex” economic appraisals reflecting a difference between ‘complex’ Case T-351/03, Schneider Electric v. Commission [2007] ECR II-2237, para. 155. See, e.g., Case T-209/01, Honeywell v. Commission [2005] ECR II-5527, para. 126: “The alleged infringement of the rights of the defence in this case relates exclusively to the aspects of the contested decision which the applicant has challenged in other respects in its other pleas in law, namely bundling and cross-subsidisation. Even if the present plea were well founded, it could therefore undermine only the pillars of the Commission’s reasoning against which those other pleas are also directed. The present plea cannot therefore have any effect on the other pillars constituting the foundation of the contested decision.” See also Case T-290/94, Kaysersberg v. Commission [1997] II-2137, para. 88: “It is settled law that the failure to comply with that rule can render the Commission’s final decision unlawful only if it is sufficiently substantial and it had a harmful effect on the legal and factual situation of the party alleging a procedural irregularity” (denied for the Commission’s failure to comply with the period of notice for convening the Advisory Committee). 20 There is at least one case, Woodpulp, which pre-dates the creation of the General Court, in which the Court of Justice employed its own economic expert to review the soundness of the Commission’s conclusions. That review led the Court of Justice to annul the Commission decision. See Joined Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85 and C-125/85 to C-129/85 A. Ahlström Osakeyhtiö and others v. Commission [1993] ECR I-1307. 18 19

120  Merger Control in European and Global Perspective economic appraisals” and merely “’difficult’ economic appraisals.”21 As Judge Forwood puts it, “[t]he mere fact that an assessment is made which requires the consideration of (possibly esoteric) economic arguments and the examination of economic evidence […] does not necessarily make an assessment subject to judicial review a ‘complex’ one, subject only to limited control, even though it may make the task of the judge extremely difficult or burdensome.”22

This distinction has led Judge Forwood to suggest that “it could be said that what the Court of Justice had in mind, at least in the first thirty years or so of its case law, was that ‘complexity’ refers more to the nature of assessment that needs to be made, rather than its technical or evidential difficulty.”23

This distinction – or at least the General Court’s ability to review complex economic facts without deferring to the Commission’s margin of discretion in respect of complex economic assessments – has been endorsed by the Courts in Tetra/Laval and most recently in Ryanair, where the General Court observed: “Whilst the Courts of the European Union recognise that the Commission has a margin of discretion with regard to economic matters, that does not mean that they must refrain from reviewing the Commission’s interpretation of information of an economic nature.”24

In the context of merger decisions, the distinction identified by Judge Forwood would suggest that questions ranging from market definition to quantitative evidence assessing the projected impact of a horizontal merger would all be classified as difficult factual/economic issues that may be subjected to thorough review, but that the ultimate decision to prohibit or permit a merger involves more of a value judgment and thus warrants judicial deference. That the complex decision to permit or prohibit a merger may be based on the sum of the merely difficult, technical economic appraisals that result in the decision being taken one way or the other does not invalidate the coherence of this analysis. What is, however, somewhat more problematic is the impact of the inherently speculative and predictive nature of merger analysis. Whereas the latter could be cited as supporting the case for greater deference, it would seem that, if anything, the inherently speculative nature of merger analysis – at least where conglomerate and/or coordinated effects are involved – has led the EU Courts to exercise even 21 See Nicholas Forwood, “The Commission’s ‘More Economic Approach’ – Implications for the Role of the EU Courts, the Treatment of Economic Evidence and the Scope of Judicial Review”, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2009: The Evaluation of Evidence and its Judicial Review in Competition Cases, Hart Publishing, 2011, pp. 255 et seq. 22 Ibid., at pp. 264–265. 23 Ibid., at p. 267. Applying his analysis, Judge Forwood would, for example, classify assessments concerning the granting of an exemption under Article 101(3), which require “value judgments” as a complex economic assessment; whereas assessments as to the applicability of Article 101(1) or whether an undertaking holds a dominant position would involve a difficult economic appraisal which may, nevertheless, be subjected to comprehensive judicial review. See ibid. at pp. 267–268. 24 Cited above note 6, at para. 30.

James S. Venit 

121  

greater vigilance when reviewing contested merger decisions by imposing a higher burden of proof on the Commission.

Judicial Deference in Cases Involving Complex Economic Assessments The legal position concerning the scope of the Court’s review of complex economic assessments is itself clear at least in its formulation and was summarized by the General Court in Ryanair. The relevant passage, already quoted earlier, bears repeating: “In addition, the Court of Justice has held that the basic provisions of the regulation, in particular Article 2, confer on the Commission a certain discretion, especially with respect to assessments of an economic nature, and that, consequently, review by the Courts of the European Union of the exercise of that discretion, which is essential for defining the rules on concentrations, must take account of the margin of discretion implicit in the provisions of an economic nature which form part of the rules on concentrations ….”25

The consequences of this deference were then spelled out in paragraphs 30 and 31 of the Ryanair judgment (also quoted earlier): “Whilst the Courts of the European Union recognise that the Commission has a margin of discretion with regard to economic matters, that does not mean that they must refrain from reviewing the Commission’s interpretation of information of an economic nature. Not only must they establish, in particular, whether the evidence relied on is factually accurate, reliable and consistent but also whether that evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it […] In addition…where the institutions have a power of appraisal, respect for the rights guaranteed by the legal order of the European Union in administrative procedures is of even more fundamental importance. Those guarantees include, in particular, the duty of the Commission to examine carefully and impartially all the relevant aspects of the individual case, the right of the person concerned to make his views known and also his right to have an adequately reasoned decision […]”

Whereas articulation of the legal position appears relatively straightforward, its implementation in practice is a somewhat more complex story. The discussion that follows will focus on the tension between the Court’s mandate to review thoroughly the accuracy and completeness of the facts and the coherence of the conclusions drawn from them by the Commission in merger decisions and the case law requirement that the Courts defer to the Commission where complex economic assessment are involved. In the Sony/BMG appeal, Advocate General Kokott described the scope of the Court’s review with respect to the accuracy, reliability and completeness of the facts and the ability of the cited facts to support the conclusions drawn from them quite broadly. According to the Advocate General, in such a review, the Court 25

Ibid., para. 29.

122  Merger Control in European and Global Perspective will, when applying the manifest error standard, undertake its own analysis of the facts and the evidence:26 “Having regard to this standard of review, it would be an error to assume that the Commission’s margin of discretion precludes the Community Courts in any event from giving their own analysis of the facts and the evidence. On the contrary, it is essential for the Community Courts to undertake such an assessment of their own where they are assessing whether the factual material on which the Commission’s decision was based was accurate, reliable, consistent and complete, and whether this factual material was capable of substantiating the conclusions the Commission drew from it. Otherwise, the Community Courts could not sensibly assess whether the Commission had stayed within the limits of the margin of discretion allowed to it or had committed a manifest error of assessment.”

This conclusion was adopted by the Court of Justice, which stated in the judgment that “it is clear from the case-law that while, in the context of the control of concentrations, a field in which the Commission has a margin of assessment with regard to economic matters, review by the Court of First Instance is limited to establishing whether the evidence relied on is factually accurate and to establishing the absence of a manifest error of assessment, it none the less remains the case that the correctness, completeness and reliability of the facts on which a decision is based may be the subject of judicial review […] Indeed, that is one of the ways in which the Community judicature can verify whether the factual and legal elements upon which the exercise of the power of assessment depends were present […]”27

However, the Advocate General did go on to note that such rigorous review does not entitle the Court to substitute its judgment for that of the Commission: “The Court of First Instance exceeds the limits of judicial review of a Commission decision in the context of merger control only where the factual and evidential position reasonably allows different assessments, the Commission adopts one of them, and the Court of First Instance none the less substitutes its own different assessment for that of the Commission.”28 26 Opinion of Advocate General Kokott in Case C-413/06 P, Bertelsmann and Sony Corporation v. Impala [2008] ECR I-4951, at para. 239. See also Bo Vesterdorf, “The European Courts’ Case Law in Merger Control”, in Abel Mateus and Teresa Moreira, eds., Competition Law and Economics: Advances in Competition Policy and Antitrust Enforcement, Kluwer Law International, 2007, chapter 13, at p. 246: “Control of primary facts by the CFI is intensive with no room for discretion on the part of the Commission. This is inherent in the nature of the control of the accuracy of facts. Either a fact is a correct fact or it is not.” 27 Case C-413/06 P, Bertelsmann and Sony Corporation v. Impala [2008] ECR I-4951, para. 69. 28 Cited above note 26, at para. 240. The different standard of judicial review with respect to the accuracy of facts, on the one hand, and their assessment by the Commission, in particular with respect to complex economic matters, on the other, was also acknowledged by Advocate General Tizzano: “With regard to the findings of fact, the review is clearly more intense, in that the issue is to verify objectively and materially the accuracy of certain facts and the correctness of the conclusions drawn in order to establish whether certain known facts make it possible to prove the existence of other facts to be ascertained. By contrast, with regard to the complex economic assessments made by the Commission, review by the Community judicature is necessarily more limited, since the latter has to respect the broad discretion inherent in that kind of assessment and may not substitute its own point of view for that of the body which is institutionally responsible for making those assessments.” Opinion of AG Tizzano in Case C-12/03 P, Commission v. Tetra Laval [2005] ECR I-987, para. 86.

James S. Venit 

123  

In theory, the distinction between the Court’s rigorous assessment of the correctness, completeness and reliability of the facts and the conclusions drawn from them, on the one hand, and the stipulation that it must not substitute its judgment for that of the Commission, on the other hand, is coherent in its articulation and can be neatly encapsulated in the notion that the Court can only annul where the Commission has committed a manifest error of assessment by drawing conclusions that are simply not plausible in light of all the relevant evidence. However, it is not selfevident how the Court can easily draw the line between thoroughly reviewing whether the facts are sufficient to support the conclusions the Commission has drawn from them, on the one hand, while at the same time respecting the obligation not to substitute its judgment for that of the Commission on the other. Respecting this distinction is likely to present serious problems in practice.29 This is because any determination as to whether the Commission’s conclusions are based on all the relevant facts and are sustainable in their light will invariably require the exercise of independent judgment. For example, the conclusion that a relevant piece of evidence has been excluded involves a judgment not only of the value and importance of that particular piece of evidence in the context of the overall evidentiary record, but also an implicit criticism of the conclusions that have been drawn by the Commission because of its neglect of that evidence. The degree of independence becomes even greater as concerns the assessment as to whether the Commission’s conclusions are sustainable in light of the evidence. Review of the case law suggests that the degree of difficulty involved in conducting such a review without second guessing the Commission may vary substantially from case to case. It further suggests that the Court has chosen the burden of proof as the judicial tool to assist it in navigating between the Scylla of rigorous factual review and the Charybdis of deference to the Commission’s complex economic assessments. Airtours v. Commission30 presents a good example of a relatively straightforward assessment, and indeed, one which seems to verge on pure factual error. In Airtours, the General Court fully re-examined the Commission’s projection of low market growth for package tours which, according to the Commission, made the relevant market conducive to coordinated effects. The Commission’s conclusion was based on a single page extract of an industry study which, the Court found, the Commission had misinterpreted by failing to take account of the actual wording of the document which clearly indicated considerable market growth in the past decades. Against this background, the Court concluded that 29 As the Court of Justice held in Case C-441/07 P, Alrosa v. Commission [2010] ECR I-5949, para. 63: “The General Court could have held that the Commission had committed a manifest error of assessment only if it had found that the Commission’s conclusion was obviously unfounded, having regard to the facts established by it.” In Alrosa (which was not a merger case but a commitments case under Article 9 of Regulation 1/2003), the General Court had objected to the remedy imposed, concluding that a less invasive alternative existed. In reversing the General Court’s judgment annulling that aspect of the Commission’s decision, the Court of Justice drew a line between the appropriateness of the remedy – which was a matter to be left to the discretion of the Commission, and a manifest error of assessment – e.g., the need for a remedy – which could fall within the competence of the General Court. 30 Cited above note 14.

124  Merger Control in European and Global Perspective “the Commission’s interpretation of the data available to it concerning growth demand was inaccurate in its disregard for the fact that the market had been marked by a clear tendency to considerable growth over the last decade in general, despite the volatile nature of demand from one year to another, and that the pace of demand growth has increased during recent years in particular. In that context of growth, and having failed to produce any more specific evidence establishing that the tendency to grow would be reversed in future years, the Commission was not entitled to conclude that market development was characterised by low growth, which was, in this instance, a factor conducive to the creation of a collective dominant position by the three remaining large tour operators.”31

Tetra Laval v. Commission32 provides perhaps the most striking example of a considerably more difficult case in which maintaining the distinction identified by Advocate General Kokott and endorsed by the Court of Justice appears particularly challenging. In Tetra/Laval, the General Court overturned the Commission’s assessment of the conglomerate effects of a merger which had been based on the Commission’s assessment of the likelihood of leveraging from aseptic carton markets, and of the consequences of such leveraging, on the market for PET packaging. The court based its conclusion on a careful examination of whether the evidence relied on by the Commission supported the decision’s findings in relation to complex factors such as (i) whether the likely levels of growth in the PET market would provide an incentive for the merging parties to leverage, and (ii) the foreseeable consequences of leveraging on the various product markets.33 On the basis of a detailed factual analysis, the court concluded that, “as regards the markets for asceptic and non-asceptic PET filling machines, the contested decision does not adduce evidence that suffices in law to show that the implementation of the modified merger would make it likely that, as a result of leveraging practices essentially on Tetra’s current carton customers who wish to switch to PET for packaging sensitive products, a dominant position on those markets would be created in the future and at least by 2005.”34

Three conclusions can be drawn from this example. First, the level of the Court’s review can be intense and, indeed, controversial, particularly when it involves the assessment of multiple factual elements underlying a single determination. Second, this intensive review can lead the Court to reject the Commission’s conclusions. Third, in doing so the Court relies on the Commission’s failure to meet its burden of proof as the judicial tool of choice to avoid overstepping the line by substituting its judgment for that of the Commission. As a legal matter, reliance on the Commission's failure to meet its burden of proof is arguably uncontroversial since such an assessment would appear to be well within the scope of the General Court's powers of review. Moreover, to the extent that such an assessment is based on the General Court's evaluation of the factual record, it also has the added benefit of insulating that assessment from Ibid., para. 133. Cited above note 12. 33 Ibid., paras. 142–251. 34 Ibid., para. 251. 31 32

James S. Venit 

125  

subsequent judicial scrutiny since the Court of Justice is limited to reviewing questions of law, but not of fact in cases brought before it on appeal. That being said, the conclusion that there is insufficient factual support for the Commission's conclusions on a key element in a merger decision involves a judgment that is far more complex than one concerning the accuracy or the completeness of the facts. There is therefore little wonder that the General Court's judgment in Tetra Laval provoked an appeal by the Commission to the Court of Justice. And, although the latter did uphold the General Court’s judgment, it did so in part by refusing to reconsider issues of fact35 and in disregard of the Advocate General’s conclusion that the General Court had exceeded its power by substituting its assessment for that of the Commission.36

IV. The impact of the speculative nature of merger control on judicial review The particularly speculative nature of merger assessment has clearly played a role in the Courts’ approach to judicial review. Thus, as the Court of Justice noted in Tetra Laval: “A prospective analysis of the kind necessary in merger control must be carried out with great care since it does not entail the examination of past events – for which often many items of evidence are available which make it possible to understand the causes – or of current events, but rather a prediction of events which are more or less likely to occur in future if a decision prohibiting the planned concentration or laying down the conditions for it is not adopted.”37

This consideration suggests, and the cases appear to support this conclusion, that: (i) the Court’s review may be considerably more independent and aggressive and the Commission’s reasoning and conclusions will be subject to intense scrutiny given the predictive and inevitably speculative nature of merger analysis; and (ii) the intensity of the Court’s analysis will be proportional to the speculative nature of the Commission’s theory of harm, with the intensity of judicial scrutiny being 35 See Case C-12/03 P, Commission v. Tetra Laval [2005] ECR I-987, para. 47: “Amongst the other examples given by it, the Commission challenges the Court of First Instance’s finding, in paragraph 289 of the judgment under appeal, that ‘fresh milk is not a product for which the marketing advantages offered by PET have any particular importance’ and its conclusions as to the cost of PET in comparison to that of carton, which are set out in paragraphs 288 and 328 of the judgment under appeal. It should be noted that these are findings of fact, which are not subject to review by the Court in appeal proceedings. It is therefore unnecessary to give a ruling on the merits of those findings by the Court of First Instance and it need be stated only that the Court of First Instance was able to base those findings on various items in the contested decision.” 36 See Opinion of Advocate General Tizzano in Case C-12/03 P, Commission v. Tetra Laval [2005] ECR I-987, para. 111: “As matters stand, it seems clear to me that, while the Court of First Instance could possibly have found that the Commission’s investigation was incomplete, or have criticised the logic, consistency and appropriateness of its reasoning, it definitely could not carry out its own assessment of the information in that institution’s possession in order to conclude that PET was ‘more expensive than carton’”. 37 Cited above note 13, at para. 42.

126  Merger Control in European and Global Perspective greater in cases involving conglomerate (and arguably coordinated) effects, as opposed to more straightforward cases of unilateral horizontal effects. The General Court has confirmed as much by indicating that, where anticompetitive harm is not the normally expected outcome or is speculative in nature, the Commission must base its decision on particularly convincing evidence, i.e., its burden of proof is considerably heightened. In the words of the court: “The Commission’s analysis of a merger producing a conglomerate effect is conditioned by requirements similar to those defined by the Court with regard to the creation of a collective dominant position […] Thus the Commission’s analysis of a merger transaction which is expected to have an anti-competitive conglomerate effect calls for a particularly close examination […] As the Court has already held, where the Commission takes the view that a merger should be prohibited because it will create or strengthen a dominant position within a foreseeable period, it is incumbent upon it to produce convincing evidence thereof […] Since the effects of a conglomerate-type merger are generally considered to be neutral, or even beneficial, for competition in the markets concerned, as is recognized in the present case […] the proof of anti-competitive conglomerate effects of such a merger calls for a precise examination, supported byconvincing evidence, of the circumstances which allegedly product those effects […]”38

This in turn raises the question of whether the heightened burden of proof which requires that the Commission put forth particularly convincing evidence in a case involving a conglomerate merger, or a more complex theory of harm like coordinated effects as in Airtours, is coupled by a stricter standard of judicial review, or whether the latter remains the same in all cases.39 This inquiry inevitably triggers another question, i.e., whether the counter-intuitive conclusion reached by the Court of Justice in Sony/BMG – to the effect that it is as hard to approve as to prohibit a merger – can be accepted at face value. On its face, Sony/BMG appears to have rejected the notion that more is required to prohibit than to permit mergers, including non-horizontal mergers, or mergers allegedly giving rise to coordinated effects. According to the Court of Justice, as a matter of principle, “there is nothing in Article 2(2) or (3) of the Regulation which states that it imposes different standards of proof in relation to decisions approving a concentration, on the one hand, and decisions prohibiting a concentration, on the other.”40

According to the Court, this overriding principle is not affected by the fact that Commission inaction would lead to the approval of a concentration (Sony/BMG at paragraph 49) nor by the speculative nature of the assessment the Commission is required to make in cases involving conglomerate mergers or, arguably, alleged coordinated effects. Cited above note 12, at para. 155. Note that, in Tetra Laval, the Court found the evidence adduced by the Commission to be insufficient in law to support the Commission’s theory of harm, but the Court did not clarify what the relevant legal standard was. See the text accompanying note 34 above. 40 Sony/BMG, cited above note 27, at para. 46. 38 39

James S. Venit 

127  

“Furthermore, it is true that, as is apparent from the Court’s case law, the decisions of the Commission as to the compatibility of concentrations with the common market must be supported by a sufficiently cogent and consistent body of evidence […] and that in the context of the analysis of a ‘conglomerate-type’ concentration the quality of the evidence produced by the Commission in order to establish that it is necessary to adopt a decision declaring the concentration incompatible with the common market is particularly important. However, it cannot be deduced from that that the Commission must, particularly where it pursues a theory of collective dominance, comply with a higher standard of proof in relation to decisions prohibiting concentrations than in relation to decisions approving them. That case-law merely reflects the essential function of evidence, which is to establish convincingly the merits of an argument or, as in the case of the control of concentrations, to support the conclusions underpinning the Commission’s decisions […] Furthermore, the fact that an issue of collective dominance does, or does not, arise, cannot of itself have an impact on the standard of proof which applies. In that regard, the inherent complexity of a theory of competitive harm put forward in relation to a notified concentration is a factor which must be taken into account when assessing the plausibility of the various consequences such a concentration may have, in order to identify those which are most likely to arise, but such complexity does not, of itself, have an impact on the standard of proof which is required. It follows that, where it has been notified of a proposed concentration pursuant to the Regulation, the Commission is, in principle, required to adopt a position, either in the sense of approving or of prohibiting the concentration, in accordance with its assessment of the economic outcome attributable to the concentration which is most likely to ensue.”41

The Court’s formulation equates the complexity of the theory of harm relied on by the Commission in coordinated effects or conglomerate cases with the requirement of a higher (the Court says “important”) quality of evidence but also states that this requirement does not impact the standard of proof. This suggests that, whereas the Court will apply the same scope and standard of review when assessing the plausibility of the Commission’s decision, the Commission’s burden of proof in such cases will be greater given the greater complexity or dubiousness of its theory of harm. As the Court of Justice stated in Tetra Laval:42 “The analysis of a ‘conglomerate-type’ concentration is a prospective analysis in which, first, the consideration of a lengthy period of time in the future and, secondly, the leveraging necessary to give rise to a significant impediment to effective competition mean that the chains of cause and effect are dimly discernible, uncertain and difficult to establish. That being so, the quality of the evidence produced by the Commission in order to establish that it is necessary to adopt a decision declaring the concentration incompatible with the common market is particularly important, since that evidence must support the Commission’s conclusion that, if such a decision were not adopted, the economic development envisaged by it would be plausible.”

41 42

Ibid., paras. 50–52. Cited above note 13, at para. 44.

128  Merger Control in European and Global Perspective In other words, the complexity and/or speculative nature of a theory of competitive harm put forward in relation to a conglomerate merger (or one allegedly giving rise to coordinated effects) is a factor which must be taken into account when assessing the plausibility of the Commission’s conclusions as to the consequences such a concentration may have. This complexity does not, of itself, have an impact on the standard of proof or the scope of judicial review which will, according to Sony/BMG, remain the same irrespective of whether the Commission is seeking to approve or prohibit the merger. In either case, the Commission’s decision will not be accepted if the facts relied on are incorrect or incomplete or the conclusions drawn by the Commission from the facts are not plausible (and vice versa). But the Court of Justice’s determination that the quality of evidence relied on by the Commission in cases involving conglomerate mergers or alleged coordinated effects is “particularly important” does strongly suggest, in the same way that greater effort is required to swim upstream than downstream, that the Commission’s burden of proof (and as regards the Court’s expectations as to the quality of the evidence on which the Commission seeks to rely) is likely to be particularly high in cases in which the Commission seeks to prohibit a merger involving conglomerate or, arguably, coordinated effects.

Conclusion The EU Courts’ formulation of the standard of review in merger cases appears to be relatively well-settled and clear – at least in its articulation. First, a Commission decision can only be annulled for errors of law, material factual or procedural errors, or where the Commission has omitted key facts or where its conclusions are manifestly not supported by the facts. The Commission is not entitled to any deference as concerns procedural issues, the correct application of the law or the accuracy and completeness of the facts on which it relies. However, given the complex nature of the assessment involved in permitting or prohibiting a merger, it is entitled to a measure of deference as concerns its assessment of the facts and evidence. In effect, this means that, in order for the Commission’s merger decision to be upheld, its conclusions must be plausible but need not be the only or arguably even the most plausible conclusion. In practice, the application of this legal standard poses particular problems with respect to whether the facts relied on by the Commission are complete, and, to an even greater degree, whether the Commission’s conclusions are sufficiently supported by the evidence as to be plausible. In theory, the EU Courts cannot substitute their judgment for that of the Commission on these matters, provided the Commission’s conclusions meet the plausibility test. However, in practice, the Courts’ assessment of the plausibility of the Commission’s conclusions would appear to allow considerable scope for second-guessing the Commission. To some extent the General Court can navigate

James S. Venit 

129  

around the constraints imposed by the deference requirement and can also avoid judicial review by the Court of Justice by basing its annulment decisions on the grounds that the Commission’s conclusions are not sufficiently supported by the evidence, which is another way of saying that the Commission has failed to meet its burden of proof under the appropriate standard of review. Second, the Courts’ standard of proof (or review) will be the same irrespective of whether the decision in question is to permit or prohibit a merger. However, it would also appear, at least in the case of non-horizontal mergers or mergers allegedly giving rise to coordinated effects, that the quality of the evidence relied on by the Commission, and thus, in effect, its burden of proof, will be higher where it seeks to prohibit such a merger. Whether the scope of EU judicial review is appropriate in an administrative law system in which the Commission enjoys the powers of both prosecutor and judge, and what precisely constitutes a complex economic assessment, are questions that will no doubt continue to be debated alongside the question of whether the General Court has been too timid or too aggressive in exercising the powers of review granted to it under the Treaty. The first question, as Theodore Fontane once said in a rather different context is, of course, ein weites Feld.43

43

Effi Briest, Theodore Fontane (1896).

V. Merger control and best practices

5 PANEL DISCUSSION

Chair: Philip Lowe

Speakers: John Boyce Rachel Brandenburger Jochen Burrichter Thomas Deisenhofer Kirsten Edwards

Cal Goldman William Kovacic Howard Shelanski Mario Siragusa Sven Völcker



328  Merger Control in European and Global Perspective

PANEL 5: Merger control and best practices Philip Lowe: This final session is about best practices. What is best, and whose practices they are, is going to be revealed to us by our three speakers. First of all, John. John Boyce: Well, I must confess that when I was asked to focus on best practices in merger control I was a little bit uncertain about what I was meant to be addressing. Was I supposed to be commenting on those guidelines that some competition agencies issue, which are labelled “best practices”? Or was I being given a bit more flexibility, to talk about what I think is the best thing competition authorities should be doing and how we, as practitioners, should do our jobs? The first thing I did, actually, was to go to the dictionary to find out what “best practices” means. I was hoping to find some sort of Shakespearian quote or something showing that it was first used by Henry II (laughter). But to my surprise, the oldest reference in the Oxford English dictionary to the use of the word “best practices” was only in 1984. That was perhaps in the days when accountants were more innovative than they now are, because it pre-dated the concentration in that industry (more laughter). According to the OED, in understanding accounting in 1984, that was the first time that the concept was used. It goes on to talk about another use of it in 1993 by John Kay, who some of you will know as an economist who wrote about corporate success.1 He explained that, if you want to run a successful corporation, one thing you might do, obviously, is to look at what other firms are doing and copy it. This strategy, he said, is more felicitously expressed as “adopting best practice”. So that’s not a bad idea. Just see what other people are doing, copy it, and that’s best practice. Well, things have moved on a little bit since then. One dictionary definition which I did quite like is from businessdictionary.com, which describes “best practices” as “methods and techniques that have consistently shown results superior than those achieved with other means, and which are used as benchmarks to strive for”. It did, however, go on to say that there is “no best practice that is best for everyone (laughter), or in every situation, and no best practice remains best for very long, as people keep on finding better ways of doing things.” Then, if you look at one of my favourite dictionary sources, Wikipedia, it’s very modern, very democratic. It comes up with similar concepts and it actually picks up some of the terminology that Bill was using earlier on. Wikipedia suggests that best practice is to be measured by its effectiveness: getting the best results. Best practice should also be about getting things done in the most efficient way, with the least amount of effort. Now, that all sounds very sensible to me, but Wikipedia actually goes on to say that there is a danger of best practices becoming a bit of a business buzz word. It says that, “as the term has become more popular, some more 1 See John Kay, Foundations of Corporate Success: How Business Strategies Add Value, OUP, 1993/1995. For additional information on Kay, see http://www.johnkay.com/about.

Panel 5 

329  

organisations have begun using the term ‘best practices’ to refer to what are in fact merely rules, causing a linguistic drift in which a new term such as ‘good ideas’ is needed to refer to what would previously have been called ‘best practices’”. I think that really sets the scene for what I think best practices should really be about. It’s all about learning from experience. That’s from your own experience, but also by looking at what other people are doing and by learning from their experiences. They might have done something wrong, so you say that’s what we want to avoid doing. It’s all about learning by doing and learning by watching. If you look at the whole area of international merger control that we’ve been talking about for the last couple of days, it’s a prime area where best practice makes sense. Increasingly, regulators are applying similar rules, and increasingly they clear similar deals, or even the same deal, so we can try to learn from what others are doing. In my paper, I try to focus on three areas where best practices make sense. The first area is best practices by agencies and by lawyers, facilitating the process of advising clients about how to get deals through. We’ve been talking about alliteration, well some time ago as a junior lawyer I learned the 6 P’s: proper preparation prevents piss poor performance (laughter). Now, increasingly that is a problem when you’re advising on international deals; where should you be notifying? I think firms are doing this regularly; you’ve got your databases, you’ve got the guides that people go to. Even on the most mundane deals, you may have to notify in a number of jurisdictions. I was just talking to Nick during the break about how, when I was a junior lawyer, the bane of our lives was being asked: “Can you review this agreement to see whether it is caught by the Restrictive Trade Practices Act in the UK?” Now, I’d have to do this and I’d think that this isn’t what I wanted to do as a competition lawyer. I wanted to deal with effects and problems. So, I think that as a legal community we were all quite pleased to see that piece of arcane legislation repealed, but what’s come along to replace that for our junior lawyers now and for the junior case handlers of the European Commission is dealing with cases that raise no competition issues whatsoever. Now, this is where I think there are some best practices out there. The ICN has been fantastic at trying to point people in the right direction by coming up with common techniques for cases, that is, both cases that shouldn’t raise problems and those that do. One of the key buzzwords for me in this area is transparency. There can be clarity as to what rules should be about, and if agencies are transparent about how their rules operate in practice that makes life much easier for us in practice. If I’ve got a couple of suggestions, my first one picks up on the point that the ICN is not a group of legislators, it’s just agencies so they don’t make any binding rules. However, I think there is still some scope for advocacy by the agencies, just making it clearer to their governments that the rules don’t make sense. It sounds as if John is having the problem the other way around at the moment, in that he’s being pushed in the direction of rules that could be more bureaucratic. I’ve talked a bit about the issue of rules, which are triggered purely by jurisdictional thresholds. I just wonder, raising the same point again, whether

330  Merger Control in European and Global Perspective there is scope for regulators – I used the word “regulator” there actually! – to exercise some self-restraint on whether to enforce the rules. Thomas, you said that you don’t have the power to do that, but to some extent you do. Let’s say a Korean company and a Japanese company set up a coffee shop in Tanzania. Say they happened to meet the thresholds and didn’t think of notifying the European Union, but someone spotted it when they were reading the newspaper in Tanzania while on holiday from the European Commission. I don’t think you would be using the Commission’s resources effectively by requiring them to file a notification. Fold up the paper and move on! (laughter) Another idea that I’ve got for the ICN is that if you hunt around the ICN’s website, you’ll find some summaries that some jurisdictions have put in, showing what their rules are, what their thresholds are. They’re pretty long, and quite difficult to understand, so I wonder whether there might be some scope for cleaning them up a bit and encouraging members to do that. You can do that behind closed doors without NGAs having to hear about it. Possibly agencies could even self-assess and spell out where the rules are ICN best practice-compliant and where they’re not. There’s nothing wrong with having rules that don’t comply with the ICN’s best practices, but it is useful for us as practitioners to know when rules don’t comply. So those are the areas where I think more can be done. A lot has been done to make life easier for us advising clients at the outset on deals. The other area of best practice is for the conduct of investigations, and we talked a lot about that in the previous session. Best practice is all about using the 3 Cs: convergence, cooperation and comity. Thomas talked about challenges, which is another C word. I like C words when it comes to competition! You were talking about the cultural challenges, the confidentiality issues – another C word that you didn’t even mention was a C word – and the challenges of communication that you have dealing with the other regulators. We heard Rachel’s more complicated MRTD acronym, but I took those letters and just threw them into the Scrabble bag and they came back as 4 Cs: mindfulness is consciousness of what the other regulators are doing; respect, well Nick said it – consideration, or courtesy or care. There you are, that’s three words (laughter). Trust is confidence – have confidence in what your other ICN members are doing; and dialogue is, of course, communication. So, there are plenty of C words out there. The last area I wanted to touch on briefly is remedies, which is an important area… Audience: Commitments! (laughter) Boyce: Yes, commitments, sorry! This is an area where, as a practitioner, it is very important to get the regulators working together because the risk of having divergent commitments can actually increase the antitrust “price” and make deals unattractive. Philip was talking right at the end of the last panel about the airlines sector, and I spent some of the past two years dealing with a transatlantic alliance case.2 I found that quite bizarre in some ways, because you have the 2

M.5747, Iberia/British Airways (14 July 2010).

Panel 5 

331  

sort of odd US practices dealing with cooperation in the airline sector, where the DOJ has got no jurisdiction at all and the Department of Transportation deals with it in an incredibly so-called “transparent” way where everything is on the docket. Also, how can you have cooperation between the Commission and the DOT under those circumstances? But you get around that somehow. The normal US approach, where orders to show cause come up quite quickly, as happened with Star Alliance and Sky Team, didn’t seem to happen in our case for some reason. That was because the Commission was cooperating so well, and not just with the Department of Transportation. I’m complimenting the Commission here, actually! (laughter) There was very close cooperation between the Commission and the DOT and there was also very close cooperation between the Commission and the DOJ. Where there didn’t seem to be close cooperation was between the DOT and the DOJ, which I found quite intriguing. I’ve stayed away from talking about best practices in the sense of guidelines because I think Jochen’s going to say a bit about that. In my paper3 I’ve tried to summarize in an annex the available best practice documents. I also commend to you, if you’re not familiar with them, the ICN’s Recommended Practices;4 I’ve summarized them in a separate annex in my paper. They’re very good and they’ve withstood the test of time, being written some five years ago. I don’t think they need to be updated much, even though one of the recommendations is that practices should be updated from time to time. The European Commission is a bit delinquent in that respect, because the Commission’s best practices for mergers pre-date the 2004 merger regulations,5 so all the references are wrong. I’ve also in an annex put down some recommended practices for lawyers, which I think we generally all try to comply with. We have guidelines like that, and I’m sure you do too, that are there to help our junior lawyers work out how you should be dealing with relatively simple cases, which may need to be notified all over the world and where you need to coordinate with local lawyers as required. To conclude, I think best practice is a great thing to strive for. If we keep on developing best practices and coming up with new, good ideas, then ultimately we will get to Utopia. But you’ve got to keep on trying to get there, because even when you think you are there, you’re not. It’s going to change, so keep on moving forward. Lowe: As school children in the North of England, we were actually taught to recite this little rhyme: “good, better, best. Never let it rest, until the good is better and the better best.” (laughter). Next we have Jochen. Jochen Burrichter: Well, it’s very difficult for me to add to what John has already said. I would like to categorize, or to analyze, the existing best practice rules which are in place. The best practices which you can observe provide guidelines 3 Boyce, “Best Practice in Merger Control: It Ain’t What You Do It’s the Way That You Do It … And That’s What Gets Results”, this volume, pp. 343 et seq. 4 http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 5 http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf.

332  Merger Control in European and Global Perspective on how merger reviews can be conducted, or should be conducted, if issued by international institutions in terms of transparency, case handling, communication, confidentiality and inter-agency coordination. If you try to categorize the existing best-practice rules, you can distinguish between the external best practices, which are issued for the benefit of merger control agencies by international institutions like the ICN, or between agencies. The second category is internal best practices, as I call them. That is, the best practices issued by the agencies themselves for the merging parties and for their own practice. The third category, which John covered at the end, is best practices issued by companies or by law firms. Let me concentrate on the two first categories. The external best practices for agencies are an important means to influence convergence. We have two categories, as I already said: firstly, the international organizations issue best practices. By far the most significant driver with regard to these best practices is the ICN, which seeks to facilitate cooperation and the exchange of know-how between competition authorities globally. The Merger Working Group published a comprehensive set of materials on best practices, which John already recommended that we should study. The most important contributions are the Guiding Principles for Merger Notification and Review6 and the Recommended Practices for Merger Notification and Review Procedures.7 Further important guidance for merger control agencies comes from the OECD. In 2005, the Council of the OECD adopted a Recommendation that touched fairness and cooperation.8 And then another document was issued by the Business and Industry Advisory Committee of the OECD; that committee published their Recommended Framework for Best Practices in International Merger Control Procedures in October 2001.9 The other important category of best practices consists of bilateral best practices, and by far the most important one is the Best Practices on Cooperation in Merger Investigations of the US-EU Merger Working Group.10 If you try to analyze the existing best practices, what is the scope? If you look at the external best practices, you see that, consistent with the rationale of external best practices, i.e., providing guidance for merger control authorities on how to shape their procedures, external best practices only establish general and broad principles for an efficient and transparent set up. These principles are usually verbalized as recommendations to agencies, therefore external best practices provide a model rather than a formal code. So you can conform to them but you are not obliged to do that. Procedural issues that are usually dealt with include recommendations on the structural alignment of notification thresholds with, for the sake of legal certainty and reduction of unnecessary transaction costs, a clear preference for revenue-based thresholds rather than market share http://www.internationalcompetitionnetwork.org/uploads/library/doc591.pdf. http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 8 OECD, Recommendation of the Council on Merger Review (2005), http://www.oecd.org/ dataoecd/3/41/40537528.pdf. In April 2012, the OECD’s Competition Committee published a report (based on the proceedings of Working Party No. 3) entitled Procedural Fairness and Transparency: Key Points. See http://www.oecd.org/dataoecd/16/34/50235955.pdf. 9 http://www.biac.org/statements/comp/BIAC-ICCMergerPaper.pdf. 10 http://www.ftc.gov/os/2011/10/111014eumerger.pdf. 6 7

Panel 5 

333  

thresholds. Moreover, the recommendations are pronounced regarding the agencies’ requirements on the timing of a notification, a clear determination of the triggering event and preferably no deadline, and the content of the notification. If you analyze the internal best practices which exist, frequently they deal with the details regarding the case team set up, the decision-making process, pre-filing guidance, communication issues – for example the timing of the Statement of Objections – guidance on the conduct of meetings, and access of third parties to such meetings. There is also relevant guidance on the right of access to the file, timing, accessible documents, entitled persons, procedures on the requirement for the initial notification, and safeguarding confidentiality with regard to the business secrets provided by the merging parties. The aims and benefits of merger best practices can be summarized very quickly. First of all, with regard to agencies, it is clear that merger control agencies benefit from the external best practices because they assist antitrust agencies with respect to a structuring of the relevant merger review procedures, therefore acting as a catalyst for better convergence. They also help to streamline the intra-agency procedures and therefore boost efficiency with respect to the merger review. The promotion of external best practices is desirable, but I think, as John said, there is still something to do. The increasing proliferation of merger control regimes and the variety of jurisdictions which affect globally active undertakings heighten the need for national agencies to publish more internal best practices on the conduct of their procedure. With regard to external best practices we have the ICN’s Recommended Practices,11 which is an excellent document that still can be improved, as John already explained. If one would try to define the minimum requirements of the internal best practices, I think it is clear that each internal best practice should deal at least with a minimum standardized set of issues which cover the most important aspects of merger control procedures. I think these are: pre-filing consultations, the scope of requests for information, the conduct of meetings, involvement of third parties, access to the file and confidentiality issues. To what extent are these best practices binding? An authority that issues such best practices normally should be bound to what they have pronounced; if there is a good reason, they may deviate from it. Another aspect that I want to mention is that in favour of the parties, or in favour of a better procedure, the binding effect should be limited. That especially applies, for example to access to file, if in appropriate cases access to file can be given earlier than you normally find in the best practices. So I would like to conclude with that and let me add that most of the proposals that John made, I would also have made. Thank you. Mario Siragusa: I will concentrate on what Jochen characterized as internal best practices, and in particular the Commission’s best practices. The Commission has made a substantial effort trying to codify and standardize its internal procedures, and there are in fact two groups of best practices: one specific to mergers,12 and 11 12

http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. Link above note 5.

334  Merger Control in European and Global Perspective a second, broader best practice for antitrust proceedings.13 On mergers, the 2004 best practices include, of course, a standard model for commitments and the trustee mandate. Then there are the more recent remedy notices,14 so that’s the block of best practices which pertain to mergers. We recently saw more best practices for antitrust proceedings published.15 There, we have of course the best practices for the submission of economic evidence16 and the guidance on the role of the Hearing Officer.17 So these are the two groups that I was talking about. It seems to me that all this documentation plays a very important and excellent role, and is inspired basically by enhancing the cooperation spirit of the relationship between the Commission and the companies involved in those various processes. That is clear, for instance, from the efforts that have been made to clarify the pre-notification process. What I find particularly important is the agreement that has been reached for the Commission to pursue fact-finding also in that process and even to talk to third parties, with the agreement of the merging entities. I think this is a very interesting development, which has clarified a very important portion of what happens during the pre-notification process. Also, I think these best practices have been inspired generally by an effort to be more transparent. The state of play meetings and the access to some of the key documents before the SO is sent all seems to me to be a step in the direction of more transparency. There have also been important decisions in the more general antitrust best practices, published more recently. In particular, I found this idea that there should be an earlier formal opening of the proceedings helpful, so that everybody knows that the proceeding is starting. Also, publications on the development of the proceeding, the issuing of the SO and the final decision, so that the opening of the procedure will be publicly announced so that everybody is informed about the progress of the proceedings. Now I would simply like to make a couple of remarks on ways in which to improve this, because I think everything can always be improved. First of all, a reflection on what the nature of all these documents is. I suppose that, particularly to the extent that they standardize proceedings that are applied by the Commission in individual cases, the Court will interpret those in the same way it has done with the Commission’s other guidelines. So we may see a development of case law 13 See now Commission Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU, 2011 OJ C308/6. 14 See Commission Notice on remedies acceptable under the Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, 2008 OJ C267/1. 15 See above note 13. 16 For the submission of economic evidence (in antitrust as well as merger cases), see http:// ec.europa.eu/competition/antitrust/legislation/best_practices_submission_en.pdf. 17  See http://ec.europa.eu/competition/consultations/2010_best_practices/hearing_officers.pdf. At the time of the 2010 Workshop proceedings, the terms of reference of the Hearing Officer were stipulated in Commission Decision on the terms of reference of hearing officers in certain competition proceedings, 2001 OJ L162/21. Those terms of reference were later expanded to some degree (e.g., during the investigative phase, or in the context of commitment proceedings under Article 9, etc.) by Decision 2011/695/EU of the President of the European Commission of 13 October 2011 on the function and terms of reference of the hearing officer in certain competition proceedings, 2011 OJ L275/29. For details, see Wouter Wils, “The Role of the Hearing Officer in Competition Proceedings before the European Commission”, 35 World Competition 431 (2012).

Panel 5 

335  

which says that a state of play meeting can take place and, if requested, must take place. So we can imagine this at least for certain aspects of these guidelines. I can imagine that there are other areas of the best practices which probably are more general and which may not be subject to judicial intervention. But this is probably something that we may see developing in the next few years. As to the possible improvements, I found that there are four areas in which I think some improvements can be made. More generally, on access to the file, I think that it should be possible to allow access from the beginning of the proceedings, both in merger cases and in antitrust cases. Once the procedure is open, and of course in the case of mergers with notification and in the case of antitrust proceedings with the formal opening of the proceedings, the file is being built. In some of our national procedures, parties have the right to access the file periodically to be informed as to the exact status of the file and the documents contained therein. This allows the party to be on a level playing field with the Commission, because when the Commission comes to know, for instance, what third parties are saying on the proposed merger, I do not see why the merging parties should not be put in exactly the same position as the Commission. This is even more important in antitrust proceedings, but it certainly could apply also to mergers. So I think that giving access to key documents before the SO is a good step in the right direction, but I propose as a next step to see whether it is possible to give full access to the file periodically during the course of the proceedings. A second suggestion relates to contacts with third parties. The Commission will of course meet and talk to third parties in the process of merger review, as in the process of antitrust proceedings. But here there is this distinction between formal and informal contacts, which I find troublesome. More generally, I think that all meetings that Commission officials have with parties – both parties to the proceedings and third parties – should somehow be recorded. There should be minutes of the meetings, summary minutes of what has been said, and this should be put in the file, with access for all interested parties. Again, this is done by some national authorities and I don’t see why this should not also be done at the Commission level. This is particularly if the results of those meetings somehow affect the Commission’s thinking and can be used by the Commission when it makes its findings. There should be some record of it in the file; otherwise there is no way it could be subject to review by the Court. The third area where improvements could be made concerns the role of the Hearing Officer. Now, I’m not sure if this can be done by way of best practices. Maybe this needs an ad hoc communication of the Commission. But certainly there were some expectations when the Commission announced the recent publication of the best practices that this could have been the occasion to somehow modify and enhance the role of the Hearing Officer. I don’t think that this has been done. The Hearing Officer’s role has been codified and explained, but I did not find anything new in the document. I’d be interested to hear whether you agree with this analysis or whether you think that there has been a change in that respect. I

336  Merger Control in European and Global Perspective certainly think that the Hearing Officer could play a more comprehensive role, not limited to procedure and access to file issues.18 My last comment is on an improvement which has been made and which probably relates more to general antitrust proceedings and not so much to merger review. It has been announced that there will be an early, official opening of the antitrust proceedings. Generally, what we have seen is that this opening is done with a very short communication, sometimes only one page or half a page as an indication of what the area is that the Commission is investigating. Not much information is given as to the scope and the real possible allegation or preliminary theory on which the opening of the case is based. To the extent that the Guidelines say that the opening of the procedure will come after an initial assessment of the case, maybe it could be useful to have a more clear indication of the nature and scope of the matter when the Commission announces the opening of the proceedings. So these are my comments on the specifics. Otherwise, I think that it is a very good initiative and I am sure that the Commission will continue to update its best practices because of the need for continual improvement. This is particularly because some of those issues covered by the best practices the Commission really have to do with due process and with the way in which the specific proceeding is going to be followed by the Commission. Thank you. Lowe: Thanks very much. John said it right at the beginning: you can’t have best practice unless you have promised to improve it regularly. For those of you who have not been to the Competition Workshop before, it has been a major driver of improvement on substantive issues and on procedural issues. The 2009 Workshop featured a barrage of criticism from some of the people around this table, but others as well, saying that the Commission didn’t do enough in this area.19 I think you can actually attribute the appearance of a document on best practice in antitrust proceedings and the submission of economic evidence and the organization of hearings precisely to the pressure you put on us on that occasion. This is just as, three or four years ago, I would say that thanks to you and to the Workshop, I was able to argue – with great difficulty – for a Guidance Paper on the application of Article 102 to unilateral conduct.20 So we have a clear indication of a cycle of debate and reaction. It’s not surprising that you don’t see a change in the precise role of the Hearing Officer, because it would have been rather odd for Neelie Kroes or myself to have left to Commissioner Almunia the test of having to do something radical. That’s up to him and his team to consider. Now, would anyone else like to comment on these best practices? 18 As seen the previous footnote, the Hearing Officer’s terms of reference were enhanced to some extent pursuant to Decision 2011/695/EU of the President of the Commission. 19 See Ehlermann and Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011. 20 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.

Panel 5 

337  

Sven Völcker: Just briefly on a couple of points. First, on Mario’s suggestion that best practices might become justiciable: I personally have my doubts, speaking under Judge Forwood’s control, of course, that the Court would ever get to that point. Looking at the past litigated merger cases, perhaps the procedural deficiencies have not been well pleaded, but they certainly have not succeeded – with the notable exception of Schneider,21 which may have been a special case. The Courts’ substantive review has been much more searching than perhaps its willingness to accept that there were certain flaws in the procedure here and there. I’m also not sure we should wish for that kind of review, because the moment it is clear that the Commission will be held to the best practices in every case by the Court, we will see a freezing of those practices and they will not evolve to the stage of being the very best. The other point is generally transparency, which I think is something the DOJ has been very outspoken on. In terms of access to file, there have been a lot of practical improvements that are very noticeable over the last year or two, down to things like an electronic index with hyperlinks to the document which saves you, in the very short time frame that you have to respond to a Statement of Objections, an enormous amount of time. I also think things like supplementary access to the file, really up until the point that the Commission needs to get close to taking its decision, are very helpful. I’m not sure we need to go to the process that the Bundeskartellamt and perhaps other national authorities have of really giving rolling access to the file. I think we all know, after having looked at these files, that 95% of it is pretty irrelevant and we would prefer in the first phase that the case team be fully focused on resolving the key issues of the case. Nevertheless, I think there is one thing that I’ve never really understood, which is that, in theory, if there are key documents in a case, you get access to them under the best practices the day after the opening of Phase II. But why do you get access to them only after the Commission has taken that decision to go into Phase II, which we all know has very significant consequences for the parties and the timing of the deal? If there is really a key document, whether it is a separate complaint or a request for information which has a 30-page response to the last question – “Do you have any concerns about this transaction?” – why not give access to that to the parties immediately? Certainly, in one case that I’ve had recently, I think that getting that information outside the framework of the best practices was extremely helpful for resolving a lot of technical issues in that case. In that vein, Rachel, I guess that the DOJ doesn’t have any best practices in terms of access to the file, because there is no file in that sense. However, I wonder whether giving access to key documents at a stage before the DOJ actually goes to court must also help the cause of transparency. Nicholas Forwood: So far as the legal status of best practice statements is concerned, there are obviously two ways that that could come before us. First of all, in the context of a plea of legitimate expectations, perhaps its limits are not quite 21

Case T-310/01, Schneider Electric SA v Commission [2002] ECR II-4071.

338  Merger Control in European and Global Perspective as broad as you might think they were, at least in relation to the sort of things that can be covered by it. What is becoming more important of course is the principle of good administration, which has now got support in the Charter of Fundamental Rights.22 Now, a person is entitled to have the administrative authorities deal with his case properly and effectively and so on. Clearly, statements as to best practice, which we have to remember is not necessarily exactly the same as good practice – best may be ideal, but good may be good enough – are clearly going to be relevant in that connection. One particular area where it may be relevant is in relation to the control of complex economic assessments in areas where there is, in some form or another, a marginal control. There, of course, as the Court reminds us in Tetra Laval,23 you verify the basic facts and so on. You also look to see, in the words of the General Confession of Sins, whether you have left undone those things that you ought to have done, or whether you have done those things you ought not to have done. In that connection, I can well see that if there is a statement in a best practice document that one ought to do X, Y, or Z, but that hasn’t been done in a particular case, then that may be the sort of thing that might be thrown in to undermine a complex assessment, to the extent that we would not otherwise be looking at it closely and forming our own view if indeed that be the case. Kirsten Edwards: In terms of submission of economic evidence, I really welcomed the best practice guidelines, but I think a lot of it is actually down to communication between the individual case team and the parties and their advisors. I’ve had very different experiences even since the best practice guidelines have been published in this respect. My best experience, which I think really demonstrated best practice on the side of the Commission and, not to be too modest, on our side too, was in Kraft/Cadbury,24 where we did a vast amount of work in Phase I. This was one of these cases where there was a merger simulation in Phase I, but the only reason we could do that much in Phase I was partly because we spent so much effort trying to pre-empt all of the questions that the case team and the Chief Economist’s Team would come up with, because we didn’t have time to go through a lengthy process of questioning. The case team was also very up front with us in that case, at every stage, on exactly what they had seen so far, what their views were at that time, and what more they needed to see to enable them to decide one way or another about it. Because of that, we did manage to get to a Phase I clearance with divestments in a way that everybody was happy with. 22 The right to good administration is provided for in Article 41 of the Charter, 2010, OJ C83/02. Article 41 provides, in its first three paragraphs, that “1. Every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions, bodies, offices and agencies of the Union. 2. This right includes: (a) the right of every person to be heard, before any individual measure which would affect him or her adversely is taken; (b) the right of every person to have access to his or her file, while respecting the legitimate interests of confidentiality and of professional and business secrecy; (c) the obligation of the administration to give reasons for its decisions. 3. Every person has the right to have the Union make good any damage caused by its institutions or by its servants in the performance of their duties, in accordance with the general principles common to the laws of the Member States.” 23 Case 13/03 P, Commission v Tetra Laval [2005] ECR I-1113. 24 M.5644, Kraft Foods/Cadbury (6 January 2010).

Panel 5 

339  

I think everyone was really proactive in that case, but I’ve been on cases where on our side we’ve just had no idea what the Commission was thinking at a particular point and I’m sure the Commission was thinking: what are these crazy economists going to put in front of them next? (laughter) So I think a large part of it is about communication, but I also think one of the challenges is actually talking to the parties themselves about best practice, because I think most people around this table and in the profession are in a lot of agreement about what best practice should be, particularly as regards economic evidence. However, when it comes to talking to your client and saying that they need to be a bit more here or there, until it comes out of the Commission’s mouth they are not willing to do anything. They are always willing to put their money where their mouth is and do as you advise, but often they think: “Why should we do this, why isn’t the Commission doing this, why is this on us?” So I think there is still a challenge in trying to get the companies who are being investigated to understand where we are coming from with the best practice. Lowe: A pre-condition for that is that the Commission has to know where it is coming from when it actually receives the original pre-notification, and starts to develop a theory or potential theories of harm and has an investigative technique which isn’t a fishing expedition. The opposite to that is when it has absolutely no theory of harm, thinks there is no problem, then it gets a key document which looks brilliant. Then it needs a non-confidential version to be able to share it with everyone else, and it’s two days before the deadline, and we can’t meet what Mario said is the right approach, which is to share key documents which are fundamental to opening Phase II. But Thomas is going to tell us about his own practical experiences. Thomas Deisenhofer: First of all, one general comment on due process, which is taken extremely seriously by Commissioner Almunia and Alexander Italianer.25 One counter-intuitive thing is that you think you can’t get enough of due process. The reality is that, depending on where you stand, there are trade-offs. There is first of all due process and speed, and effectiveness. Then there are tradeoffs between the due process of the one and the due process of the other, which is the most important thing. Confidentiality involves tradeoffs on all sides too. All these points that you have made are extremely delicate points where the difficulty is to find the right balance and to calibrate it correctly, so that you provide as much due process as you can while maintaining speed and while maintaining the rights of third parties. Depending on where you stand as a lawyer, obviously you will defend more this side than the other side, but due process is not an unlimited resource. The second thing is how seriously we take it in practice. It’s not just what is written in the rule that’s important, but what we do. I think Sven mentioned 25 See, e.g., Joaquín Almunia, “Fair process in EU competition enforcement”, SPEECH/11/396 of 30 May 2011; Alexander Italianer, “Safeguarding due process in antitrust proceedings”, Fordham speech, 23 September 2010.

340  Merger Control in European and Global Perspective key documents; it’s true that there is currently no rule that key documents before Phase I can be given, but as Philip said, on the one hand the case team has in reality 15 days from notification during which a lot of time is spent waiting for the replies to questions we try to send out the day the notification arrives. We send them out so that the companies have a bit of time to reply, then we try to coordinate inside. So there might be a document coming up, and then the internal coordination, and then on day 15 we have to speak to the parties. I think it’s right to tell the parties where we’re going, but if on top of that you then have to produce non-confidential versions of key documents, with the difficulties of getting them from third parties, and then manage possible remedies or whatever, that would be a challenge if it were to become mandatory. What is happening in practice, as you’ve rightly described, is that if it’s possible, and if the complainant has no problem with it, then directly or indirectly you can promptly identify relevant technical issues. Routinely, at least at my level, we try to give non-confidential versions of documents even in Phase I in order to sort out things because the ultimate objective is to get to the issues as soon as possible. My last point is on attitude. It’s all about attitude: attitude of the enforcer, and of the parties and third parties. If there is a give and take, and if people are reasonable, then solutions are normally found beyond the written best practices. Bill Kovacic: I have a response to Sven’s question posed to Rachel, and she can certainly address it herself as well. I think there’s a basic tradeoff inherent in the US system, that is that parties like the fact that we can’t do anything that hurts until we’re in front of a judge and a judge says so. That inhibits access to the file, and I think if we had the capacity to take the decision in the first instance it would be much more forthcoming. The tradeoff in the system design is that as long as we have an adversarial contest to resolve the matter, there is always going to be an inhibition to putting all of the evidentiary cards face up. If we had a process that resembled that of the Commission, if the original Hart-Scott vision had been in place, or if the FTC for example could take the steps that the European Commission could take, I imagine that we would have a similar bundle of disclosure safeguards ex ante, but the possibility of litigation makes it hard to be as revealing as we might be. Thank you. Rachel Brandenburger: All I would add is that, from my experience in private practice counselling transatlantically, although the procedures were different and one spent quite a lot of time explaining to clients why it worked this way and that way, I’m not aware of cases in front of either US agency where I felt that I did not understand what the theory of harm was, or what the evidence was in front of the US agencies, even if I did not have access to the file. If you have other experiences, then we need to hear about them. Howard Shelanski: You may remember the proposals that Charles James made in 2001–2002, where he said that the DOJ was willing to put things on a different

Panel 5 

341  

path if it had an earlier common revelation of information, especially in instances where a consensus might be achievable. I don’t know exactly what the experience with that turned out to be, but it was a commitment to put more information forward earlier. I hear continuing complaints as well, Sven, about whether, even without access to the file, we are forthcoming enough. I don’t know whether that’s because we don’t say enough on our side at the FTC, or whether we say it quite clearly but the parties don’t want to believe it. Cal Goldman: Just one or two quick additions. When I was running the Canadian Competition Bureau I consulted with my US colleagues about issuing news releases even when we took no-challenge decisions. I was cautioned that if we did that, then we would end up being sued. Well, we’ve come a long way. The Canadian and the US agencies now issue all kinds of information. That’s a best practice. Secondly, the US authorities, especially the FTC, should be commended for stakeholder consultations on an international scale. I suggest that what the FTC has done on global hearings, consulting the global community and not just stakeholders in the US, is a very good example of trying to achieve as much input through a form of best practice as is possible. It’s a model for other jurisdictions. Thirdly, I’d like to make a small point concerning the OECD. Peer review at the OECD does not generally include BIAC. I think it should. It would be constructive to consider including BIAC at least in the country reviews up until the point where there is any discussion of confidential enforcement actions. I don’t see any harm, I can see only good coming of it. Thank you. Lowe: Thank you very much. I think we’ll draw a line under that discussion of best practices. I just want to say one or two things in conclusion of this day and a half. In June of last year, I took an executive decision that we should discuss mergers. Not because I thought that there would be a huge amount of disagreement on many issues, but because it was an opportunity to discuss the new guidelines envisaged in the US and the UK. I think we did have a very stimulating discussion on the use of methodologies, including market definition issues and other investigative techniques. I didn’t notice a great deal of alarm, but certainly some caution about using those techniques in the right context, and recognizing too that virtually all our techniques need to complement each other to arrive at a robust result. On the standards of assessment and evidential requirements before the EU Courts, I think we’re still going to debate that quite a lot. At the moment I don’t see at least the Commission in Europe, or the EU Courts, actually moving on any particular issue of structure or procedure. Maybe that will become necessary if indeed we have another spate of Court activity with respect to certain cases, and there are some in the pipeline. On the procedural issues of convergence particularly, I thought that we would not expect to have a great deal of advance compared with the work already done in the ICN context and elsewhere, but I did see, in both yesterday’s session and this morning, continuing emphasis on evaluating performance and

342  Merger Control in European and Global Perspective indeed confronting ourselves with either the criticism of over-enforcement or under-enforcement. What that would mean in terms of ongoing work inside our agencies almost certainly needs clarification. We emphasized how important it was to build up a relationship of cooperation and confidence between those people who are actually going to tackle the same issues in the same markets. I did raise a small alarm bell, which I think will become larger, that the more our markets become global, the more pure competition will be mixed up, at least in the public perception and in the government’s perception, with other issues such resource scarcity, security of supply, and general concerns about concentration. These may take any agency, if put under pressure, beyond what it would regard as the normal standards of proof necessary to prohibit a merger. I’m just saying that this is an alarm bell, and it may be less loud if we are able to look more closely at the longer-term effects of further concentration in global markets. It comes back to the issue of what the strength of potential competition is outside any particular jurisdiction, and what the serious assumptions being made by one jurisdiction are in taking a decision in relation to the effects in its own jurisdiction. There, thank goodness that there is the Bundeskartellamt to look at BHP Billiton. With the Commission, obviously. But, without some constructive cooperation between the agencies in relation to transactions like that, I think that there would be a serious issue of government failure of addressing major transactions, which have maximum impact in practically all jurisdictions. On the last issues of best practice, we can only get better. Looking ahead, Mel and I will be thinking about next year’s subject, and we’ll undoubtedly choose a more controversial one where we can scratch each other’s eyes out on procedural issues or inter-institutional issues. I thank you very much, particularly those who have come a long way to be with us to share in this discussion. As I say, the Workshop hasn’t been as bloody as some of the previous ones. Last year, when we discussed due process there were far greater, far more powerful epithets used than anything in the last 36 hours, but those lively discussions did lead us to interesting conclusions.26 We can say that there’s a lot of progress being made, but there are some alarm bells ringing and we shouldn’t be complacent. I thank you again for your participation, and as always we will try to set an agenda next year to match the standards set by Claus Ehlermann and all of you in times gone by. Thank you (applause).

26

See Ehlermann and Marquis, Evaluation of Evidence and its Judicial Review, cited above note 19.

Sven Völcker1

Dare to Defer? Towards Greater Procedural Efficiency in Multijurisdictional Merger Remedies

1. Introduction Inter-agency coordination in the design of remedies in multijurisdictional merger cases has long been a reality. Agency press releases routinely tout successful cooperation with sister agencies around the world in multijurisdictional cases. While regulators and advisors involved in such cases will sometimes admit privately that day-to-day cooperation was less than perfect, stakeholders are clearly working hard, and it seems by and large successfully, on avoiding inconsistent remedies demands. Yet the increasing number of agencies who are involved in remedies discussions in one and the same transaction suggests the potential limits of a fully cooperative approach. Remedies discussions almost inevitably come at a late stage of an already long review process, which typically runs up against statutory time limits. The remedies in question may undermine the commercial logic of the deal and give rise to difficult negotiations between the parties in parallel to discussions with the regulators. Negotiating remedies even with a single agency can quickly turn into a 24/7 task. Involving multiple regulators across time zones makes such discussions considerably more complex, even if time lines can be fully aligned, which is not always possible. Inevitably, the slowest jurisdiction determines the overall pace, and the jurisdiction with the most burdensome requirements determines the scope of the remedy package. And even a fully coordinated remedy can give rise to disagreements at the implementation stage, such as who is the most suitable buyer of divestiture assets. Multiple agencies negotiating and policing a largely identical remedies package is almost by necessity duplicative and inefficient. In an ideal world, situations in which there would seem to be a common solution to a common competition problem should be dealt with by a work-sharing agreement between enforcers. Such a work-sharing agreement could take different forms. One or more jurisdictions with a more limited nexus to the transaction, or the remedies required, could simply “step aside” and defer to the judgment of (an)other jurisdiction(s) as to whether and, if so, what kind of remedy is required. One or more jurisdictions could wait until a remedy has been formalized in another jurisdiction, and then take the existence of that remedy into account in its decision to close its investigation. Another option would be an explicit “positive comity” approach, under which 1

WilmerHale, Brussels.

134  Merger Control in European and Global Perspective one or more jurisdictions explicitly request one or more “lead” jurisdictions to take their interests into account when fashioning a remedy. A middle-of-the-road approach would be to have one or more jurisdictions take the lead in the remedies negotiations, but with the other agencies closely involved in the discussions and being able to “opt out” and seek a further-reaching remedy if they believe it necessary to preserve competition within their respective jurisdictions. This paper discusses the extent to which such conceptually appealing ideas can be implemented in the real world. It briefly explores the extent to which agencies already defer to remedies (or remedies negotiations) in other jurisdictions today. Using the EU Merger Regulation as an example, the paper then seeks to identify the legal constraints to various forms of work sharing among jurisdictions. Last, the paper discusses some practical suggestions for implementing a work-sharing approach.

2. Agency practice with respect to remedies agreed in other jurisdictions There are several publicized examples of agencies across the world having recognized remedies that were formalized in other jurisdictions in the context of their forward-looking assessment of proposed transactions. Most recently, the US DOJ’s press release announcing that it was closing its investigation into the Cisco/ Tandberg transaction explicitly made reference to the (worldwide) commitments that the parties had given to the European Commission.2 In Dresser/Halliburton (admittedly twelve years ago), the Commission followed a similar approach when it noted in its Phase I clearance decision that the parties had already agreed a divestiture remedy with the US Department of Justice, that the relevant undertaking was attached to the parties’ formal notification, and that “on this factual basis, the competitive overlap on the market for drilling fluids will be removed”.3 In that case, it seems that the parties had waited to complete their notification before the Commission, filed several months earlier, until the divestiture commitment with the DOJ had been negotiated.4 National competition authorities in Europe have similarly deferred to remedies agreed and/or implemented in other jurisdictions.5 2 Press Release, U.S. Department of Justice, “Justice Department Will Not Challenge Cisco’s Acquisition of Tandberg” (29 March 2010), http://www.justice.gov/atr/public/press_ releases/2010/257173.htm (“The EC also announced today that it has cleared the transaction. Cisco has made commitments to facilitate interoperability between its telepresence products and those of other companies as part of the EC’s merger clearance process…. The department has taken the commitments into account, along with various market factors, such as the evolving nature of the telepresence business, in reaching its decision to close its investigation.”). The press release cites the case as “a model of international cooperation between the United States and the European Commission.” 3 M.1140, Dresser/Halliburton (6 July 1998), para. 7. 4 Ibid., para. 6 5 See the OFT’s decision in ME/1024/04 – General Electric/Invision Technologies and the German Cartel Office’s decision in B 5-130/97 – Federal Mogul/T&N.

Sven Völcker 

135  

Notwithstanding their apparent willingness to take into account other agencies’ remedies on a case-by-case basis, most agencies have been reluctant to make any policy statements in this regard. A notable exception is the Canadian Competition Bureau, which set out its approach in its 2006 Information Bulletin on Merger Remedies.6 The Bulletin acknowledges that the Bureau will sometimes not take separate action to formalize negotiated remedies in Canada, even when a transaction is expected to result in a substantial lessening of competition in Canada. According to the Bulletin, the Bureau may rely on “the remedies initiated through formal proceedings by foreign jurisdictions when the asset(s) that are subject to divestiture, and/or conduct that must be carried out as part of a behavioural remedy, are primarily located outside of Canada.”7 Conversely, notwithstanding Assistant Attorney General Christine Varney’s recent statements that appeared strongly to endorse the Canadian approach,8 neither the US agencies nor the European Commission has issued any comparable policy guidance. The Commission’s Remedies Notice does not mention the possibility of taking into account remedies negotiated in other jurisdictions, nor do the “Best Practices on Cooperation in Merger Investigations” issued by the US-EU Merger Working Group. The US-EU Cooperation Agreement of 1991 contains only basic “positive comity” principles,9 whereas the 1998 Agreement, which expands those principles at least slightly, does not apply to mergers.10 The lack of policy statements by the major agencies on both sides of the Atlantic may have to do with the perceived lack of the issue’s practical importance, as these agencies may typically expect to be “lead” jurisdictions for most matters falling within their jurisdictions, which are geographically broad and thus for the most part co-extensive with the relevant geographic market, whereas the Canadian Competition Bureau may more often be confronted with cases for which the relevant geographic market is North America. However, a recent survey of national competition authorities in Europe – which should be used to dealing with cases in which the relevant geographic market is larger than their own jurisdiction – also suggests considerable reluctance to give any formal recognition to remedies agreed in other jurisdictions.11 6 Competition Bureau Canada, Information Bulletin on Merger Remedies in Canada, 22 September 2006, paras..75–80. 7 Ibid., para. 78. 8  Christine Varney, “Coordinated Remedies: Convergence, Cooperation, and the Role of Transparency”, speech, Paris, 15 February 2010, http://www.justice.gov/atr/public/speeches/255189. htm (“That is exactly the sentiment I wish to express, and whatever advance coordination is possible to promote this end is certainly to be recommended, in my view. It is not a matter of passing the buck. It is an effort to respect each others’ sovereignty and to acknowledge each others’ good faith efforts to secure outcomes that are best for consumers worldwide.”) 9 Article V of the Agreement between the Government of the United States of America and the Commission of the European Communities regard the application of their competition laws, 1995 OJ L95/47. 10 Article II(4)(a) of the Agreement between the European Communities and the Government of the United States of America on the application of positive comity principles in the enforcement of their competition laws, 1998 OJ L173/28. 11 See answers to the question: “If a remedy has already been agreed in another jurisdiction, is the Authority willing to waive the demand for a formal remedy?”, Reports on Merger Remedies, The e-Competitions Bulletin, www.concurrences.com/merger_remedies_qa.php3.

136  Merger Control in European and Global Perspective Rather, the apparent divergence between a practical willingness to recognize remedies in other jurisdictions on a case-by-case basis, and the reluctance to formulate policy guidance seems to be driven by a concern that such guidance may be seen as a sign of weakness that could be exploited by the merging parties and perhaps other jurisdictions. Indeed, there have been cases in which agencies have suspected that the parties have deliberately sought to accelerate review in a different “major” jurisdiction perceived to be more sympathetic to their case, especially where the nexus between that jurisdiction and the parties is strong, in the expectation that other agencies will see no other realistic option than to accept the remedy already agreed, or indeed no remedy at all if the transaction is cleared unconditionally. Over time, however, agencies should be able to overcome such “moral hazard” concerns. Agencies could make it clear in their policy guidance that “deference” will only be considered where it has been in a position to give timely input into the design of the remedy. Increasing cooperation in various international fora such as the ICN and the need to work together on cases on a day-to-day basis should help discourage agencies from presenting others with a fait accompli. The Canadian Competition Bureau’s Bulletin – and especially the Bureau’s leadership in its speeches on the subject12 – have made it very clear that the Bureau will examine carefully any remedy agreed elsewhere to ensure that it addresses concerns with respect to the Canadian market, and it is not apparent that Canadian merger enforcement has suffered from its declared policy.

3. Legal constraints on work sharing between jurisdictions at the remedies stage An agency’s ability to participate in international work sharing at the remedies stage is obviously limited by the procedural and substantive legal framework within which it operates. Those frameworks differ considerably, in particular in terms of the degree of prosecutorial discretion and flexibility with respect to time lines. However, it is submitted that, with a modicum of creativity and pragmatism, legal constraints need not prevent sensible forms of work sharing among agencies. This applies even to an administrative system that operates under fairly tight legal constraints, such as that of the European Commission. Confidentiality constraints? Unless an agency is content to step out of the process of remedies negotiations completely by way of a “presumptive deferral” (which most will be reluctant to do),13 work sharing at the remedies stage will 12 See Sheridan Scott, “Canadian Perspectives on the Role of Comity in Competition Law Enforcement in a Globalized World – To Defer, or Not To Defer? Is that the question?”, speech, Washington, D.C., 29 March 2006, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/02049.html29. 13 See, e.g., Scott, cited previous footnote, p. 14 (“Our enforcement approach does not, however, include a determination at the outset of a merger review to presumptively abide by decisions of our foreign counterparts, without ascertaining the competition issues arising in Canada and how they ought to be addressed”.).

Sven Völcker 

137  

inevitably require a fair degree of exchange of confidential information. However, experience in the joint negotiation of remedies has shown that this is not really a concern. It is in the nature of remedies discussions that most of the relevant confidential information is within the sphere of the parties to the transaction. Agencies routinely ask – and receive – waivers from the merging parties to exchange such information. It is not inconceivable that third parties (for example, a joint venture partner of one of the merging parties) might have confidential information that is relevant to the success of a contemplated divestiture. However, it would appear rare that third parties would refuse to give waivers in such situations. Timing constraints? Coordinated remedies negotiations take time, and awaiting the outcome of remedies negotiations in another jurisdiction requires that the statutory review period for taking enforcement action does not run out, lest the agency considering whether to defer lose all leverage in the process. Jurisdictions such as the United States and Canada that have the power to challenge transactions ex post even after the expiration of statutory waiting periods naturally have more freedom in this respect, even though the risk of potential divestiture assets deteriorating will normally lead to a strong preference for resolving remedies negotiations before the expiry of the statutory waiting periods. The EU Merger Regulation, like many other merger regimes, is characterized by relatively tight time limits, in particular at the remedies stage, both in Phase I and Phase II. While the Merger Regulation provides for the automatic extension of time limits when remedies proposals are submitted, these extensions apply to the submission of formal remedies proposals to the Commission itself, and are not triggered by remedies submitted to other jurisdictions. However, as experience has shown, at least with some degree of cooperation by the parties, the Merger Regulation’s statutory time limits need not be much of a constraint. Indeed, the statutory time limits under the Merger Regulation only begin to run (or, as the case may be, they restart) once the Commission receives a complete notification. In Dresser/Halliburton,14 the Commission apparently had a full notification for several months before declaring it formally complete upon submission of the remedies offered to the US DoJ as part of the notification. Such a pragmatic approach may have its limits because one may question the extent of the Commission’s investigative powers vis-à-vis third parties under the Merger Regulation prior to formal notification. However, nothing would seem to prevent the parties from formally withdrawing and re-notifying their proposed transaction at any stage of the process. Such an approach would seem to make sense in particular where the remedy will be fully implemented before the Commission takes its decision (i.e. a “fix-it-first” remedy), and thus modifies the proposed concentration originally notified. Of course, unless they have consulted with the Commission about such a remedy beforehand, the parties proceed at the risk that the Commission will not view it as sufficient for an unconditional clearance. 14

See note 3 above.

138  Merger Control in European and Global Perspective Where delaying the (re-)start of the statutory time limits or “resetting the clock” is not desired, the Merger Regulation offers other means of gaining additional timing flexibility. In particular, in Phase II cases the statutory deadlines can be extended by as much as 20 working days with the consent of the parties.15 Moreover, where the parties argue that the Commission should take into account the remedies agreed – or about to be agreed – with another agency, this should normally be a material factor in the Commission’s assessment. Any delay in the parties’ response to the Commission’s pertinent requests for information – which could for example be due to the fact that remedies discussions in another jurisdiction have not yet been finalized – would give the Commission a legitimate reason to “stop the clock” with a formal decision under Article 11(3) of the Merger Regulation until it receives a complete response.16 Constraints on recognition of remedies agreed with another agency as part of the substantive assessment? Where, as in the United States and Canada, the reviewing agency has full discretion over whether to challenge a merger in court, it need not be concerned, as a legal matter, about how to justify its deference to a remedy agreed in a foreign jurisdiction. In contrast, the European Commission operates under an administrative system that requires it to motivate clearance decisions, and in doing so it must meet a legal standard comparable to that for prohibitions.17 In this context, the question is to what extent an agency can treat a remedy agreed with an agency in another jurisdiction on a par with remedies formalized under its own rules as part of its substantive assessment. Where it is certain that the remedy will be fully implemented before or at the time the transaction closes, i.e. where the remedy is of a “fix-it-first” nature, it should not be controversial that the “concentration” to be assessed by the Commission is the transaction as modified by the remedy. The situation is arguably more complex when the remedy will only be implemented after closing, even where the buyer of divestiture assets has already agreed to a binding purchase agreement (“upfront buyer”). In Dresser/Halliburton, the Commission seemed to assume without more that the divestiture remedy agreed with the DOJ would in fact be implemented. The Commission did not even treat the relevant market as “affected” in its decision.18 Indeed, where a binding remedy has been agreed and there is every indication that it will in fact be implemented in short order, there is no reason to deny the Commission the possibility to take this fact into account as part of its prospective analysis. The outcome of the Commission’s assessment routinely turns on predictions of the post-merger conduct of the parties, competitors and customers, See Article 10(3)(2) of the Merger Regulation. In Case COMP/M.3216, Oracle/PeopleSoft (26 October 2004), the Commission (under admittedly somewhat unusual circumstances) stopped the clock for more than a year pending the provision of additional information by the parties even after the issuance of the Statement of Objections, which allowed the Commission to take into account the evidence presented in – and the outcome of – the US trial. 17 See Case C-413/06 P, Bertelsmann and Sony Corporation of America v. Impala [2008] ECR I-4951, para. 43. 18 See note 3 above. 15 16

Sven Völcker 

139  

without such conduct being the subject of legal obligations vis-à-vis anyone. In Oracle/Sun, the Commission’s decision to clear the transaction turned in part on “commitments” Oracle gave to third parties, and to some extent even on public pledges that were not enforceable as a matter of law.19 The ECJ’s judgment in Tetra Laval suggests that the legality of predicted post-merger conduct is of relevance in the prospective assessment.20 Clearly, there is some tension between the Commission’s general ability to take into account remedies agreed in other jurisdictions as part of its overall prospective analysis of the transaction’s likely effects, and the requirements imposed by the Courts21 and the Commission itself through its Remedies Notice. It is understandable that the Commission would be reluctant to accept a remedy in another jurisdiction that falls far short of the Commission’s own requirements. Two types of considerations might be appropriate here: • Equivalence of remedies in the other jurisdiction. The Commission may want to consider to what extent the design and enforceability of the foreign remedy conforms to the general standards in the Commission’s Remedies Notice. In terms of enforceability, the key considerations would seem to be whether the failure to implement the commitments is subject to independent sanctions (equivalent to Article 14(2)(d) of the Merger Regulation), and whether a breach of the commitments could lead to the revocation of the clearance decision, the transaction having to be re-notified and the authority ultimately being entitled to unwind it as a last resort (equivalent to Articles 6(3)(b) and 8(6)(b)/(7) of the Merger Regulation). For jurisdictions with a well-developed remedies policy, such as the United States and Canada, it should be relatively straightforward to agree that safeguards are equivalent to those existing under the Merger Regulation. • Commission’s ability to revoke its own clearance decision. Pursuant to Articles 6(3)(a) and 8(6)(a) of the Merger Regulation, the Commission may revoke a clearance decision where “the decision is based on incorrect information for which one of the undertakings is responsible or where it has been obtained by deceit”. The scope of these provisions is largely untested.22 However, where there is a clear breach of a commitment given in another jurisdiction – a M.5529, Oracle/Sun Microsystems, (21 January 2010), paras. 630–658. See Case C-12/03 P, Tetra Laval v. Commission [2005] ECR I-987, paras. 74–75 (prospective analysis must take account of disincentives to stipulated post-merger conduct, including possibility that such conduct is illegal, even though the Commission is not required to conduct a detailed analysis of national legal regimes, the likelihood of detection and remedial action by national competition authorities, and ensuing financial penalties). 21 In Case T-210/01, General Electric v. Commission [2005] ECR II-5575, the General Court suggested that the Commission must be able to determine “with certainty” that the remedies proposed by the parties can be implemented (para. 555). On the other hand, in Case T-87/05, EDP v. Commission [2005] ECR II-3745, the General Court held that any commitment should be treated as a modification of the concentration, and that it is for the Commission to demonstrate that the concentration as modified meets the criteria for a prohibition decision (para. 65). 22 The only known case is the parties’ failure to disclose a significant competitive overlap that led to a divestiture once the transaction was re-notified (M.1543, Sanofi/Synthélabo (2 July 1999)). 19 20

140  Merger Control in European and Global Perspective commitment that the parties have communicated to the Commission and on the basis of which they are proposing that the Commission should clear the transaction without requiring a remedy of its own – the Commission would arguably have grounds to revoke its clearance decision. Constraints on accepting remedies for the benefit of other jurisdictions? Just like other jurisdictions’ laws, the EU Merger Regulation’s purpose is to protect “effective competition within the Common Market” and not beyond. One could thus ask whether, in the design of its remedies, the Commission has the power to accommodate concerns that are unique to other jurisdictions. Binding agreements with other jurisdictions that contain positive comity principles, such as the 1991 EU-US agreement,23 could give some legal foundation to such an approach. However, it will in any case be rare that a work-sharing approach at the remedies stage is considered without the consent and strong support of the notifying parties, if only because it is up to the parties to propose commitments in any given jurisdiction, and the Commission cannot unilaterally expand the scope of commitments it is offered. In Cementbouw, the General Court suggested that the Commission has a fair amount of discretion to accept commitments even if they may go beyond what is strictly required to ensure compatibility with the common market.24 Thus, for example, even if a proposed EU divestiture remedy extended to product lines whose divestiture would not strictly be required under the Merger Regulation, but is included by the parties to satisfy the concerns of a competition authority in another jurisdiction, the Commission would not be prevented from accepting such a remedy.

4. Practical steps towards facilitating work sharing in future cases Obviously, not every multijurisdictional merger case is suitable for the kind of work sharing at the remedies stage envisioned here. Even where multiple jurisdictions are affected, the resulting competition problems and solutions are not always global. For example, pharmaceutical mergers will usually have multiple product line overlaps, but remedies often need to be tailored along national lines due to different licensing agreements and regulatory constraints. However, as the globalization of markets and global consolidation continue apace, the need for developing a principled approach to work sharing will grow. Pro-active measures that could be taken by competition authorities would include the following: • Developing a common understanding of core safeguards that would constitute minimum requirements for deference. The decision to defer to 23 See Article V of the 1991 Agreement (which provides that “the notified Party will consider whether […] to expand ongoing enforcement activities, with respect to the anticompetitive activities identified in the notification”) 24 Case T-282/02, Cementbouw Handel & Industrie BV v. Commission [2006] ECR II-319.

Sven Völcker 

141  

remedies negotiated with other jurisdictions is one that requires familiarity with, and confidence in, the remedies policy and practice of those agencies. It is perfectly understandable that an agency considering whether to show deference would expect the other agency to have a remedies policy with an equivalent (albeit not necessarily identical) standard when it comes to designing and policing remedies, for example in terms of protecting divestiture assets against deterioration, vetting possible buyers, or the use of monitoring trustees. It would probably make sense for a core group of competition authorities with developed remedies policies to take the lead in developing common principles for such safeguards. • Developing common principles for work allocation. Based on their experience with multijurisdictional cases, a core group of agencies could develop a set of principles that would help identify situations in which deference to the remedies negotiated with one authority would be particularly suitable. The location of the divestiture assets, as identified in the Canadian Information Bulletin, is certainly one relevant factor, albeit not the only conceivable one. Another equally if not more important consideration would appear to be in which jurisdiction the competition problems identified are such as to require the most far-reaching remedy. One could refer to this as the “remedial centre of gravity”. For example, if a remedy for competition concerns in jurisdiction A requires a divestiture of product lines 1, 2 and arguably 3 as a package to a single buyer, whereas jurisdiction B requires only a divestiture of product line 3, it should not be decisive whether the production assets for product line 3 are located in jurisdiction B. • Outreach to clarify mutual expectations. As mentioned in the introduction, competition authorities’ hesitation to make policy statements on deferring to remedies agreed with other jurisdictions may in part be motivated by fear of being perceived as weak and easily taken advantage of by companies and other jurisdictions. At the same time, on the occasions when cooperation between agencies has not been smooth, it has not been uncommon to shift blame to the parties. Regardless of the extent to which such blame shifting is justified in an individual case, as competition authorities develop ideas for work sharing at the remedies stage – which by necessity is a collaborative venture with the parties to a transaction – continuous outreach and dialogue with the competition bar and other stakeholders should be encouraged.

Ian S. Forrester, QC, LL.D1

Post Plures Unum Streamlining and Simplifying Merger Procedures in an Era of Multijurisdictional Merger Filings

There is a close link between procedure and substantive outcome in any legal or regulatory structure. My contention is that the merger control regime practised in Brussels for the past two decades has become a success, despite gloomy predictions at its birth, has survived the testing fires of judicial review with honour, and is now so well-regarded that companies often prefer to be governed by it and organise their affairs to come within its scope. I have two policy successes to note, and two modest suggestions to make. The first success is rather fundamental: there is a system which functions well, with an astonishingly high level of non-rivalrous cooperation by agencies within Europe and with other non-European jurisdictions. The second success is to note the use made of the Merger Regulation to restrain Member State protection of national champions. As to the suggestions, one is to propose that effort be devoted to making national filing thresholds less irrational and more predictable; the other is that the rules governing when referrals of mergers between the Commission and national authorities be modified and relaxed so as to make them easier to use. Assessing the wisdom in competition law terms of mergers or “concentrations” is accepted as an economic and regulatory necessity. Mergers and merger control are facts of economic life. However, the wide acceptance of that concept has led to a multiplication of merger control regimes all over the world, which in turn has created problems of its own. Genuine cooperation between enforcers is always important but is particularly acute as to mergers since a single negative decision may put an entire deal in jeopardy. Divergence between friendly jurisdictions in assessing international transactions is embarrassingly public and is longremembered. The name is remembered longer than the topic of disagreement, as is the case of, for example, GE/Honeywell. But, even if there is no great battle of principle, a multiple obligation to notify a deal imposes significant costs in terms of time and money. Preparing parallel filings is burdensome. Deciding whether and where to file, multiple filing fees, gathering and translating slightly different information and documents, filling in 1 Queen’s Counsel at the Scots Bar; Honorary Professor, University of Glasgow; White & Case, Brussels. Warm thanks are expressed to my colleagues, Pascal Berghe, avocat at the Brussels Bar, and Ian Reynolds, solicitor, England and Wales, and others unnamed, who cheerfully shared their knowledge gained from many nights of ensuring that mergers were safely delivered, for their contribution to this paper. The opinions expressed are wholly personal.

144  Merger Control in European and Global Perspective forms in several languages and waiting for blessings collectively amount to a tax on business which is difficult to justify in the absence of a genuine link with the relevant jurisdictions.

Historical development of EU and international merger control regimes Let us begin at the beginning. The drafters of the Treaty of Rome responsible for the competition Articles considered a wide range of topics, including discrimination on religious and national grounds, the powers of monopolists and other no longer fashionable topics,2 but they did not consider at any length the possible necessity to prohibit mergers between competitors. The EEC Treaty, unlike the ECSC Treaty, was silent on merger control, despite the important powers it gave to the Commission under Articles 85 and 86 (now 101 and 102 TFEU). Indeed, the creation of a general Community-wide system applying to all industries across the board was not seriously discussed at the drafting sessions in 1956 at Val Duchesse in the suburbs of Brussels. The delegates of the six founding countries had plenty of difficulty finding consensus on the banning of abuses of monopolylike power and on the conditional prohibition of anti-competition agreements. We may conclude that mergers were left for another day. However the Commission quite soon acknowledged a lacuna in the competition regime, and from 1966 began to advocate for the enactment of an EU merger control regime.3 The Commission, followed by the Court, also adopted a creative interpretation of Articles 85 and 86 EEC, which enabled it to approach the control of certain mergers by the back door. The Continental Can decision prevented a dominant undertaking from acquiring control of a competitor. The Court annulled the Commission’s decision on the facts, but upheld in principle the widening of the Commission’s power to control concentrations that strengthened an existing dominant position.4 However, Article 86 EEC was an imperfect tool to review concentrations, as it applied only to preexisting dominant positions. For lack of a better instrument, the Commission also sought to rely on Article 85 EEC. Thus, in Philip Morris,5 the prohibition of restrictive agreements between competitors was stretched creatively. The Commission decision under appeal was, essentially, not to condemn an agreement whereby (after modifications following 2 Ian Forrester, “The Modernisation of EC Antitrust Policy: Compatibility, Efficiency, Legal Security”, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2000: The Modernisation of EC Antitrust Policy, Hart Publishing, 2001. 3 Memorandum on the concentration of enterprises in the common market, EEC Competition Series Study No 3, 1966. Also published in 2 Revue trimestrielle de droit européen 651 (1966). 4 Case 6/72, Continental Can v. Commission [1973] ECR 125; Götz Drauz and Christopher Jones, EU Competition Law, Volume II – Mergers and Acquisitions, Claeys & Casteels, eds., 2006, p. 2. 5 Case 142/84, BAT and RJ Reynolds v. Commission [1987] ECR 4487.

Ian S. Forrester  145   long negotiations with the Commission) Philip Morris acquired a substantial stake in Rothmans tobacco on terms intended to make it a “passive” investment. The Court upheld the Commission’s decision, but in so doing it struck a glancing blow at long-established and important doctrines and assumptions concerning the nonapplicability of Article 85 of the Treaty to corporate mergers and acquisitions and, more generally, to companies buying shareholdings in each other. It handed the Commission a potentially powerful weapon to extend control over corporate mergers and acquisitions in the Community. In fact, immediately afterwards, the Commissioner for Competition threatened to begin applying Article 85 to such transactions if the Member States did not adopt the Commission’s proposal for an instrument permitting it to vet them (versions of which had been pending since 1972). The crucial language of the Court’s judgment was sweeping and open-ended, albeit somewhat Delphic: “37. Although the acquisition by one company of an equity interest in a competitor does not in itself constitute conduct restricting competition, such an acquisition may nevertheless serve as an instrument for influencing the commercial conduct of the companies in question so as to restrict or distort competition on the market on which they carry on business. 38. That will be true in particular where, by the acquisition of a shareholding or through subsidiary clauses in the agreement, the investing company obtains legal or de facto control of the commercial conduct of the other company or where the agreement provides for commercial co-operation between the companies or creates a structure likely to be used for such co-operation. 39. That may also be the case where the agreement gives the investing company the possibility of reinforcing its position at a later stage and taking effective control of the other company. Account must be taken not only of the immediate effects of the agreement but also of its potential effects and of the possibility that the agreement may be part of a long-term plan. 40. Finally, every agreement must be assessed in its economic context and in particular in light of the situation on the relevant market. Moreover, where the companies concerned are multinational corporations which carry on business on a world-wide scale, their relationships outside the Community cannot be ignored. It is necessary in particular to consider the possibility that the agreement in question may be part of a policy of global co-operation between the companies which are party to it. […] 45. In such a market situation the Commission must display particular vigilance. It must consider in particular whether an agreement which at first sight provides only for a passive investment in a competitor is not in fact intended to result in a takeover of that company, perhaps at a later stage, or to establish co-operation between the companies with a view to sharing the market. Nevertheless, in order for the Commission to hold that an infringement of Article 85 has been committed, it must be able to show that the agreement has the object or effect of influencing the competitive behaviour of the companies on the relevant market.”

146  Merger Control in European and Global Perspective The Court accepted that the various undertakings given to the Commission by Philip Morris and Rembrandt (the seller which retained a majority shareholding in Rothmans International) were sufficient to prevent direct concertation. It also considered whether Philip Morris’s stake in Rothmans International “requires the companies involved to take into consideration the other party’s interest when determining their commercial policy” and declined to overturn the Commission’s assessment that it did not. Comments on the judgment ranged from, at one extreme, the suggestion that it probably changed nothing and further clarification was needed (A.H. Hermann, the then-competition policy guru of the Financial Times) to, on the other, the idea that in some kinds of markets share transactions leading to control of the company whose shares are acquired, including mergers, would be caught by Article 85(1), with the result that the Commission’s proposed merger control regulation would need to be recast as a group exemption (Valentine Korah of University College London). There were also many expressions of concern about the validity of share transfers in previously implemented transactions which might now turn out to be caught by Article 85(1). The least that can be said is that there was a range of situations involving mergers and acquisitions where the possible application of Article 85 had now to be considered in a way which had been unnecessary before the judgment. The Commission had a clear interest in maintaining those doubts until it gained some sort of explicit power to control mergers and acquisitions. How the law would develop seemed wide open, and presumably depended on whether the Commission succeeded in getting those powers. The uncertainties surrounding the Philip Morris judgment were instrumental in getting the EU Merger Regulation approved by the Council. The same year as Continental Can, the Commission proposed a regulation on the control of concentrations.6 Some Member States were reluctant to have merger control executed at Community level rather than by their own authorities.7 Two revised drafts were brought to the Council8 before a fourth proposal was finally adopted in 1989, two years after the Philip Morris judgment and after tough negotiation. Thus was Regulation 4064/89 enacted. This Regulation gave rise to a “rich and extensive jurisprudence”, by contrast to other areas where few formal decisions were taken (such as in abuse of dominance cases).9 The 1989 Regulation was based on principles that are still fundamental under the 2004 Regulation: the competence of the Commission to control mergers with a European dimension, mandatory notification of these concentrations, application of a market-based test, and timely decision-making.10 6 Commission Proposal for a Regulation of the Council of Ministers on the Control of Concentrations between Undertakings, 1973 OJ C92/1. 7 Richard Whish, Competition Law, 6th edition, OUP, 2009, at p. 818. 8 1982 OJ C36/3; 1986 OJ C324/5. 9 Nicholas Levy, “EU Merger Control: From Birth to Adolescence”, 26 World Competition 195, 201 (2003). 10 Ibid., p. 197.

Ian S. Forrester  147  

The system works successfully Recollect where we came from. In the 1980s, DG IV was burdened with an extraordinary interpretation of its own powers under Article 85 EEC, according to which many routine provisions of commercial contracts were caught by the prohibition contained in Article 85(1), void under Article 85(2) and finable, and could only be cured by an exemption, which was a very uncommon procedural creature. So proceedings before DG IV were very slow and formal decisions rare, often less than 10 per year for the whole European Community. The process was painfully slow, perfectionist and conservative. Mergers are put together in secret and are rarely very stable until the deal has actually been signed and consummated. The longer the two parties are bound to each other by a promise to agree once regulatory approvals have been obtained, the more likely it is that one of them will wish to withdraw. So it is worth remembering that in 1988 there was widespread doubt over whether the Commission had the productivity, courage and rigour necessary to deliver prompt, useful, pragmatic decisions within one month. This fear proved to be unfounded. The first Director, Colin Overbury, was a former army officer who favoured action, courtesy and pragmatism.11 Contrary to the doubters’ fears, the Commission demonstrated that it was able to manage numerous notifications and deliver decisions within a reasonable period of time. At the time, merger control regimes existed in Canada, Germany, the US and half-a-dozen other countries. Since then, most ICN members (more than 90 countries) have adopted merger regimes. So Europe is not unique in having decided to set one up, but it is unique in achieving one which compels and relies upon cooperation between countries. The 2004 reform aimed at a more effective jurisdictional organisation and a fresh substantive test. Regulation 139/2004 changed the dominance test for the prohibition of mergers that would “significantly impede effective competition” (SIEC) (the previous formulation was whether a concentration would “create or strengthen a dominant position”). The new formulation includes, but is not limited to, creating or strengthening dominance. The 2004 Regulation also developed the mechanisms of cooperation and deference. Under the previous regime, referrals were possible from the Commission to a national authority, at the request of the Member States.12 The new Regulation introduced new mechanisms to avoid multijurisdictional notification. If it would be notifiable in at least three Member States,13 the parties can, prior to the notification, ask the Commission to review the transaction. They can also do the reverse and ask the Commission to have a Member State review their operation, if the merger concerns a distinct market within this Member State.14 Member States can also refer He denied that Form CO was named after him. Article 9, Regulation 4064/89. 13 Article 4 (5), Regulation 139/2004. 14 Article 4 (4), Regulation 139/2004. 11

12

148  Merger Control in European and Global Perspective a concentration to the Commission, if it affects trade between the Member States or competition within the territory of the Member State(s).15 The experience of this cooperation has been remarkably smooth. There is, however, a turbulent special circumstance, which is the second success to be highlighted.

The unexpected importance of Article 21 Member State governments often wish to protect national champions or to avoid losses of domestic jobs by discouraging the acquisition of control over domestic employers. The legal apparatus for challenging favouritism is quite burdensome: although the Commission has the power to take the matter to the Court of Justice pursuant to the Treaty’s infringement procedure (Article 258 TFEU, ex-Article 226 EC), the procedure is slow, highly politicised and formal. The Commission may have well-founded complaints, but years are likely to elapse before the ECJ has pronounced and the State has complied. Challenges to “Golden Share” arrangements, whereby the State retains ultimate control of a politically sensitive industry, have succeeded on several occasions.16 None of these various challenges arrived at a rapid conclusion, and each allowed a fair amount of obfuscation, invocation of public policy, governmental lobbying and the like. Article 21 of the Merger Regulation offers a completely different technique, which has been unexpectedly (to my surprise at least) deployed successfully in several cases. According to Article 21(4), Member States “could take appropriate measures” “to protect legitimate interests” (such as security, media plurality, public policy) in the context of a merger. That said, they are obliged to inform the Commission, which has the procedural power to challenge the measure.17 In a Article 22, Regulation 139/2004. Case C-98/01, Commission v. UK [2003] ECR I-4641 and Case C-282/04, Commission v. Netherlands [2006] ECR I-9141 are only two examples. The Volkswagen case (Case C-112/05, Commission v. Germany [2007] ECR I-8995) was much celebrated for similar reasons. 17 Articles 21(3) and 21(4) provide as follows: “(3). No Member State shall apply its national legislation on competition to any concentration that has a Community dimension. The first subparagraph shall be without prejudice to any Member State’s power to carry out any enquiries necessary for the application of Articles 4(4), 9(2) or after referral, pursuant to Article 9(3), first subparagraph, indent (b), or Article 9(5), to take the measures strictly necessary for the application of Article 9(8). (4). Notwithstanding paragraphs 2 and 3, Member States may take appropriate measures to protect legitimate interests other than those taken into consideration by this Regulation and compatible with the general principles and other provisions of Community law. Public security, plurality of the media and prudential rules shall be regarded as legitimate interests within the meaning of the first subparagraph. Any other public interest must be communicated to the Commission by the Member State concerned and shall be recognised by the Commission after an assessment of its compatibility with the general principles and other provisions of Community law before the measures referred to above may be taken. The Commission shall inform the Member State concerned of its decision within 25 working days of that communication.” 15 16

Ian S. Forrester  149   succession of battles with Spanish and Portuguese interventions, which furnished an occasion to clarify the interpretation of Article 21, the Commission emerged both substantively and procedurally the clear winner. The first controversy in which Article 21 was put to the test involved the attempted repelling of market forces by the Portuguese government.18 It held 12% of the shares of Cimpol, a cement company, and wished to block the acquisition of Cimpol by Holderbank, the Swiss giant. Cimpol was on the way to being privatised, and Portugal regarded the bid as inimical to its strategic vision for the company. Portugal failed to respect the procedures established by Article 21(3) of the Merger Regulation, and was successfully challenged by the Commission. The Court rendered a judgment which was heavily favourable to the Commission, brusquely rejecting Portugal’s contention that the appropriate remedy would be an action under Article 226 EC (now Article 258 TFEU). “55. If, as the Advocate-General has rightly pointed out in paragraph 51 of his Opinion, the Commission were reduced, in the absence of any communication by the Member State concerned, to the sole option of bringing an action for failure to fulfil obligations under Article 226 EC, it would be impossible to obtain a Community decision within the short time-limits laid down by the Merger Regulation, with a consequent increase in the risk that such a decision may be taken only after national measures have already irretrievably prejudiced the merger with a Community dimension. 56. Moreover, the Portuguese Government’s interpretation would render the third subparagraph of Article 21(3) of the Merger Regulation ineffective by giving Member States the possibility of easily circumventing the controls enacted by that provision. 57. It follows that, for the power to review public interests other than those specified in the second subparagraph of Article 21(3) of the Merger Regulation, to be effective, the Commission must be recognised as having the power to rule by decision as to the compatibility of those interests with the general principles and other provisions of Community law, whether or not those interests have been communicated to it.”19

There was a subsequent case involving the Spanish state’s efforts to resist the acquisition by the German energy company E.ON of Endesa, a Spanish energy company, and to favour the bid by the Spanish incumbent, Gas Natural.20 The conflict was fought on a number of battlefields: there was a flurry of litigation in the Spanish judicial and administrative hierarchy, with the competition and regulatory authorities taking different approaches, the Supreme Court suspending an act of the Council of Ministers, and other skirmishes. For present purposes, what is interesting is that the Commission won in asserting its procedural competence to insist on respect for Article 21(4). The ball was passed from the European Commission to national government to national competition authority to national regulator and ultimately back to the Commission. So the structure – whereby the Member State can intervene and can distort, but must say so, and must say why – has worked well. Frank Benyon has 18 19 20

Case C-42/01, Portugal v. Commission [2004] ECR I-6079. Ibid., paras. 55–57. Case C-196/07, Commission v. Spain [2008] ECR I-41.

150  Merger Control in European and Global Perspective provided a robust, Commission-friendly account of these controversies in a short book.21 This was – as I have noted – an unexpected benefit of the Merger Regulation.

Streamlining the notification thresholds Reducing the number of filings There is no worldwide legal standard constraining state discretion over what must be filed by merging parties, so a country is free to adopt its own merger control regimes. The principle of territoriality is well established in public international law. The General Court in Gencor confirmed that the application of a domestic merger control regime to an international transaction “is justified under public international law when it is foreseeable that a proposed concentration will have an immediate and substantial effect” in that jurisdiction.22 In other words, notification thresholds should be designed to catch only those transactions that have an appropriate nexus with the jurisdiction concerned.23 But, correspondingly, it will often be easier to comply with a state’s exorbitant requirements than to challenge their validity under public international law. It would certainly be disproportionate to adopt notification thresholds so broad as to require notification regardless of the effect of the merger on the territory of the reviewing jurisdiction. The aim of merger control must be to scrutinise those transactions that could significantly lessen competition there. In the absence of any appropriate nexus with the local market, states ought to be ready to defer to other jurisdictions better placed to review the transaction. Merger control regimes should not impose unnecessary transaction costs on merging parties. I respectfully voice the unworthy fear that merger controls are in some countries a convenient source of income (and possibly also a source of statistical information). In Europe only, the sums payable vary from zero (EU) to 42,000 euros in Serbia, 50,000 euros in Germany and Hungary, up to 60,000 euros in Italy and Spain and up to GBP 90,000 for large transactions in the UK. Worldwide, the amounts can be even heftier.24 The ICN Recommended Practices provide that merger notification thresholds should incorporate appropriate standards of materiality as to the level of local nexus required for merger notification.25 In other words, each jurisdiction should 21 Frank Benyon, Direct Investment, National Champions and EU Treaty Freedoms: From Maastricht to Lisbon, Hart Publishing, 2010. 22 Case T-102/96, Gencor v. Commission [1999] ECR II-753, para. 90. 23 ICN 2002 Recommended Practices for Merger Notification Procedures, I.A. 24 For instance, the filing fees can reach up to USD 280,000 for large transactions in the US; SGD 100,000 (approx. 55,000 euros) in Singapore; and ZAR 350,000 (approx. 34,500 euros) in South Africa. 25 ICN 2006 Recommended Practices for Merger Notification Procedures, I.B.

Ian S. Forrester  151   aim at not screening transactions which are unlikely to result in any domestic appreciable competition effects. The thresholds should make reference to the revenues or assets of at least two parties to the transaction, or at the very least to the acquired business within the territory of the reviewing jurisdiction.26 The absence of any explicit reference to the local activities of the target would force companies with significant activities in one jurisdiction to notify there all acquisitions they make, even though the deals have no effect in the jurisdiction in question (e.g., the French oil company filing in France its acquisition of a petrol station in Mexico). Conversely, the absence of reference to the local activities of the acquirer would require the notification of the acquisition of any large company within the jurisdiction even though the acquirer was not present on that market (the Brazilian oil company acquires its first and only petrol station in France). In each case, this transaction is unlikely to produce any material effect within the jurisdiction. Similarly, in a well-tuned regulatory world, thresholds should be well-adapted to the size of the local market. They should be regularly reviewed in light of the evolution of the economy of the jurisdiction. Too low a figure requires the notification of mergers without material effect on the territory. Too high a threshold risks being too permissive. The level of turnover threshold chosen properly reflects the competition policy of each jurisdiction. Compare, for example, France’s notification threshold requiring that each of at least two parties achieve a local turnover of 50 million euros with Belgium’s requirement of a local turnover of 40 million euros. In light of the size of the economies, Belgium’s merger control regime is more permissive than that of France. A balance must be found between total enforcement (catching every single transaction that could potentially affect the jurisdiction’s territory) and the cost for society in terms of administrative burden for both the parties and the competition authorities. Inevitably, notification thresholds are marginally under-inclusive. Since mergers are generally viewed as presumptively pro-competitive, merger control regimes should only aim at catching most of the potentially problematic transactions, but not necessarily all of them. The cost for society of over-inclusive notification thresholds should be higher than accepting that a few transactions will escape scrutiny. For instance, it is true that the combination of two large competitors, each of which is active only on its local market, could potentially raise some competition issues. However, removing the requirement of a local nexus would impose an undue burden on many non-problematic mergers. These ideas may seem rather bland, at least within the European Union. However, it is worth reminding ourselves that not so long ago, the notification threshold in Germany, the largest European economy, could be met even if one party to the transaction achieved no local sales there at all, as long as the other party to the transaction had sufficiently large activities on the German market. It is still the case today in no less than 10 Member States that the notification threshold can be met even if only one party derives sales from the jurisdiction in question 26

Ibid., I.C.

152  Merger Control in European and Global Perspective (Austria, Cyprus, Czech Republic, Denmark, Greece (post-merger filing), Ireland, Italy, Slovakia, Slovenia and the United Kingdom (not compulsory)). Under the circumstances, there is still clearly a need for further streamlining of merger control notification thresholds within the EU.

Simplifying the notification thresholds Reducing the number of filings is one step to streamline merger control procedure, but not the only one. Merging parties need something close to legal certainty. They wish to be able to determine easily where to file, especially since failure to notify can lead to the imposition of significant sanctions. For instance, in February 2010 the Dutch NMa fined two companies a total of 707,000 euros for their late notification of the acquisition of joint control over a bio-ethanol trader.27 It also fined NPM Capital 580,000 euros for failure to notify the increase of its shares in Buitenfood.28 In 2007, the Italian AGCM imposed a fine of 309,000 euros on LIDL for having failed to notify not less than 103 acquisitions of shops over a period of five years.29 Greece fined a Cypriot company, Sea Star, more than 3.7 million euros for failing to notify its acquisition of a controlling share in rival shipping company Anek.30 In the US, the CEO of Liberty Media agreed to pay a fine of USD 1.4 million for failure to report the acquisition of shares when Liberty Media was spun off.31 Finally, the European Commission imposed a fine of 20 million euros on Electrabel for late notification of its de facto acquisition of control over Compagnie Nationale du Rhône.32 Merger notification in “small” or “new” jurisdictions should not be lightly overlooked by merging parties. In October 2009, the Jersey Competition Regulatory Authority imposed a fine of GBP 25,000 on Lufthansa for failing to notify the acquisition of British Midland.33 On 28 September 2009, Bulgaria fined Aidatur approximately 420,000 euros for failing to notify its acquisition of a 90% share in Kardjali Tabak.34 Romania35 and the Slovak Republic36 have also imposed fines for failure to notify in recent years. The above list of examples, which is not exhaustive, constitutes a general reminder about the obligation to notify concentrations. The question can also be raised as to whether some of the fines imposed are proportionate in a case of failure to observe a procedural requirement. NMa, press release of 24 February 2010 regarding fines imposed on Trafigura and Sibco. Ibid. AGCM, Procedure n. 16809 – C 8094 LIDL ITALIA/ RAMI D’AZIENDA, 10 March 2007. 30 Decision of 4 February 2009. 31 FTC, Malone/Discovery Holding, Decision of 23 June 2009, FTC File No. 0810219. 32 M.4994, Electrabel /Compagnie Nationale du Rhône (10 June 2009). 33 JCRA Decision C332/08 of 15 October 2009. 34 The Bulgarian competition authority also imposed a BGN 75,000 fine (approx. 38,000 euros) on Oktopod Invest Holding EAD and Slanchev Briag on 12 July 2010. 35 The Romanian competition authority fined Ameropa Holding RON 32,265 on 20 May 2009. 36 The Slovak competition authority fined Scholz 5,100 euros on 4 May 2010. 27 28 29

Ian S. Forrester  153   If the obligation for merging parties to comply with the law and notify their concentrations is undeniable, I submit that the possibility to impose large fines also creates a duty for competition authorities to enact clear, transparent and easyto-assess notification thresholds. Legal certainty is a basic principle protected by Article 7 of the ECHR, which states that: “No one shall be held guilty of any criminal offence on account of any act or omission which did not constitute a criminal offence under national or international law at the time when it was committed.”

The European Court of Human Rights in Strasbourg has consistently held that the “offence should be clearly described by law” if penalties are to be imposed legitimately.37 Individuals must be able to know “from the wording of the relevant provisions and, if need be, with the assistance of the courts’ interpretation” what acts and omissions will make them liable.38 Unfortunately, in addition to the number of filings required, the way notification thresholds are devised is commonly a source of uncertainty or complexity. A classic example of the lack of precision in the drafting of legal provisions is whether the relevant threshold refers to worldwide or local turnover. Does the concept of turnover refer to the relevant legal entity or the entire group, and for what reference period? Too often, a plain reading of the text does not give the answer. Decisional practice of the relevant authorities can obviously help, but this misses the point: parties ought not to have to scratch their heads over simple principles to determine whether and where they need to notify. Second, as recommended by the ICN, the notification thresholds should be based on objectively quantifiable criteria that are clearly defined and based on readily available data.39 The jurisdictional assessment of where to file does not touch the merits of the case; it is simply a determination of where the assessment on the merits will take place. We need objective and straightforward criteria linked to the expected effects of the transaction. The goal should not be to avoid the monster in the video game, but to have a deal reviewed by well-placed authorities. Market sharebased tests are thus not appropriate to decide whether or not a deal should be notified in jurisdictions with compulsory pre-merger notification systems: such tests are too prone to controversies. Market share-based tests are better in the subsequent assessment of the transaction once the reviewing jurisdictions have been identified. However, despite these ICN Recommended Practices, it must be noted that Portugal, Slovenia and Spain have such notification thresholds. (The UK presents a different situation, as notification is not compulsory: an erroneous market definition does not risk incurring a fine for failure to notify.) The notification thresholds should also be based on readily available information. In practice, this means that the measurement tool should correspond, to the extent possible, to information contained in the companies’ accounts. This 37 38 39

Handyside v. UK, [1974] 17 YB 228. Kokkinakis v. Greece, (1993) 17 EHRR 397, para. 52. ICN 2006 Recommended Practices for Merger Notification Procedures, II.B.

154  Merger Control in European and Global Perspective has two advantages. First, it greatly simplifies the analysis, as the parties do not have to reconstruct the needed data. Second, it gives the competition authorities the assurance that the figures have been audited. The issue can be illustrated with the Endesa case.40 In 2005, Gas Natural announced its intention to make a public bid to acquire the totality of the shares of Endesa, another Spanish energy company. It notified the transaction to the Spanish authorities. However, Endesa considered that the merger had an EU dimension and lodged a complaint with the Commission, which was rejected. Endesa appealed to the Court of First Instance on the ground that the Commission had miscalculated the relevant turnover. Endesa argued that the Commission should have taken into consideration figures prepared under the International Financial Reporting Systems, and not figures produced pursuant to the Spanish Generally Accepted Accounting Principles. The applicable Commission Notice was silent on the question. The Court of First Instance held that the relevant set of accounts was the one established and verified in accordance with the then applicable legislation. Since the applicable rules in Spain were the Spanish GAAP, the Commission was right to rely on accounting figures based on these principles and not on IFRSprepared figures. This judgment makes perfect sense. It would be disproportionate for competition authorities to require companies to prepare a new set of accounts based on different principles for the purpose of calculating the relevant sales (antique dealers, banks, farmers and oil companies do it differently). The principles of simplicity and transparency require that the thresholds be calculated on the basis of figures readily available to the parties. Finally, identical concepts should ideally be construed identically. It is pointless for all jurisdictions to enact thresholds in terms of “turnover” if the way to calculate the turnover significantly differs from one jurisdiction to another. In this regard there is no legal principle which can be invoked to force international harmonisation. However, common sense and a proper wish to avoid unnecessary complications should favour some sort of global harmonisation. The ICN seems to me to be the perfect forum to discuss and agree on common definitions. Of course, governments make the rules, and officials can be the victims of rules just like companies and their lawyers. But I respectfully suggest that there are certain criteria in certain jurisdictions which deserve reappraisal.41 Case T-417/05, Endesa v. Commission [2006] ECR II-2533. For instance, the seller’s entire group turnover is to be included in the turnover figures to assess whether a transaction is notifiable in Brazil or in Taiwan, and to test whether the “size of parties” test is met in Canada. Even within the EU, the entire turnover of the seller’s capital group needs to be considered for the purpose of calculating the parties’ turnover under Polish merger control rules. This goes against the logic: if the seller entirely loses control over the target (and thus not in a change from sole control to joint control), it should have no relevance for merger control purposes. It is respectfully submitted that these tests should be harmonised with the rules applied by most countries in the world. Another example of problematic divergence is Austria, where the turnover of a 25% subsidiary is included in the turnover of an undertaking concerned (regardless of whether the subsidiary is controlled by the undertaking concerned). 40 41

Ian S. Forrester  155  

Streamlining the referral mechanisms Introduction The referral process is designed to complement the application of jurisdictional thresholds described above. These thresholds are arbitrarily defined metrics designed to capture concentrations of sufficient size to warrant supranational or national review. Inevitably, there will be times when the Commission’s or NCAs’ turnover thresholds capture transactions that are inappropriate for review at supranational or national level. Recital 11 of the EU Merger Regulation describes the referral mechanisms, which provide for the allocation or re-allocation of cases between NCAs and the Commission, as an “effective corrective mechanism”. Second, one of the signs of the Commission’s success is the fact that some consumers of merger control services prefer the specialised, energetic, concentrated blast of attention which is available in Brussels to the uncertainties of multiple national reviews. Other such consumers may wish to take advantage of the generally shorter pre-notification times at national level, or the typically shorter forms and information requirements in certain countries, compared to Commission standards. In addition, practitioners and parties may consider that transactions could receive more appropriate reviews at national level (where remedies negotiations or substantive reviews may be more flexible or transparent) than at Commission level. Since the adoption of Regulation 4064/89 in 1990, successive Commissions have attempted to facilitate the use of referral mechanisms to ensure that cases are reviewed at the most appropriate level. The most significant step forward in respect of the referral mechanisms was made with the recasting of the Regulation in 2004 (resulting in Regulation 139/2004), which provided for a system of voluntary pre-notification referrals from Commission to Member States and vice versa. The recast Regulation introduced these voluntary referral mechanisms “to further improve the efficiency of the system for the control of concentrations within the Community”, in particular to identify or safeguard “which is the authority more appropriate for carrying out the investigation, the benefits inherent in a “one-stop shop system” and the importance of legal certainty with regard to jurisdiction.”42

42

EU Merger Regulation, Recital 16.

156  Merger Control in European and Global Perspective

Overview of referral mechanisms Pre-notification Referral from Commission to Member States Article 4(4) of the Merger Regulation provides a voluntary mechanism allowing concentrations to be referred to Member States at the request of the parties. The Commission, where it considers that the concentration may significantly affect competition in a distinct market within a Member State, may decide to refer all or part of the case to the relevant NCA, which then applies national competition law. The parties make a reasoned submission to the Commission (using “Form RS”) providing information about the parties and the transaction and markets affected, and explain why they believe the concentration to be eligible for referral. The Commission then forwards the Form RS to all Member States. The candidate Member State informs the Commission whether it acquiesces to the requested referral. Silence is deemed to be agreement. If a relevant Member State disagrees with a referral request, then the Commission may not refer the case to it. Where no disagreement is expressed, the decision to make a referral is left to the Commission’s discretion. The Commission must take this decision within 25 working days of receiving the Form RS, failing which it will be deemed to have made the referral. Referral from Member States to Commission Article 4(5) of the Merger Regulation provides a voluntary mechanism which allows for concentrations without a Community dimension which could otherwise be reviewed in three Member States to be referred to the Commission. In the absence of disagreement, the concentration is deemed to have a Community dimension and is referred to the Commission. The parties send the Commission a Form RS, informing it that in principle the concentration attracts review by three NCAs and should therefore be reviewed by the Commission. The Commission sends the Form RS to all Member States immediately. If at least one Member State that is competent to review the merger disagrees with the referral, then the Commission cannot accept the reference. Where no disagreement is expressed within 15 working days, the case is deemed to be referred to the Commission. No Member State may then apply its national merger control rules to a concentration that the Commission will review under this provision.

Ian S. Forrester  157  

Post-notification Referral from Commission to Member States Article 9 of the Merger Regulation provides a mechanism for referring cases to Member States (which have already been notified to the Commission) where a Member State is better positioned to carry out the investigation. A referral may be made of the entire transaction or only part, depending on the transaction’s competitive effects. A Member State may request (at its own initiative or after an invitation to do so by the Commission) that a case be referred to it in either of the following circumstances: (i) the concentration must “threaten to affect significantly competition in a market” and the market in question must be within the requesting Member State and “present all the characteristics of a distinct market”, or (ii) the concentration must “affect competition in a market” and the market in question must be within the requesting Member State and “present all the characteristics of a distinct market” and “does not constitute a substantial part of the common market”. Sometimes both circumstances are argued.43 Any such request must be made within 15 working days of the Member State receiving a copy of the notification from the Commission. Referral from Member States to Commission Article 22 was originally intended to allow Member States without merger control legislation to refer a case to the Commission. Under the revised Article 22, one or more Member States may request the Commission to examine a concentration without a Community dimension that affects trade between Member States and threatens to affect competition significantly within the territory of the Member States concerned. Other Member States to join a referral request. Member States joining a request automatically relinquish jurisdiction. Any Member State has the right to join the initial referral and (in contrast to Article 4(5)), there is no minimum number of jurisdictions to complete the referral process. Finally, the Commission can take jurisdiction only for the part of a transaction which was referred by the Member States. The parts of the transaction which were not referred remain to be reviewed at national level. The request from the NCAs must be made within 15 working days from the date the transaction was notified to the Member State under national merger control rules or otherwise made known to it.44 The Commission will inform all Member States, which then have a further period of 15 working days to join the initial request. National time limits for review of transactions are suspended until the Commission decides whether or not to assume jurisdiction. The Commission has a period of ten days in which to make this decision. Unlike under the Article Article 9 of the EU Merger Regulation. As with Article 9 of the recast Regulation, the Commission may, on its own initiative, decide to inform one or more Member States that it believes that a transaction they are reviewing meets the criteria for reference under Article 22 and may invite them to make a reference request. 43 44

158  Merger Control in European and Global Perspective 4(5) procedure, Member States that do not join the request may apply national merger control rules to the transaction.

Comments on pre-notification referrals The functioning of Article 4(4) Why so much data? The volume and scope of the information required in a Form RS is, in my view, excessive in light of the stated aim of the referral procedure, which is meant to be the determination of jurisdiction. It should not require detailed market analysis as to the likely impact on competition. Why carry out a substantive review? There is no need both to have a dog and to bark. An abridged version of the Form RS should focus on the issues most relevant to determining jurisdiction and to alleviating the burden on the parties concerned. The Form should prune away the pointless elements. For example, Section 5.4 foresees the submission of information with respect to the five largest independent customers and suppliers; information which normally is relevant only for market testing, a post-notification phenomenon generally begun in the first weeks of the procedure. The information requirements in the Form RS should concentrate on demonstrating that the relevant markets are national and potentially sub-national in scope so that the concentration is more suitable for national review. Moreover, it appears to be the case that the parties could be required to provide additional information “beyond the legal requirements specified in the Form RS”.45 This obscure and opaque requirement should be abandoned, given the already wide scope of the Form RS. Why so long? NCAs are given 15 working days from receipt of the Form RS in which to consider a referral request. Since at this stage the assessment deals merely with a preliminary question in relation to the substantive assessment of a concentration, 15 working days seem overly protracted, particularly as a regular Phase I investigation takes 25 working days to handle both jurisdiction and substantive issues. A further potential factor which prolongs the review process is the fact that the 15 working day review period begins only after the NCAs have received a hard copy of the Form RS, which may be up to three working days from the date the Commission received it. Moreover, different NCAs may receive the Form RS on different dates, resulting in different time lines in different Member States. Electronic submission is not provided for by the Merger Regulation or the Implementing Regulation. I note that such electronic submissions have been introduced in some EU jurisdictions (Italy) and outside the EU (Brazil). The Commission now market tests “electronically” 45

Commission Notice on Case Referral in respect of concentrations, 2005 OJ C56/02, para. 63.

Ian S. Forrester  159   in the merger process and accepts electronic submission of documents in both the merger and antitrust spheres. The European Courts also accept electronic service for many documents. The Commission and NCAs should adopt similar procedures. Time is typically of the essence in merger matters. These issues mean that in practice the time necessary to review parties’ referral requests has become overly long. Between 2004 and 2007, the median time between the initial submission of a Form RS in a first phase clearance decision was 61 to 70 working days. This period does not include any pre-notification discussions required between parties and the Commission.46 Contrast this period with the usual first phase Commission review period of 25 working days, plus pre-notification timetable. The time periods concerned are a burdensome bar to the effective use of the procedures. The time has come to streamline the process. Long lead times were appropriate for the days when the rules and processes were uncertain. Given that, in practice, the Commission handles referrals well within the 25 working days that are provided for, the statutory timeline can without risk be shortened so as to allow the parties concerned to plan the merger notification process with more certainty. Finally, from the Commission statistics published for the period 2004 to September 2010, there appears to be a discernible trend away from the use of the Article 4(4) pre-notification processes. This could be because parties are deterred from use of the procedure by added time, cost and complexity. Between 2007 and 2008 the number of notifications dropped by around an eighth, but the number of pre-notification requests for referral under Article 4(4) fell by around half. This would tend to suggest that parties are discouraged from utilising these mechanisms.47 In light of the foregoing analysis, I suggest that the NCAs should have 10 working days to decide whether to object to a referral, and that the Commission should communicate the Form RS in electronic form to the NCAs on the same day it is received.

Functioning of Article 4(5) Information Requirements As with Article 4(4), an application for a referral under Article 4(5) is made by way of a Form RS. As pointed out above, the information requirements contained in the Form RS are substantial, mirroring large parts of the Form CO, and require significant effort by the notifying parties to complete. In the event that the Commission does not assume jurisdiction under Article 4(5), much of the effort spent by the notifying parties in supplying the detailed information requested in the Form RS may be wasted. 46 See Staff Working Paper accompanying the Communication from the Commission to the Council, Report on the Functioning of Regulation No 139/2004 (COM(2009) 281 final), 30 June 2009 (“Staff Working Paper”), p. 26. 47 See Commission merger statistics, http://ec.europa.eu/competition/mergers/statistics.pdf.

160  Merger Control in European and Global Perspective The mandatory provision of such information is unnecessary at this stage of the process, since the Article 4(5) procedure, as with Article 4(4), is a jurisdictional assessment. There is no need for an in-depth substantive analysis. The guiding criterion contained in Article 4(5) is simply that the notifying party can establish that three Member States require notification. In this context, the need to provide detailed information, such as customer and supplier contact details and a detailed analysis of the markets, does not seem to be relevant at such an early stage of the procedure, when only jurisdictional matters are being addressed. Consequently, I would recommend introducing a new Form RS containing less burdensome requirements and focusing on the elements necessary to establish whether the main criterion has been met. The information requested in Sections 5.4–5.10 seems unnecessary at this stage. Timing The timing arguments raised above in respect of Article 4(4) also apply to Article 4(5), in the sense that the timeframes laid down in the Merger Regulation go beyond what is reasonably required for a jurisdictional assessment of a merger control filing. Indeed, under the Regulation, a Phase I investigation (that is, the vast majority of notifications) covering both jurisdictional and substantive matters is carried out within 25 working days of notification, yet NCAs are granted 15 working days to consider an application under Article 4(5). Indeed, the average time from the receipt of the Form RS to a first phase decision under the Article 4(5) process between 2004 and 2008 was around 46–60 working days.48 More than twice as much time is required for referral cases, as opposed to “own jurisdiction” cases.49 The period under Article 4(5) is therefore disproportionately long when compared to the time allocated for a full jurisdictional and substantive analysis. Such an extensive time period also imposes a significant additional delay on deal timetables, which are generally already very tight. The fact that the time period runs from the date an NCA receives a hard copy of the Form RS, and not from the date on which it is sent by the Commission, further compounds the problem in that it leads to different deadlines in the different concerned Member States – indeed many notifications are sent by post due to the size of the electronic submissions. This latter problem could be solved by requiring applications to be sent electronically to the NCAs and providing that time starts to run when the Commission transmits the electronic files (see also above). The inefficiencies created by the 15 working day period are particularly burdensome on parties to a transaction which does not involve any affected market. Such a transaction would generally qualify for a simplified procedure review (where they exist) in the EU or, in any event, would be cleared by competition authorities in the EU within the initial review period (Phase I). By contrast, a referral to the Commission, followed by a Phase 1 clearance, would be 48 49

Staff Working Paper, p. 30. Ibid., p. 37.

Ian S. Forrester  161   much more time-consuming. The transaction would be clear, at best, after 15 + 25 (= 40) working days, and would require the parties to fill in two forms, the Form RS and then a Short Form CO. Through the “national route”, there may be many notification forms to be completed (but some or all of them may be relatively short, for example in Austria or (an EEA jurisdiction) Norway). The various review periods would be running in parallel and not cumulatively. Since time is often a crucial factor in simple transactions (for example, acquisitions by private equity companies with limited overlapping portfolios), it is understandable why the referral process is not attractive. I would suggest that a more appropriate period for the referral process under Article 4(5) would be a 10 working day period from receipt of the Form RS by the Member States.

General conclusions on the “new” pre-notification referral processes The introduction of the pre-notification referral mechanisms by the 2004 reform was a significant addition to the merger control process in the EU. It brought much needed flexibility and allowed the parties to carry out an analysis of the most suitable jurisdiction(s) in which to notify. That option is very important, particularly given the possibility for NCAs to ask for a referral to or from the Commission post-notification. However, this promised flexibility has not entirely lived up to expectations: whereas the case-referral mechanisms in the Merger Regulation should be devised and applied in a way that allows them to work in an effective, timely and seamless manner, there are two main issues which sometimes make pre-notification referrals unattractive: (i) the time they take, and (ii) the burdensome nature of the Form RS. NCAs should be under tighter deadlines to accept or object to referral requests by the parties. Furthermore, the amount of information requested in the Form RS is disproportionate to its purpose, which is the determination of the jurisdiction competent to review the transaction. Parties are concerned with three key issues: (1) the amount of time regulatory processes will take and the delay’s impact on deal timelines; (2) how much money the process will require; and (3) legal certainty as to their chances of a positive result for their transaction. While the stated intention of the pre-notification referral mechanisms is to enhance both regulatory efficiency and the procedure for parties, it is doubtful that these processes are working at an optimal level. Procedural reforms, of the types detailed above, are required.

162  Merger Control in European and Global Perspective

Post-notification referrals Article 9 An Article 9 referral request, although an NCA is entitled to request it, can cause major disruption to the planning of a transaction. By way of example, the Blackstone Group/NHP Plc concentration was notified to the Commission on 9 December 2004. The Office of Fair Trading (OFT) in the UK requested a referral of the case in accordance with Article 9(2)(b) of the Merger Regulation.50 The Commission granted referral of the case on 1 February 2005 and the concentration was cleared in the UK on 5 April 2005. Had the OFT not requested referral of the case, clearance in Phase I without remedies would have been granted by the Commission around mid-late January 2005: four months “wasted”. This is, firstly, because NCAs have 15 working days after the notification to make a request for referral under Article 9. Secondly, the Commission has until the end of Phase I (if the request is made in Phase I) or until the end of Phase II (if the request is made in Phase II) to accept or reject the referral. Particularly in Phase II, the fact that the Commission is left with so much time in which to decide on the referral request is a source of major legal uncertainty for the parties. One way to remedy this would be to impose an obligation on the Commission to decide on the referral within 10 working days from the date at which the referral has been made. Between 2004 and 30 September 2010 there were 36 Article 9 requests. I expected that the introduction of Article 4(4) would result in a lower number of Article 9 requests in the long term. This is because the parties to a transaction may decide to file for an Article 4(4) referral if they expect one or more national authorities to otherwise use Article 9 in a post-notification referral. However, the expected reduction in Article 9 referrals has not taken place. The equal number of Article 9 requests after 2004 may be explained by the enlargement of the EU in May 2004.51 One concern with Article 9 referrals arises from the fact that, when the Commission accepts a referral by decision, it must establish in the decision that the transaction “threatens to affect significantly” or “affect competition” in a relevant market presenting all the characteristics of a distinct market in the Member State concerned. The merging parties will therefore have to start the new notification process before the NCA with an adverse decision on substance to contend with. Absent a legal challenge before the General Court, the merging parties are left with no means to counter the conclusions of the Commission, especially since they do not have access to the Commission’s file at this stage. In such cases, it would be justified for the parties to have access to the main pieces of evidence on which the Commission has based its Article 9 decision, similar to the process when the parties respond to an Article 6(1)(c) decision. An alternative solution would be for the Commission to establish that the transaction 50 The Commission’s decision to refer the case is described in Press Release IP/05/128 of 2 February 2005 (M.3669). See also OFT, ME/1544/05 – Blackstone Group/NHP plc. 51 See Commission merger statistics, at http://ec.europa.eu/competition/mergers/statistics.pdf.

Ian S. Forrester  163   potentially has a “significant impact” in a relevant market presenting all the characteristics of a distinct market in the Member State concerned.

Article 22 Over the past 15 years, the Article 22 procedure has been invoked only 24 times. It is one of the least used referral mechanisms.52 However, due to the deadlines, it is potentially the most disruptive: • the provision allows for a period of 40 working days from the date a transaction is made known or notified to an NCA for the Commission to decide whether or not to accept jurisdiction; and • once the Commission has accepted jurisdiction, it may require a notification to be submitted; it then benefits from a further 25 working days to issue a decision. Thus, at best, parties lose close to two months once an Article 22 referral is made. In my view, this is not a reasonable timeframe within which to establish jurisdiction. One remedy would be to limit the time granted (a) to a Member State to refer under Article 22, (b) to a Member State to join an Article 22 request, and (c) for the Commission to decide whether or not to accept, to 10 working days respectively (and thus 30 working days in total, instead of 40). The highly disruptive impact that an Article 22 referral can have on a transaction is well illustrated by Thrane & Thrane/Nera, in which the UK’s competition authority made an Article 22 referral after having received a complaint from a third party.53 The Commission accepted jurisdiction and issued a decision some six months after the merger had been cleared by the Norwegian competition authority. This case highlights the unpredictable nature of this procedure, since parties could be in a position where their transaction is cleared by a number of NCAs, yet they are still at risk of an authority (in this circumstance to which a notification was not mandatory) referring the transaction to the Commission. This in turn means not only a significant delay for the transaction timetable, but more importantly the uncertainty of waiting for a competition clearance that was not envisaged and potentially (presumably) not foreseen in the transaction documents. As to whether the introduction of Article 4(5) has had any impact on the functioning of Article 22 is difficult to determine. Article 22 referrals were rarely used prior to the Regulation coming into force (the procedure was used only seven times by NCAs over a period of 14 years) and yet it has been used 18 times subsequent to the introduction of the new Merger Regulation in 2004 (over a period of only seven years).54 This is probably best explained by the enlargement of the Ibid. OFT, ME/2684/06 – Thrane and Thrane A/S /Nera ASA. See also M.4465, Thrane & Thrane/ Nera (21 March 2007). 54 See Commission merger statistics, available at http://ec.europa.eu/competition/mergers/statistics. pdf. 52 53

164  Merger Control in European and Global Perspective EU – and possibly the inexperience of some of the newer competition authorities, which may have been tempted to refer complex cases to the Commission. In the long term, I expect that the use of Article 4(5) by the parties should lead to a reduction in Article 22 referrals, though given the trends in relation to Article 9, described above, the future for Article 22 remains unclear.

Conclusions In each section above, I have identified certain shortcomings of the current referral mechanisms. I recognize I am not alone in noting and voicing these concerns.55 In fact, the Commission itself has not been slow to identify the failings of the current processes (in particular the pre-notification referral mechanisms), though it has stopped short of proposing workable solutions for the future. “[I]t can be concluded that there is still more scope for "one-stop-shop" review. In fact, there are still a large number of cases with a cross-border interest that are neither caught by Article 1(3) nor referred under Article 4(5). Should further improvements of the pre-notification referral mechanisms be feasible, this would most likely result in fewer multiple filings and therefore lower cost for the business community. Conversely, there may be some scope for more referrals in the direction of the Member States in application of Article 4(4).”56

I mention above that the Commission has at regular intervals (1990, 1998 and 2004) reviewed the scope and efficacy of the Merger Regulation. This is to be commended. The time is opportune to continue the process of regeneration and improvement by altering these useful, but at present imperfect, mechanisms to the benefit of the Commission, NCAs and the wider business community. Authorities which once were rivals are normally now allies. Their responsiveness can be counted upon. Time limits established in the expectation of some timeconsuming formalism can be shortened. Thresholds which are obscure should be made clear. Greedy authorities which make too much money from filings should be gently encouraged to be prudent. And we should celebrate a modest European success – the functioning under effective judicial control of a continent-wide system for reviewing significant mergers.

55 56

See in particular, Staff Working Paper, Sections 3.1.4; 3.2.2.4; 4.1.3; and 4.2.3. Ibid., para. 160.

Calvin S. Goldman*

Contemporary US-Canada Cross-border Merger Review: Tradeoffs in Policy Objectives between Harmonization, Simplification and Accountability

I. Introduction • In the past year and a half, there have been substantial changes to the way in which mergers are reviewed in Canada, primarily stemming from: o First, the recent amendments to the Competition Act that have aligned the timing of Canadian and US merger review. Canada now has a US-style second request process in place – within a 30-day initial waiting period following pre-merger notification, the Commissioner of Competition (“Commissioner”) can request any relevant additional information (a procedure known as a “supplementary information request”). Once the request is complied with, a second 30-day waiting period begins. o Second, Canada’s Commissioner as of January 1, 2010, Melanie Aitken, who has articulated a strong enforcement agenda1 and supported the development of updated policies and new guidelines such as the Merger Review Process Guidelines, the Fees and Service Standards Policy for Mergers and Merger-Related Matters (“Merger Policy”), the Fees and Service Standards Handbook for Mergers and Merger-Related Matters (“Merger Handbook”) and the Procedures Guide for Notifiable Transactions and Advance Ruling Certificates under the Competition Act (“Procedures Guide”).2 • As a result of the above, the Canadian competition bar is witnessing more intense and thorough review of strategic mergers.

* Calvin S. Goldman is a Partner and Co-Chair of the Competition, Antitrust & Foreign Investment Group at Blake, Cassels & Graydon LLP, Toronto, Canada. From May 1986 until October 1989, Mr. Goldman was the Director of the Competition Bureau (a position now called “Commissioner”), Ottawa, Canada. 1 See Emily Gray, “Aitken Appointed Canadian Competition Commissioner”, Global Competition Review (6 August 2009) (quoting Aitken as stating “Canadians need to see that there is a vigilant cop on the beat”), http://www.globalcompetitionreview.com/news/article/18484/aitken-appointedcanadian-competition-commissioner. 2 See Competition Bureau, Competition Bureau Releases Updated Merger Guidance Documents (22 October 2010), http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03303.html.

166  Merger Control in European and Global Perspective • I propose to discuss the recent convergence on procedural issues in Canadian and US merger review, as well as continued areas of divergence.3 Next, I’ll briefly touch on some opportunities for further streamlining and simplifying the merger review process. And, finally, I’ll close with some observations on the inherent tradeoffs in merger review between convergence, simplification, and accountability.

II. Recent convergence on procedural issues in Canadian and US Merger Review • The 2009 amendments and subsequent regulations and guidance documents promote convergence between US and Canadian merger review on several levels. Specifically, important changes to merger review procedure include: o The Commissioner’s new authority to issue a supplementary information request (“SIR”), a Second Request-like tool that allows the Commissioner to demand any information or documents “relevant” to her analysis of a transaction and prohibits parties from closing until 30 days after they have certified a complete response to the Competition Bureau’s (“Bureau”) information demands.4 • The Bureau’s Merger Review Process Guidelines adopt many of the same protocols as those used by US agencies in issuing Second Requests – e.g., 2-year relevant periods for records, 3-year relevant periods for data, and a maximum number of 30 custodians.5 o Revised waiting periods to match the waiting periods under the US HSR Act – an initial 30-day waiting period following notification, with another potential waiting period preventing closing until 30 days after compliance with a SIR if one is issued.6 o Elimination of the distinction between a “long” and “short” form notification, providing for only a single form for all notifiable transactions going forward, consistent with notification practice in the US.7 o Revision of the terminology of the “6.1” regulatory provision governing the filing of transaction documents to effectively mimic the US “4(c)” document filing requirement.8 • Commissioner Aitken has also been a proponent of closer inter-agency cooperation between the Bureau and the US agencies. In cross-border 3 For an overview of the Canadian competition law regime, see Julie Soloway and Cassandra Brown, “Canada”, in Ilene Knable Gotts, ed., The Merger Control Review, 2010, pp. 68 et seq.. 4 Competition Act, §§ 114(2) and 123. 5 Competition Bureau, Merger Review Process Guidelines (2009) (“MRPGs”), §§ 3.3.1–3.3.2, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03128.html. 6 Ibid., § 123(1). 7 Notifiable Transactions Regulations, SOR/87-348, as am., § 16. That said, the Canadian and US HSR merger notification forms continue to be different in layout and in the information demanded. See note 20 below. 8 Ibid.

Calvin S. Goldman 

167  

investigations, parties are usually asked to provide waiver letters. In fact, the updated Merger Handbook states that “parties are strongly encouraged to provide foreign agencies with waivers permitting them to share information with the Bureau as early in the review process as possible.”9 • The Commissioner has also affirmed the importance of harmonizing US and Canadian remedies, stating: “[W]here we at the Bureau are satisfied that a remedy designed and/or implemented in another jurisdiction will adequately address Canadian issues, we will agree to support or use as a starting point the remedies agreed to in those jurisdictions.”10 • Thus, from a procedural perspective, Canadian and US merger review law are closer than ever before.11

III. Areas of continued procedural divergence • Despite the synchronization of many technical aspects of the merger review legal regimes in Canada and the US, institutional and cultural differences will continue to produce divergent practices by local lawyers. • First, the Canadian SIR process is more focused than the US process, which is based on a “model” Second Request.12 In the US, the agencies issue a wide Second Request, which the parties routinely negotiate down through strategies such as “quick look”13 and advanced/rolling production. By contrast, the Bureau 9 Competition Bureau, Merger Handbook, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/ eng/03295.html, § 2.1. 10 See Competition Bureau, Speaking Notes for Melanie Aitken, Speech to the Northwinds Professional Institute 2009 Competition Law and Policy Forum: Cambridge, Ontario, 12 February 2009 (“Northwinds Speech”), http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/02994.html; Blakes Bulletin, Commissioner Speaks Out on Comity and Co-operation at New York’s Harmonie Club, 17 December 2009. 11 For further discussion of the SIR reforms in Canada, the impact in Canada of the 2010 revisions to the U.S. Horizontal Merger Guidelines, and the coordination of remedies in multi-jurisdictional mergers, see Cal Goldman, Julie Soloway, and Dustin Kenall, “Contemporary Canadian Merger Review: New Challenges In Multi-Jurisdictional Transactions”, presented at the International Bar Association, Vancouver, Canada, 5 October 2010; and Julie Soloway, “Best Practices and Practical Tips for Navigating the New SIR Process”, Ontario Bar Association – Navigating Canada’s New Competition Law Regime – Practical Advice for Guiding Your Clients to Success, Toronto, Ontario, 5 November 2010. 12 See Competition Bureau, Responses to the Consultation on the Draft Enforcement Guidelines on the Revised Merger Review Process—Canadian Chamber Submission on the Draft Merger Review Guidelines (26 May 2009) (reporting roundtable discussion where the Bureau indicated that it had “rejected the U.S. approach of using a ‘model’ second request; rather each supplemental information request will be focused on specific issues and be unique to the relevant merger. This approach in part reflects the reality that the Bureau has more limited resources to manage a ‘second request’ process than its U.S. counterparts”), http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03107.html. 13 In a “quick look,” the parties will supply a response to potentially dispositive questions in the Second Request but delay full compliance, anticipating that the partial response will eliminate any concerns the reviewing agency might have and will thus render moot the need to undertake a full response.

168  Merger Control in European and Global Perspective (due in part to more limited resources) works more with parties from the very beginning to try to tailor its requests to the parties’ document and data retention practices. That said, the Bureau will not negotiate with parties on the question of what types of data or documents it considers relevant or necessary.14 It will consult only on how best to access them. • Second, timing agreements are becoming increasingly popular with the Bureau as a tool for negotiating an extension to the initial waiting period in lieu of issuing a SIR to delay closing. In the US, timing agreements are more commonly used for obtaining advance or rolling production agreements after a Second Request has been issued.15 Timing agreements can be useful to parties and the Bureau as a means of coordinating and structuring a review by providing, for example, deadlines for the Bureau to conclude its initial assessment, deadlines for parties to respond to any deficiencies noted in an information request, and commitments for rolling production or making employees available for interviews.16 However, by amending the Competition Act to provide for an initial 30-day waiting period with a potential second waiting period triggered only by the issuance of a SIR, Parliament expressed an intent to align the US and Canadian review timelines and to limit the ability of the Bureau to delay the closing of a transaction on its own.17 Accordingly, it would not be appropriate for the Bureau to extract an agreement delaying closing purely for the purposes of completing its own review (rather than for the purpose of acquiring necessary information) when the parties have complied with all the notification requirements per the statutory initial waiting period and related regulations. • Third, some miscellaneous technical details continue to differentiate the Canadian and US review process. For example, the Bureau requires parties to index their SIR responses.18 And, of course, the Canadian notification form differs significantly from the US HSR form in terms of layout and required information.19 • Fourth, enforcement authority in the US is not just divided but fragmented among several different players. MRPGs, cited above note 5, § 3.2. See, e.g., Christine Varney, “Procedural Fairness”, speech, Fiesole, 12 September 2009, http:// www.justice.gov/atr/public/speeches/249974.pdf. at 8–9 (describing purposes for which timing agreements are sought and used). 16 See MRPGs, § 3.4. 17 Section 100 of the Competition Act still permits the Commissioner to apply to the Competition Tribunal (“Tribunal”) for an interim order prohibiting closing but requires an evidentiary showing from the Commissioner that her investigation is ongoing, more time is required to complete it, and closing would substantially impair the Tribunal’s ability to impose a remedy. Where this evidence has been inadequate, incomplete, or unpersuasive, the Tribunal has refused to grant an interim injunction. See Canada (Commissioner of Competition) v. Labatt Brewing Co. (2007), 2007 Comp. Trib. 9, aff’d 2008 FCA 22. 18 MRPGs, § 3.3.4. 19 For example, the HSR form (in contrast to the Canadian form) requires the filing party to provide sales revenues by NAICS (“North American Industry Classification System”) code and to provide information regarding certain prior acquisitions occurring within the past five years. 14 15

Calvin S. Goldman 

169  

o The US Federal Trade Commission (“FTC”) and the US Department of Justice (“DOJ”) have their own clearance process for reviewing mergers, and, to some extent, different expertise in different industries. The standard the FTC must meet to obtain a preliminary injunction in court is slightly lower than that which the DOJ must meet.20 Additionally, the FTC can bring administrative hearings in which it prosecutes and decides a case, with deferential judicial review.21 o State Attorneys General can bring their own challenges to mergers, regardless of the decisions of or status of an ongoing review by the federal agencies.22 o Private plaintiffs can, and often do, also file suit under the Clayton Act.23 o Thus, whereas, in Canada, competition counsel are focused on only one enforcer, in the US, counsel must keep in mind the implications of review by several different types of enforcers with different expertise, motives, and powers. • Fifth, there are many more licensed attorneys in the US than Canada (about one lawyer for every 421 people in Canada versus one lawyer for every 265 people in the US).24 This leads to different document review models, with Canadian firms relying more on in-house articling students and US firms relying more on temporary third-party contract attorneys. • Finally, sixth, because of the prevalence of private actions in the US, the US bar is more familiar with sophisticated, complex, expensive economic evidence.25 Likewise, the US agencies (because of their deeper economic bench) are more adept at requesting and analyzing such evidence.

20 Section 13(b) of the FTC Act provides that a court may grant a 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.” 15 U.S.C. § 53(b). This is a less rigorous standard than that applied to the DOJ, which must meet the traditional common law standard requiring likelihood of success on the merits, a showing of irreparable harm if the injunction does not issue, that granting the injunction will not cause undue harm to the private parties, and that the public interest favors such relief. See FTC v. Whole Foods Market, Inc., 533 F.3d 869, 875 (D.C. Cir. 2008). 21 15 U.S.C. § 45(b). See J. Thomas Rosch, “A Peek Inside: One Commissioner’s Perspective on the Commission’s Roles as Prosecutor and Judge”, Santa Fe, 3 July 2008 (discussing FTC’s administrative hearing authority and procedures), http://www.ftc.gov/speeches/rosch/080703nera.pdf. 22 15 U.S.C. § 26. 23 Ibid. 24 See Kate Lunau, Maclean’s, Where’s a lawyer when you need one? (2 February 2009) (reporting that “Canada has 16 common law schools, the same number it had three decades ago, when the population was smaller by a third”), http://www2.macleans.ca/2009/02/02/where%E2%80%99s-alawyer-when-you-need-one. 25 In the US, private actions are available for violations of both Section 1 (joint conduct) and Section 2 (monopolization) of the Sherman Act. 15 U.S.C. § 15. In Canada, private actions are available only for criminal violations of the Competition Act and violations of Tribunal orders. Competition Act, § 36. Criminal violations of the Competition Act are principally limited to cartel conduct and bid rigging, although, prior to recent amendments, they also included resale price maintenance, price discrimination, and predatory pricing.

170  Merger Control in European and Global Perspective

IV. Opportunities for simplifying and streamlining cross-border Canadian and US merger review • The introduction of SIRs in Canada has promoted convergence; however, this has arguably come at the cost of impeding the simplification and streamlining of the merger review process. o A 2005 survey on the second request process compiled by the American Bar Association’s (“ABA”) Section of Antitrust Law reports that: (1) median compliance costs were US$3.3 million; (2) investigations lasted an average of seven months; and (3) production averaged 583,000 pages of e-mail; 555,000 pages of other documents, and 13 gigabytes of electronic data.26 Others have reported average compliance costs of between US$5 and US$10 million,27 with the most complex cases reaching US$20 million.28 o Figures compiled by a leading national law firm indicate that the average time frame for SIR compliance is approximately seven weeks, while the average cost of compliance is over C$1 million in legal fees. The quantity of documents reviewed ranged from approximately 100,000 in one case to over two million in another. Documents submitted to the Bureau also encompassed two extremes: with approximately 9,000 documents submitted in one case to well over 100,000 submitted in another.29 o Thus, given the relative sizes of the US and Canadian economies, the Canadian SIR process is on its way to matching the US second request process in complexity and cost. • Absent new legislation (in either Canada or the US), there are some limited opportunities for simplification in cross-border mergers. o Waivers: • Confidentiality waivers are now de rigueur in cross-border transactions.30 The Bureau doesn’t believe it needs a waiver to share information with other agencies, but others (e.g., the US and European agencies) require a waiver to share their information with the Bureau. • Cross-border exchanges can be of great assistance to the domestic agency. Not only can the foreign agency supplement the domestic agency’s information, but the agencies can discuss theories of the case and if one 26 Data Regarding the Burden Involved in Responding to HSR Second Request Investigations (Letter to Antitrust Modernization Commission, dated 22 February 2007), http://govinfo.library.unt. edu/amc/public_studies_fr28902/merger_pdf/070222_aba_mergers.pdf. 27 Steven Sunshine and David Wales, “Statement at AMC Merger Enforcement Hearing” (17 November 2005), http://govinfo.library.unt.edu/amc/commission_hearings/pdf/Sunshine_Statement. pdf, at 4. 28 Cecile Kohrs Lindell, Daily Deal, Majoras Hopes to Streamline Reviews (11 May 2005). 29 This does not include additional non-legal costs, such as fees for third-party experts needed to collect data to assess efficiencies, third-party e-discovery providers needed for clients unable to respond to a SIR through in-house personnel, and costs of indexing responses as required by Bureau policy. 30 See Merger Handbook, cited above note 9.

Calvin S. Goldman 

171  

agency is further ahead than the other in its investigation and thinking, the agency that is further behind may be able to benefit from the leader’s more advanced knowledge to further develop its own case. • Commissioner Aitken has repeatedly emphasized the importance of interagency cooperation, and has stressed her commitment to building upon the Bureau’s “strong partnerships within the international competition community.”31 The Commissioner has acknowledged the “enormous value” derived from integrating the Bureau’s remedies with those of other antitrust agencies and has stated that she will work closely with her US counterparts at the FTC and at the Department of Justice, to facilitate parallel investigations in merger review.32 • Indeed, the Commissioner recently stated that “Ticketmaster, Merck and Pfizer are all shining examples of how regular contact and deep trust between the US Department of Justice/Federal Trade Commission and the Bureau, on multiple levels, can lead to a settlement that works on both sides of the border and to the definite benefit of Canadians.”33 o Timing: • Now that the initial 30-day review periods are aligned in the US and Canada, it is critical for counsel to coordinate and identify issues to take off the table upfront to help avoid a SIR or Second Request. The Bureau will be increasingly attuned to the progress of a filing in the US (and any lag time between when a notification is filed in the US and in Canada), especially with regard to whether the US agencies have issued a Second Request. • In time sensitive and/or complex cases, an early (pre-filing) meeting with parties and their counsel can often be helpful for the Bureau, in terms of providing a preliminary understanding of the issues and assistance in narrowing the issues. (The Bureau’s Guidelines encourage such early consultations “prior to, or as soon as possible after, submitting a merger notification.”).34 • After a filing has been made in Canada, essentially you have a two-week time frame in which to pitch your best case to the Bureau. After that, it will start deciding whether to issue a SIR and what questions to include in a SIR. You will get a chance to consult with the Bureau on draft SIR questions to minimize your burden by revising the questions to align with the data and document retention policies of your client, but there will be no opportunity to amend the SIR on substantive aspects. • Commissioner Aitken has indicated her commitment to engage in major matters and to be accessible at key junctures in major reviews including, in appropriate cases, prior to transaction announcement, during the review Northwinds Speech, cited above note 10. Ibid. 33 Melanie Aitken, “Keynote Dinner Address”, Cambridge, Ontario, 3 February 2010, http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03205.html. 34 MRPGs, § 2.3. 31 32

172  Merger Control in European and Global Perspective process and in the final resolution of clearances or the negotiation of consent agreements. She encourages CEOs and other company decisionmakers from the parties to come and meet with her so that the key issues can be scoped out early in the process. o Coordination: • It is critical that Canadian counsel coordinate with US counsel and US economists in data analysis and document collection during the SIR/ Second Request process. For example, if the US agencies are performing diversion ratio analysis, the Bureau is probably going to want access to relevant underlying data in Canada to perform its own analysis. Delay in providing such data or analyses may mean clearance timelines begin to diverge. You do not want to have to explain to a client why they are clear to close in the US but in Canada the Bureau is still reviewing the transaction. • Notification filings do not need to take place concurrently, but parties need to take account of the process in other jurisdictions. As noted above, the Canadian and US agencies will take note of each other’s investigations and preliminary findings. However, counsel should be aware of other considerations related to coordination. For example, now that the notification requirements regarding 4(c)/6.1 documents are aligned, the benefits of a simultaneous US/Canada filing include not having to conduct a second round of document searches—in a large transaction where multiple filings are being made in several continents, this can be a substantial relief to clients.35 Also, the parties may want room to see how the issues play out in key jurisdictions such as the US and EU and then file subsequently in other jurisdictions. • Coordination can be challenging where the substantive analysis and review of a case is different than in Canada. Different jurisdictions have different substantive standards for reviewing mergers. Thus, care and caution must be exercised in ensuring positions taken in one jurisdiction do not undermine positions in other jurisdictions. For example, the new Horizontal Merger Guidelines in the US suggest that the US agencies’ approach to merger analysis may differ from most other jurisdictions (in that they could lean away from reliance on market definition).36 In response to these revisions (the first in almost 20 years), the Bureau announced in September 2010 that it would re-examine its own Merger Enforcement Guidelines (even though they were only six years old).37 A draft of the new 35 This would require carefully planning the 4(c)/6.1 document search to encompass custodians in Canada and the US, but often this is already done in cross-border transactions. 36 Federal Trade Commission and U.S. Department of Justice, Horizontal Merger Guidelines (2010), at 7 (“The Agencies’ analysis need not start with market definition. Some of the analytical tools used by the Agencies to assess competitive effects do not rely on market definition, although evaluation of competitive alternatives available to customers is always necessary at some point in the analysis.”), http://www.justice.gov/atr/public/guidelines/hmg-2010.pdf. 37 Competition Bureau, Open Consultation — Merger Enforcement Guidelines, http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/02671.html.

Calvin S. Goldman 

173  

Guidelines was published for comment in June 2011.38 Different countries also have different factual bases underlying any competitive effects analysis, such as different market structure. While these are all factual, substantive considerations, clearly they implicate procedural questions about when and where to file and how to calibrate advocacy submissions.

V. Conclusion: We face unavoidable tradeoffs in cross-border merger review • Convergence is not a goal in itself. Convergence can work against simplification and streamlining when countries are converging to a more burdensome, intrusive standard (e.g., the adoption of SIRs). • Likewise, the simplification and streamlining of the merger review process can work against confidentiality and flexibility. For example, requiring waivers can streamline a review but at the expense of confidentiality and the ability to have tactical flexibility across jurisdictions. • Finally, accountability can work against simplification and streamlining when it results in more independent review of agencies’ information requests. For example, judicial or third-party review of the scope of a SIR would enhance accountability but complicate the merger review process. • In summary, convergence is obviously an important goal—perhaps the most important goal in substantive antitrust analysis, as it is not ideal for one regulator to clear a deal and another to block it on the same set of facts. However, with respect to merger review procedures, convergence is but one of many goals, each posing tradeoffs. Procedural reforms appropriate in one jurisdiction may not be appropriate in another, as different countries have different institutional structures and different levels of expertise. Some agencies may face an accountability deficit because of lack of judicial oversight, others may not have a strong economic bench, and yet others may have such a robust private enforcement system that imbuing the government regulator with extraordinary discovery and reviewing powers may be unnecessary. • In the US and Canada, we have seen worthwhile convergence in several areas (waiting periods, notification documents, and coordination on remedies) and questionable convergence on others (the introduction of the Second Request in Canada). In the end, then, I would advocate a progressive but conservative approach to procedural reform, something like “first do no harm”. It is good to experiment with new techniques, but let’s not forget that the impact of our experiments can be tremendous. If a merger review regime is functioning well but not perfectly, it might be prudent to leave well enough alone, or, at least, to focus on incremental improvement through minor changes rather than radical surgery. 38

See http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03384.html.

V. Merger control and best practices

5 PANEL DISCUSSION

Chair: Philip Lowe

Speakers: John Boyce Rachel Brandenburger Jochen Burrichter Thomas Deisenhofer Kirsten Edwards

Cal Goldman William Kovacic Howard Shelanski Mario Siragusa Sven Völcker



328  Merger Control in European and Global Perspective

PANEL 5: Merger control and best practices Philip Lowe: This final session is about best practices. What is best, and whose practices they are, is going to be revealed to us by our three speakers. First of all, John. John Boyce: Well, I must confess that when I was asked to focus on best practices in merger control I was a little bit uncertain about what I was meant to be addressing. Was I supposed to be commenting on those guidelines that some competition agencies issue, which are labelled “best practices”? Or was I being given a bit more flexibility, to talk about what I think is the best thing competition authorities should be doing and how we, as practitioners, should do our jobs? The first thing I did, actually, was to go to the dictionary to find out what “best practices” means. I was hoping to find some sort of Shakespearian quote or something showing that it was first used by Henry II (laughter). But to my surprise, the oldest reference in the Oxford English dictionary to the use of the word “best practices” was only in 1984. That was perhaps in the days when accountants were more innovative than they now are, because it pre-dated the concentration in that industry (more laughter). According to the OED, in understanding accounting in 1984, that was the first time that the concept was used. It goes on to talk about another use of it in 1993 by John Kay, who some of you will know as an economist who wrote about corporate success.1 He explained that, if you want to run a successful corporation, one thing you might do, obviously, is to look at what other firms are doing and copy it. This strategy, he said, is more felicitously expressed as “adopting best practice”. So that’s not a bad idea. Just see what other people are doing, copy it, and that’s best practice. Well, things have moved on a little bit since then. One dictionary definition which I did quite like is from businessdictionary.com, which describes “best practices” as “methods and techniques that have consistently shown results superior than those achieved with other means, and which are used as benchmarks to strive for”. It did, however, go on to say that there is “no best practice that is best for everyone (laughter), or in every situation, and no best practice remains best for very long, as people keep on finding better ways of doing things.” Then, if you look at one of my favourite dictionary sources, Wikipedia, it’s very modern, very democratic. It comes up with similar concepts and it actually picks up some of the terminology that Bill was using earlier on. Wikipedia suggests that best practice is to be measured by its effectiveness: getting the best results. Best practice should also be about getting things done in the most efficient way, with the least amount of effort. Now, that all sounds very sensible to me, but Wikipedia actually goes on to say that there is a danger of best practices becoming a bit of a business buzz word. It says that, “as the term has become more popular, some more 1 See John Kay, Foundations of Corporate Success: How Business Strategies Add Value, OUP, 1993/1995. For additional information on Kay, see http://www.johnkay.com/about.

Panel 5 

329  

organisations have begun using the term ‘best practices’ to refer to what are in fact merely rules, causing a linguistic drift in which a new term such as ‘good ideas’ is needed to refer to what would previously have been called ‘best practices’”. I think that really sets the scene for what I think best practices should really be about. It’s all about learning from experience. That’s from your own experience, but also by looking at what other people are doing and by learning from their experiences. They might have done something wrong, so you say that’s what we want to avoid doing. It’s all about learning by doing and learning by watching. If you look at the whole area of international merger control that we’ve been talking about for the last couple of days, it’s a prime area where best practice makes sense. Increasingly, regulators are applying similar rules, and increasingly they clear similar deals, or even the same deal, so we can try to learn from what others are doing. In my paper, I try to focus on three areas where best practices make sense. The first area is best practices by agencies and by lawyers, facilitating the process of advising clients about how to get deals through. We’ve been talking about alliteration, well some time ago as a junior lawyer I learned the 6 P’s: proper preparation prevents piss poor performance (laughter). Now, increasingly that is a problem when you’re advising on international deals; where should you be notifying? I think firms are doing this regularly; you’ve got your databases, you’ve got the guides that people go to. Even on the most mundane deals, you may have to notify in a number of jurisdictions. I was just talking to Nick during the break about how, when I was a junior lawyer, the bane of our lives was being asked: “Can you review this agreement to see whether it is caught by the Restrictive Trade Practices Act in the UK?” Now, I’d have to do this and I’d think that this isn’t what I wanted to do as a competition lawyer. I wanted to deal with effects and problems. So, I think that as a legal community we were all quite pleased to see that piece of arcane legislation repealed, but what’s come along to replace that for our junior lawyers now and for the junior case handlers of the European Commission is dealing with cases that raise no competition issues whatsoever. Now, this is where I think there are some best practices out there. The ICN has been fantastic at trying to point people in the right direction by coming up with common techniques for cases, that is, both cases that shouldn’t raise problems and those that do. One of the key buzzwords for me in this area is transparency. There can be clarity as to what rules should be about, and if agencies are transparent about how their rules operate in practice that makes life much easier for us in practice. If I’ve got a couple of suggestions, my first one picks up on the point that the ICN is not a group of legislators, it’s just agencies so they don’t make any binding rules. However, I think there is still some scope for advocacy by the agencies, just making it clearer to their governments that the rules don’t make sense. It sounds as if John is having the problem the other way around at the moment, in that he’s being pushed in the direction of rules that could be more bureaucratic. I’ve talked a bit about the issue of rules, which are triggered purely by jurisdictional thresholds. I just wonder, raising the same point again, whether

330  Merger Control in European and Global Perspective there is scope for regulators – I used the word “regulator” there actually! – to exercise some self-restraint on whether to enforce the rules. Thomas, you said that you don’t have the power to do that, but to some extent you do. Let’s say a Korean company and a Japanese company set up a coffee shop in Tanzania. Say they happened to meet the thresholds and didn’t think of notifying the European Union, but someone spotted it when they were reading the newspaper in Tanzania while on holiday from the European Commission. I don’t think you would be using the Commission’s resources effectively by requiring them to file a notification. Fold up the paper and move on! (laughter) Another idea that I’ve got for the ICN is that if you hunt around the ICN’s website, you’ll find some summaries that some jurisdictions have put in, showing what their rules are, what their thresholds are. They’re pretty long, and quite difficult to understand, so I wonder whether there might be some scope for cleaning them up a bit and encouraging members to do that. You can do that behind closed doors without NGAs having to hear about it. Possibly agencies could even self-assess and spell out where the rules are ICN best practice-compliant and where they’re not. There’s nothing wrong with having rules that don’t comply with the ICN’s best practices, but it is useful for us as practitioners to know when rules don’t comply. So those are the areas where I think more can be done. A lot has been done to make life easier for us advising clients at the outset on deals. The other area of best practice is for the conduct of investigations, and we talked a lot about that in the previous session. Best practice is all about using the 3 Cs: convergence, cooperation and comity. Thomas talked about challenges, which is another C word. I like C words when it comes to competition! You were talking about the cultural challenges, the confidentiality issues – another C word that you didn’t even mention was a C word – and the challenges of communication that you have dealing with the other regulators. We heard Rachel’s more complicated MRTD acronym, but I took those letters and just threw them into the Scrabble bag and they came back as 4 Cs: mindfulness is consciousness of what the other regulators are doing; respect, well Nick said it – consideration, or courtesy or care. There you are, that’s three words (laughter). Trust is confidence – have confidence in what your other ICN members are doing; and dialogue is, of course, communication. So, there are plenty of C words out there. The last area I wanted to touch on briefly is remedies, which is an important area… Audience: Commitments! (laughter) Boyce: Yes, commitments, sorry! This is an area where, as a practitioner, it is very important to get the regulators working together because the risk of having divergent commitments can actually increase the antitrust “price” and make deals unattractive. Philip was talking right at the end of the last panel about the airlines sector, and I spent some of the past two years dealing with a transatlantic alliance case.2 I found that quite bizarre in some ways, because you have the 2

M.5747, Iberia/British Airways (14 July 2010).

Panel 5 

331  

sort of odd US practices dealing with cooperation in the airline sector, where the DOJ has got no jurisdiction at all and the Department of Transportation deals with it in an incredibly so-called “transparent” way where everything is on the docket. Also, how can you have cooperation between the Commission and the DOT under those circumstances? But you get around that somehow. The normal US approach, where orders to show cause come up quite quickly, as happened with Star Alliance and Sky Team, didn’t seem to happen in our case for some reason. That was because the Commission was cooperating so well, and not just with the Department of Transportation. I’m complimenting the Commission here, actually! (laughter) There was very close cooperation between the Commission and the DOT and there was also very close cooperation between the Commission and the DOJ. Where there didn’t seem to be close cooperation was between the DOT and the DOJ, which I found quite intriguing. I’ve stayed away from talking about best practices in the sense of guidelines because I think Jochen’s going to say a bit about that. In my paper3 I’ve tried to summarize in an annex the available best practice documents. I also commend to you, if you’re not familiar with them, the ICN’s Recommended Practices;4 I’ve summarized them in a separate annex in my paper. They’re very good and they’ve withstood the test of time, being written some five years ago. I don’t think they need to be updated much, even though one of the recommendations is that practices should be updated from time to time. The European Commission is a bit delinquent in that respect, because the Commission’s best practices for mergers pre-date the 2004 merger regulations,5 so all the references are wrong. I’ve also in an annex put down some recommended practices for lawyers, which I think we generally all try to comply with. We have guidelines like that, and I’m sure you do too, that are there to help our junior lawyers work out how you should be dealing with relatively simple cases, which may need to be notified all over the world and where you need to coordinate with local lawyers as required. To conclude, I think best practice is a great thing to strive for. If we keep on developing best practices and coming up with new, good ideas, then ultimately we will get to Utopia. But you’ve got to keep on trying to get there, because even when you think you are there, you’re not. It’s going to change, so keep on moving forward. Lowe: As school children in the North of England, we were actually taught to recite this little rhyme: “good, better, best. Never let it rest, until the good is better and the better best.” (laughter). Next we have Jochen. Jochen Burrichter: Well, it’s very difficult for me to add to what John has already said. I would like to categorize, or to analyze, the existing best practice rules which are in place. The best practices which you can observe provide guidelines 3 Boyce, “Best Practice in Merger Control: It Ain’t What You Do It’s the Way That You Do It … And That’s What Gets Results”, this volume, pp. 343 et seq. 4 http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 5 http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf.

332  Merger Control in European and Global Perspective on how merger reviews can be conducted, or should be conducted, if issued by international institutions in terms of transparency, case handling, communication, confidentiality and inter-agency coordination. If you try to categorize the existing best-practice rules, you can distinguish between the external best practices, which are issued for the benefit of merger control agencies by international institutions like the ICN, or between agencies. The second category is internal best practices, as I call them. That is, the best practices issued by the agencies themselves for the merging parties and for their own practice. The third category, which John covered at the end, is best practices issued by companies or by law firms. Let me concentrate on the two first categories. The external best practices for agencies are an important means to influence convergence. We have two categories, as I already said: firstly, the international organizations issue best practices. By far the most significant driver with regard to these best practices is the ICN, which seeks to facilitate cooperation and the exchange of know-how between competition authorities globally. The Merger Working Group published a comprehensive set of materials on best practices, which John already recommended that we should study. The most important contributions are the Guiding Principles for Merger Notification and Review6 and the Recommended Practices for Merger Notification and Review Procedures.7 Further important guidance for merger control agencies comes from the OECD. In 2005, the Council of the OECD adopted a Recommendation that touched fairness and cooperation.8 And then another document was issued by the Business and Industry Advisory Committee of the OECD; that committee published their Recommended Framework for Best Practices in International Merger Control Procedures in October 2001.9 The other important category of best practices consists of bilateral best practices, and by far the most important one is the Best Practices on Cooperation in Merger Investigations of the US-EU Merger Working Group.10 If you try to analyze the existing best practices, what is the scope? If you look at the external best practices, you see that, consistent with the rationale of external best practices, i.e., providing guidance for merger control authorities on how to shape their procedures, external best practices only establish general and broad principles for an efficient and transparent set up. These principles are usually verbalized as recommendations to agencies, therefore external best practices provide a model rather than a formal code. So you can conform to them but you are not obliged to do that. Procedural issues that are usually dealt with include recommendations on the structural alignment of notification thresholds with, for the sake of legal certainty and reduction of unnecessary transaction costs, a clear preference for revenue-based thresholds rather than market share http://www.internationalcompetitionnetwork.org/uploads/library/doc591.pdf. http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 8 OECD, Recommendation of the Council on Merger Review (2005), http://www.oecd.org/ dataoecd/3/41/40537528.pdf. In April 2012, the OECD’s Competition Committee published a report (based on the proceedings of Working Party No. 3) entitled Procedural Fairness and Transparency: Key Points. See http://www.oecd.org/dataoecd/16/34/50235955.pdf. 9 http://www.biac.org/statements/comp/BIAC-ICCMergerPaper.pdf. 10 http://www.ftc.gov/os/2011/10/111014eumerger.pdf. 6 7

Panel 5 

333  

thresholds. Moreover, the recommendations are pronounced regarding the agencies’ requirements on the timing of a notification, a clear determination of the triggering event and preferably no deadline, and the content of the notification. If you analyze the internal best practices which exist, frequently they deal with the details regarding the case team set up, the decision-making process, pre-filing guidance, communication issues – for example the timing of the Statement of Objections – guidance on the conduct of meetings, and access of third parties to such meetings. There is also relevant guidance on the right of access to the file, timing, accessible documents, entitled persons, procedures on the requirement for the initial notification, and safeguarding confidentiality with regard to the business secrets provided by the merging parties. The aims and benefits of merger best practices can be summarized very quickly. First of all, with regard to agencies, it is clear that merger control agencies benefit from the external best practices because they assist antitrust agencies with respect to a structuring of the relevant merger review procedures, therefore acting as a catalyst for better convergence. They also help to streamline the intra-agency procedures and therefore boost efficiency with respect to the merger review. The promotion of external best practices is desirable, but I think, as John said, there is still something to do. The increasing proliferation of merger control regimes and the variety of jurisdictions which affect globally active undertakings heighten the need for national agencies to publish more internal best practices on the conduct of their procedure. With regard to external best practices we have the ICN’s Recommended Practices,11 which is an excellent document that still can be improved, as John already explained. If one would try to define the minimum requirements of the internal best practices, I think it is clear that each internal best practice should deal at least with a minimum standardized set of issues which cover the most important aspects of merger control procedures. I think these are: pre-filing consultations, the scope of requests for information, the conduct of meetings, involvement of third parties, access to the file and confidentiality issues. To what extent are these best practices binding? An authority that issues such best practices normally should be bound to what they have pronounced; if there is a good reason, they may deviate from it. Another aspect that I want to mention is that in favour of the parties, or in favour of a better procedure, the binding effect should be limited. That especially applies, for example to access to file, if in appropriate cases access to file can be given earlier than you normally find in the best practices. So I would like to conclude with that and let me add that most of the proposals that John made, I would also have made. Thank you. Mario Siragusa: I will concentrate on what Jochen characterized as internal best practices, and in particular the Commission’s best practices. The Commission has made a substantial effort trying to codify and standardize its internal procedures, and there are in fact two groups of best practices: one specific to mergers,12 and 11 12

http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. Link above note 5.

334  Merger Control in European and Global Perspective a second, broader best practice for antitrust proceedings.13 On mergers, the 2004 best practices include, of course, a standard model for commitments and the trustee mandate. Then there are the more recent remedy notices,14 so that’s the block of best practices which pertain to mergers. We recently saw more best practices for antitrust proceedings published.15 There, we have of course the best practices for the submission of economic evidence16 and the guidance on the role of the Hearing Officer.17 So these are the two groups that I was talking about. It seems to me that all this documentation plays a very important and excellent role, and is inspired basically by enhancing the cooperation spirit of the relationship between the Commission and the companies involved in those various processes. That is clear, for instance, from the efforts that have been made to clarify the pre-notification process. What I find particularly important is the agreement that has been reached for the Commission to pursue fact-finding also in that process and even to talk to third parties, with the agreement of the merging entities. I think this is a very interesting development, which has clarified a very important portion of what happens during the pre-notification process. Also, I think these best practices have been inspired generally by an effort to be more transparent. The state of play meetings and the access to some of the key documents before the SO is sent all seems to me to be a step in the direction of more transparency. There have also been important decisions in the more general antitrust best practices, published more recently. In particular, I found this idea that there should be an earlier formal opening of the proceedings helpful, so that everybody knows that the proceeding is starting. Also, publications on the development of the proceeding, the issuing of the SO and the final decision, so that the opening of the procedure will be publicly announced so that everybody is informed about the progress of the proceedings. Now I would simply like to make a couple of remarks on ways in which to improve this, because I think everything can always be improved. First of all, a reflection on what the nature of all these documents is. I suppose that, particularly to the extent that they standardize proceedings that are applied by the Commission in individual cases, the Court will interpret those in the same way it has done with the Commission’s other guidelines. So we may see a development of case law 13 See now Commission Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU, 2011 OJ C308/6. 14 See Commission Notice on remedies acceptable under the Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, 2008 OJ C267/1. 15 See above note 13. 16 For the submission of economic evidence (in antitrust as well as merger cases), see http:// ec.europa.eu/competition/antitrust/legislation/best_practices_submission_en.pdf. 17  See http://ec.europa.eu/competition/consultations/2010_best_practices/hearing_officers.pdf. At the time of the 2010 Workshop proceedings, the terms of reference of the Hearing Officer were stipulated in Commission Decision on the terms of reference of hearing officers in certain competition proceedings, 2001 OJ L162/21. Those terms of reference were later expanded to some degree (e.g., during the investigative phase, or in the context of commitment proceedings under Article 9, etc.) by Decision 2011/695/EU of the President of the European Commission of 13 October 2011 on the function and terms of reference of the hearing officer in certain competition proceedings, 2011 OJ L275/29. For details, see Wouter Wils, “The Role of the Hearing Officer in Competition Proceedings before the European Commission”, 35 World Competition 431 (2012).

Panel 5 

335  

which says that a state of play meeting can take place and, if requested, must take place. So we can imagine this at least for certain aspects of these guidelines. I can imagine that there are other areas of the best practices which probably are more general and which may not be subject to judicial intervention. But this is probably something that we may see developing in the next few years. As to the possible improvements, I found that there are four areas in which I think some improvements can be made. More generally, on access to the file, I think that it should be possible to allow access from the beginning of the proceedings, both in merger cases and in antitrust cases. Once the procedure is open, and of course in the case of mergers with notification and in the case of antitrust proceedings with the formal opening of the proceedings, the file is being built. In some of our national procedures, parties have the right to access the file periodically to be informed as to the exact status of the file and the documents contained therein. This allows the party to be on a level playing field with the Commission, because when the Commission comes to know, for instance, what third parties are saying on the proposed merger, I do not see why the merging parties should not be put in exactly the same position as the Commission. This is even more important in antitrust proceedings, but it certainly could apply also to mergers. So I think that giving access to key documents before the SO is a good step in the right direction, but I propose as a next step to see whether it is possible to give full access to the file periodically during the course of the proceedings. A second suggestion relates to contacts with third parties. The Commission will of course meet and talk to third parties in the process of merger review, as in the process of antitrust proceedings. But here there is this distinction between formal and informal contacts, which I find troublesome. More generally, I think that all meetings that Commission officials have with parties – both parties to the proceedings and third parties – should somehow be recorded. There should be minutes of the meetings, summary minutes of what has been said, and this should be put in the file, with access for all interested parties. Again, this is done by some national authorities and I don’t see why this should not also be done at the Commission level. This is particularly if the results of those meetings somehow affect the Commission’s thinking and can be used by the Commission when it makes its findings. There should be some record of it in the file; otherwise there is no way it could be subject to review by the Court. The third area where improvements could be made concerns the role of the Hearing Officer. Now, I’m not sure if this can be done by way of best practices. Maybe this needs an ad hoc communication of the Commission. But certainly there were some expectations when the Commission announced the recent publication of the best practices that this could have been the occasion to somehow modify and enhance the role of the Hearing Officer. I don’t think that this has been done. The Hearing Officer’s role has been codified and explained, but I did not find anything new in the document. I’d be interested to hear whether you agree with this analysis or whether you think that there has been a change in that respect. I

336  Merger Control in European and Global Perspective certainly think that the Hearing Officer could play a more comprehensive role, not limited to procedure and access to file issues.18 My last comment is on an improvement which has been made and which probably relates more to general antitrust proceedings and not so much to merger review. It has been announced that there will be an early, official opening of the antitrust proceedings. Generally, what we have seen is that this opening is done with a very short communication, sometimes only one page or half a page as an indication of what the area is that the Commission is investigating. Not much information is given as to the scope and the real possible allegation or preliminary theory on which the opening of the case is based. To the extent that the Guidelines say that the opening of the procedure will come after an initial assessment of the case, maybe it could be useful to have a more clear indication of the nature and scope of the matter when the Commission announces the opening of the proceedings. So these are my comments on the specifics. Otherwise, I think that it is a very good initiative and I am sure that the Commission will continue to update its best practices because of the need for continual improvement. This is particularly because some of those issues covered by the best practices the Commission really have to do with due process and with the way in which the specific proceeding is going to be followed by the Commission. Thank you. Lowe: Thanks very much. John said it right at the beginning: you can’t have best practice unless you have promised to improve it regularly. For those of you who have not been to the Competition Workshop before, it has been a major driver of improvement on substantive issues and on procedural issues. The 2009 Workshop featured a barrage of criticism from some of the people around this table, but others as well, saying that the Commission didn’t do enough in this area.19 I think you can actually attribute the appearance of a document on best practice in antitrust proceedings and the submission of economic evidence and the organization of hearings precisely to the pressure you put on us on that occasion. This is just as, three or four years ago, I would say that thanks to you and to the Workshop, I was able to argue – with great difficulty – for a Guidance Paper on the application of Article 102 to unilateral conduct.20 So we have a clear indication of a cycle of debate and reaction. It’s not surprising that you don’t see a change in the precise role of the Hearing Officer, because it would have been rather odd for Neelie Kroes or myself to have left to Commissioner Almunia the test of having to do something radical. That’s up to him and his team to consider. Now, would anyone else like to comment on these best practices? 18 As seen the previous footnote, the Hearing Officer’s terms of reference were enhanced to some extent pursuant to Decision 2011/695/EU of the President of the Commission. 19 See Ehlermann and Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011. 20 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.

Panel 5 

337  

Sven Völcker: Just briefly on a couple of points. First, on Mario’s suggestion that best practices might become justiciable: I personally have my doubts, speaking under Judge Forwood’s control, of course, that the Court would ever get to that point. Looking at the past litigated merger cases, perhaps the procedural deficiencies have not been well pleaded, but they certainly have not succeeded – with the notable exception of Schneider,21 which may have been a special case. The Courts’ substantive review has been much more searching than perhaps its willingness to accept that there were certain flaws in the procedure here and there. I’m also not sure we should wish for that kind of review, because the moment it is clear that the Commission will be held to the best practices in every case by the Court, we will see a freezing of those practices and they will not evolve to the stage of being the very best. The other point is generally transparency, which I think is something the DOJ has been very outspoken on. In terms of access to file, there have been a lot of practical improvements that are very noticeable over the last year or two, down to things like an electronic index with hyperlinks to the document which saves you, in the very short time frame that you have to respond to a Statement of Objections, an enormous amount of time. I also think things like supplementary access to the file, really up until the point that the Commission needs to get close to taking its decision, are very helpful. I’m not sure we need to go to the process that the Bundeskartellamt and perhaps other national authorities have of really giving rolling access to the file. I think we all know, after having looked at these files, that 95% of it is pretty irrelevant and we would prefer in the first phase that the case team be fully focused on resolving the key issues of the case. Nevertheless, I think there is one thing that I’ve never really understood, which is that, in theory, if there are key documents in a case, you get access to them under the best practices the day after the opening of Phase II. But why do you get access to them only after the Commission has taken that decision to go into Phase II, which we all know has very significant consequences for the parties and the timing of the deal? If there is really a key document, whether it is a separate complaint or a request for information which has a 30-page response to the last question – “Do you have any concerns about this transaction?” – why not give access to that to the parties immediately? Certainly, in one case that I’ve had recently, I think that getting that information outside the framework of the best practices was extremely helpful for resolving a lot of technical issues in that case. In that vein, Rachel, I guess that the DOJ doesn’t have any best practices in terms of access to the file, because there is no file in that sense. However, I wonder whether giving access to key documents at a stage before the DOJ actually goes to court must also help the cause of transparency. Nicholas Forwood: So far as the legal status of best practice statements is concerned, there are obviously two ways that that could come before us. First of all, in the context of a plea of legitimate expectations, perhaps its limits are not quite 21

Case T-310/01, Schneider Electric SA v Commission [2002] ECR II-4071.

338  Merger Control in European and Global Perspective as broad as you might think they were, at least in relation to the sort of things that can be covered by it. What is becoming more important of course is the principle of good administration, which has now got support in the Charter of Fundamental Rights.22 Now, a person is entitled to have the administrative authorities deal with his case properly and effectively and so on. Clearly, statements as to best practice, which we have to remember is not necessarily exactly the same as good practice – best may be ideal, but good may be good enough – are clearly going to be relevant in that connection. One particular area where it may be relevant is in relation to the control of complex economic assessments in areas where there is, in some form or another, a marginal control. There, of course, as the Court reminds us in Tetra Laval,23 you verify the basic facts and so on. You also look to see, in the words of the General Confession of Sins, whether you have left undone those things that you ought to have done, or whether you have done those things you ought not to have done. In that connection, I can well see that if there is a statement in a best practice document that one ought to do X, Y, or Z, but that hasn’t been done in a particular case, then that may be the sort of thing that might be thrown in to undermine a complex assessment, to the extent that we would not otherwise be looking at it closely and forming our own view if indeed that be the case. Kirsten Edwards: In terms of submission of economic evidence, I really welcomed the best practice guidelines, but I think a lot of it is actually down to communication between the individual case team and the parties and their advisors. I’ve had very different experiences even since the best practice guidelines have been published in this respect. My best experience, which I think really demonstrated best practice on the side of the Commission and, not to be too modest, on our side too, was in Kraft/Cadbury,24 where we did a vast amount of work in Phase I. This was one of these cases where there was a merger simulation in Phase I, but the only reason we could do that much in Phase I was partly because we spent so much effort trying to pre-empt all of the questions that the case team and the Chief Economist’s Team would come up with, because we didn’t have time to go through a lengthy process of questioning. The case team was also very up front with us in that case, at every stage, on exactly what they had seen so far, what their views were at that time, and what more they needed to see to enable them to decide one way or another about it. Because of that, we did manage to get to a Phase I clearance with divestments in a way that everybody was happy with. 22 The right to good administration is provided for in Article 41 of the Charter, 2010, OJ C83/02. Article 41 provides, in its first three paragraphs, that “1. Every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions, bodies, offices and agencies of the Union. 2. This right includes: (a) the right of every person to be heard, before any individual measure which would affect him or her adversely is taken; (b) the right of every person to have access to his or her file, while respecting the legitimate interests of confidentiality and of professional and business secrecy; (c) the obligation of the administration to give reasons for its decisions. 3. Every person has the right to have the Union make good any damage caused by its institutions or by its servants in the performance of their duties, in accordance with the general principles common to the laws of the Member States.” 23 Case 13/03 P, Commission v Tetra Laval [2005] ECR I-1113. 24 M.5644, Kraft Foods/Cadbury (6 January 2010).

Panel 5 

339  

I think everyone was really proactive in that case, but I’ve been on cases where on our side we’ve just had no idea what the Commission was thinking at a particular point and I’m sure the Commission was thinking: what are these crazy economists going to put in front of them next? (laughter) So I think a large part of it is about communication, but I also think one of the challenges is actually talking to the parties themselves about best practice, because I think most people around this table and in the profession are in a lot of agreement about what best practice should be, particularly as regards economic evidence. However, when it comes to talking to your client and saying that they need to be a bit more here or there, until it comes out of the Commission’s mouth they are not willing to do anything. They are always willing to put their money where their mouth is and do as you advise, but often they think: “Why should we do this, why isn’t the Commission doing this, why is this on us?” So I think there is still a challenge in trying to get the companies who are being investigated to understand where we are coming from with the best practice. Lowe: A pre-condition for that is that the Commission has to know where it is coming from when it actually receives the original pre-notification, and starts to develop a theory or potential theories of harm and has an investigative technique which isn’t a fishing expedition. The opposite to that is when it has absolutely no theory of harm, thinks there is no problem, then it gets a key document which looks brilliant. Then it needs a non-confidential version to be able to share it with everyone else, and it’s two days before the deadline, and we can’t meet what Mario said is the right approach, which is to share key documents which are fundamental to opening Phase II. But Thomas is going to tell us about his own practical experiences. Thomas Deisenhofer: First of all, one general comment on due process, which is taken extremely seriously by Commissioner Almunia and Alexander Italianer.25 One counter-intuitive thing is that you think you can’t get enough of due process. The reality is that, depending on where you stand, there are trade-offs. There is first of all due process and speed, and effectiveness. Then there are tradeoffs between the due process of the one and the due process of the other, which is the most important thing. Confidentiality involves tradeoffs on all sides too. All these points that you have made are extremely delicate points where the difficulty is to find the right balance and to calibrate it correctly, so that you provide as much due process as you can while maintaining speed and while maintaining the rights of third parties. Depending on where you stand as a lawyer, obviously you will defend more this side than the other side, but due process is not an unlimited resource. The second thing is how seriously we take it in practice. It’s not just what is written in the rule that’s important, but what we do. I think Sven mentioned 25 See, e.g., Joaquín Almunia, “Fair process in EU competition enforcement”, SPEECH/11/396 of 30 May 2011; Alexander Italianer, “Safeguarding due process in antitrust proceedings”, Fordham speech, 23 September 2010.

340  Merger Control in European and Global Perspective key documents; it’s true that there is currently no rule that key documents before Phase I can be given, but as Philip said, on the one hand the case team has in reality 15 days from notification during which a lot of time is spent waiting for the replies to questions we try to send out the day the notification arrives. We send them out so that the companies have a bit of time to reply, then we try to coordinate inside. So there might be a document coming up, and then the internal coordination, and then on day 15 we have to speak to the parties. I think it’s right to tell the parties where we’re going, but if on top of that you then have to produce non-confidential versions of key documents, with the difficulties of getting them from third parties, and then manage possible remedies or whatever, that would be a challenge if it were to become mandatory. What is happening in practice, as you’ve rightly described, is that if it’s possible, and if the complainant has no problem with it, then directly or indirectly you can promptly identify relevant technical issues. Routinely, at least at my level, we try to give non-confidential versions of documents even in Phase I in order to sort out things because the ultimate objective is to get to the issues as soon as possible. My last point is on attitude. It’s all about attitude: attitude of the enforcer, and of the parties and third parties. If there is a give and take, and if people are reasonable, then solutions are normally found beyond the written best practices. Bill Kovacic: I have a response to Sven’s question posed to Rachel, and she can certainly address it herself as well. I think there’s a basic tradeoff inherent in the US system, that is that parties like the fact that we can’t do anything that hurts until we’re in front of a judge and a judge says so. That inhibits access to the file, and I think if we had the capacity to take the decision in the first instance it would be much more forthcoming. The tradeoff in the system design is that as long as we have an adversarial contest to resolve the matter, there is always going to be an inhibition to putting all of the evidentiary cards face up. If we had a process that resembled that of the Commission, if the original Hart-Scott vision had been in place, or if the FTC for example could take the steps that the European Commission could take, I imagine that we would have a similar bundle of disclosure safeguards ex ante, but the possibility of litigation makes it hard to be as revealing as we might be. Thank you. Rachel Brandenburger: All I would add is that, from my experience in private practice counselling transatlantically, although the procedures were different and one spent quite a lot of time explaining to clients why it worked this way and that way, I’m not aware of cases in front of either US agency where I felt that I did not understand what the theory of harm was, or what the evidence was in front of the US agencies, even if I did not have access to the file. If you have other experiences, then we need to hear about them. Howard Shelanski: You may remember the proposals that Charles James made in 2001–2002, where he said that the DOJ was willing to put things on a different

Panel 5 

341  

path if it had an earlier common revelation of information, especially in instances where a consensus might be achievable. I don’t know exactly what the experience with that turned out to be, but it was a commitment to put more information forward earlier. I hear continuing complaints as well, Sven, about whether, even without access to the file, we are forthcoming enough. I don’t know whether that’s because we don’t say enough on our side at the FTC, or whether we say it quite clearly but the parties don’t want to believe it. Cal Goldman: Just one or two quick additions. When I was running the Canadian Competition Bureau I consulted with my US colleagues about issuing news releases even when we took no-challenge decisions. I was cautioned that if we did that, then we would end up being sued. Well, we’ve come a long way. The Canadian and the US agencies now issue all kinds of information. That’s a best practice. Secondly, the US authorities, especially the FTC, should be commended for stakeholder consultations on an international scale. I suggest that what the FTC has done on global hearings, consulting the global community and not just stakeholders in the US, is a very good example of trying to achieve as much input through a form of best practice as is possible. It’s a model for other jurisdictions. Thirdly, I’d like to make a small point concerning the OECD. Peer review at the OECD does not generally include BIAC. I think it should. It would be constructive to consider including BIAC at least in the country reviews up until the point where there is any discussion of confidential enforcement actions. I don’t see any harm, I can see only good coming of it. Thank you. Lowe: Thank you very much. I think we’ll draw a line under that discussion of best practices. I just want to say one or two things in conclusion of this day and a half. In June of last year, I took an executive decision that we should discuss mergers. Not because I thought that there would be a huge amount of disagreement on many issues, but because it was an opportunity to discuss the new guidelines envisaged in the US and the UK. I think we did have a very stimulating discussion on the use of methodologies, including market definition issues and other investigative techniques. I didn’t notice a great deal of alarm, but certainly some caution about using those techniques in the right context, and recognizing too that virtually all our techniques need to complement each other to arrive at a robust result. On the standards of assessment and evidential requirements before the EU Courts, I think we’re still going to debate that quite a lot. At the moment I don’t see at least the Commission in Europe, or the EU Courts, actually moving on any particular issue of structure or procedure. Maybe that will become necessary if indeed we have another spate of Court activity with respect to certain cases, and there are some in the pipeline. On the procedural issues of convergence particularly, I thought that we would not expect to have a great deal of advance compared with the work already done in the ICN context and elsewhere, but I did see, in both yesterday’s session and this morning, continuing emphasis on evaluating performance and

342  Merger Control in European and Global Perspective indeed confronting ourselves with either the criticism of over-enforcement or under-enforcement. What that would mean in terms of ongoing work inside our agencies almost certainly needs clarification. We emphasized how important it was to build up a relationship of cooperation and confidence between those people who are actually going to tackle the same issues in the same markets. I did raise a small alarm bell, which I think will become larger, that the more our markets become global, the more pure competition will be mixed up, at least in the public perception and in the government’s perception, with other issues such resource scarcity, security of supply, and general concerns about concentration. These may take any agency, if put under pressure, beyond what it would regard as the normal standards of proof necessary to prohibit a merger. I’m just saying that this is an alarm bell, and it may be less loud if we are able to look more closely at the longer-term effects of further concentration in global markets. It comes back to the issue of what the strength of potential competition is outside any particular jurisdiction, and what the serious assumptions being made by one jurisdiction are in taking a decision in relation to the effects in its own jurisdiction. There, thank goodness that there is the Bundeskartellamt to look at BHP Billiton. With the Commission, obviously. But, without some constructive cooperation between the agencies in relation to transactions like that, I think that there would be a serious issue of government failure of addressing major transactions, which have maximum impact in practically all jurisdictions. On the last issues of best practice, we can only get better. Looking ahead, Mel and I will be thinking about next year’s subject, and we’ll undoubtedly choose a more controversial one where we can scratch each other’s eyes out on procedural issues or inter-institutional issues. I thank you very much, particularly those who have come a long way to be with us to share in this discussion. As I say, the Workshop hasn’t been as bloody as some of the previous ones. Last year, when we discussed due process there were far greater, far more powerful epithets used than anything in the last 36 hours, but those lively discussions did lead us to interesting conclusions.26 We can say that there’s a lot of progress being made, but there are some alarm bells ringing and we shouldn’t be complacent. I thank you again for your participation, and as always we will try to set an agenda next year to match the standards set by Claus Ehlermann and all of you in times gone by. Thank you (applause).

26

See Ehlermann and Marquis, Evaluation of Evidence and its Judicial Review, cited above note 19.

Andreas Mundt and Andreas Bardong*

Comity, Cooperation and International Convergence – Recent Developments in German Merger Control

I. Introduction Convergence, Cooperation, Comity.1 These are three fundamental principles for the international community of competition authorities. They are in some aspects aspirational, i.e. they mark the direction of competition authorities’ efforts rather than binding rules of black letter law and describe an evolution in the application of competition law. The implementation of these principles has come a long way, on the international as well as on the European and German levels. On the international level, the ICN has been a major success, not least in the merger area. The ICN’s membership has increased tremendously, from 14 founding jurisdictions in 2001 to more than 100 member authorities today.2 The ICN’s recommended practices regarding merger control have significantly contributed to procedural and substantive convergence. The regular exchange of experiences at all levels and in different forums of the ICN are crucial to share know-how and to build up consensus on best practices. Different working groups have created a wealth of handbooks, reports and other documents on key issues of merger control.3 The ICN documents have become important points of reference and they play a focal role in enhancing convergence. The ICN complements the excellent work done in the framework of the OECD.4 Developments on the regional level in Europe are also remarkable. Many jurisdictions have copied elements of the European Merger Control Regulation, which was in some respects inspired by German merger control. Many smaller jurisdictions in the EU have matured. Cooperation in general between national authorities is increasing, and this is also true of cooperation in the merger area. For example, national authorities are reflecting, in a working group at the EU * Andreas Mundt is President of the Bundeskartellamt, Bonn. Andreas Bardong works at the Bundeskartellamt as Head of Section for Merger Control Policy. 1 See, e.g., J. Thomas Rosch, “The Three Cs, Convergence, Comity, and Coordination”, St. Gallen International Competition Law Forum, 10–11 May 2007, http://www.ftc.gov/speeches/ rosch/070510stgallen.pdf. 2 See the ICN’s member directory, http://www.internationalcompetitionnetwork.org/members/ member-directory.aspx. 3 The ICN’s work product can be accessed on the ICN website. See http://www. internationalcompetitionnetwork.org/working-groups/current/merger.aspx. 4 To consult the OECD’s work product in the field of merger control, see http://www.oecd.org/topi c/0,3699,en_2649_34715_1_1_1_1_37463,00.html.

214  Merger Control in European and Global Perspective level, on how to develop the case information system that was established in the context of the European Competition Authorities (“ECA”) in 2001. Currently, the case information system consists of the so-called “ECA Notices”, which contain basic case information (parties, sector or products concerned, contact details of case handlers) and are exchanged among competition authorities in the EEA.5 Germany takes an active role to support this development as a vice chair of this working group. The other vice chairs are the Irish Competition Authority and the EU Commission. At a national level, there have been important developments with regard to substantive convergence. The reform of merger guidelines at the level of the EU, in the UK, US, and France shows that the analysis of markets and competitive harm brought about by mergers has converged significantly worldwide. Germany, Canada and Ireland are currently revising their merger guidelines, too. In Germany, the case practice has evolved a lot over the 10 years since the BKartA issued its last edition of substantive merger guidelines. Today there is a lot of common ground with the other revised guidelines. This paper will focus on developments in Germany with regard to comity. It is organized as follows: (i) in the second part of the paper, the requirements of the principle of non-intervention and the effects doctrine are outlined briefly, and then we discuss the related questions raised, in particular, in the Phonak/ReSound and the BHP/Rio Tinto cases; (ii) in the third part, comity and the requirements of local nexus are outlined and the introduction of a second domestic turnover threshold in 2009 in Germany as well as its impact on case load and effective enforcement are discussed; thereafter, we reflect on the consequences a further increase of the second domestic turnover threshold would have; and (iii) in the last part of the paper, we look at the prospects for further substantive convergence in the context of the next revision of the German Act against Restraints of Competition.

II. Principle of non-intervention and effects doctrine In problematic multi-jurisdictional merger cases, parties sometimes raise the question of whether conducting a merger procedure in a foreign-to-foreign merger or intervening against the merger is in line with public international law.

5 The “ECA Principles” concerning merger case referrals (January 2005) are available at http:// www.bundeskartellamt.de/wDeutsch/download/pdf/Merkblaetter/Merkblaetter_englisch/ECA_ Principles.pdf.

Andreas Mundt and Andreas Bardong  215  

1. Scrutiny of mergers In the Phonak/ReSound case,6 for instance, the merging parties’ lawyers argued that the centre of gravity of the merger was not in Germany but abroad. The companies’ main headquarters are in Denmark and Switzerland and the parties claimed that their turnover in Germany amounted to less than 10% of their total turnover. Manufacturing sites are mainly outside of Germany and the merger was notifiable neither in Denmark nor in Switzerland. On this basis, the parties’ lawyers claimed that public international law, and more precisely the principle of non-intervention, precluded the exercise of jurisdiction over the merger in Germany.7 This raises the question of whether merger control must be limited to countries where the parties or targets have their headquarters, their legal place of incorporation, or manufacturing sites. It also raises the question of whether merging parties’ turnover has to exceed 10% or maybe 50% of total turnover in any jurisdiction that intends to apply its merger control procedures to the transaction. The correct answer to all these questions is clearly no. Most jurisdictions, courts and scholars agree with the effects doctrine.8 It is also incorporated into German law in § 130 Section 2 of the Act against Restraints of Competition, which provides that the Act “shall apply to all restraints of competition having an effect within the area of application of this Act, also if they were caused outside” this area. Accordingly, it is sufficient that a merger has a relevant impact on the territory of the intervening state, no matter where its premises are. Therefore, the question is whether the merger has an appreciable effect on markets that cover all or part of the state, for example if the companies have more than de minimis turnover in that state. This was also the approach of the Higher Regional Court of Düsseldorf in the Phonak/ReSound case.9 The Higher Regional Court of Düsseldorf also confirmed that the application of the German merger control rules not excluded even if the parties’ claim were correct that the merger’s centre of gravity lies outside of Germany.10 This is an appropriate 6 See Bundeskartellamt, Decision of 11 April 2007, WuW, DE-V 1365 – Phonak/ReSound; OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477 – Phonak II; BGH, judgment of 20 April 2010, KVR 1/09, WuW, DE-R 2905 – Phonak/GN Store. 7 OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477, 2483 – Phonak II. 8 See, e.g., BGH, judgment of 12 July 1973, KRB 2/72, WuW/E BGH 1276, 1279 – Ölfeldrohre; BGH, judgment of 29. May 1979, KVR 2/78, WuW/E BGH 1613, 1614 – Organische Pigmente; BGH, judgment of 25 September 2007, KVR 19/07, WuW/E DE-R 2133 – Sulzer/Kelmix para.18; OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477, 2485 – Phonak II (reversed on other grounds: BGH, judgment of 20 April 2010, KVR 1/09, WuW, DE-R 2905 – Phonak/ GN Sto­re). See also Case T-102/96, Gencor v. Commission [1999] ECR, II-753 paras. 89–92, and Baetge, Globalisierung des Wettbewerbsrechts, (Mohr Siebeck, 2008), 268 et seq. 9 OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477, 2485 – Phonak II. On this issue, the decision was indirectly confirmed by the Bundesgerichtshof, the German Federal Court of Justice, BGH, Judgment of 20 April 2010, KVR 1/09, WuW, DE-R 2905 – Phonak/ GN Sto­re. The decision of the OLG Düsseldorf was reversed on other grounds. 10 OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477, 2484 – Phonak II. The Court left open the factual question of where the transaction’s centre of gravity was indeed located.

216  Merger Control in European and Global Perspective approach to the treatment of international mergers in domestic jurisdictions. For these types of transactions it is typical that merging parties’ turnover is spread over many countries, often with no single country accounting for 50 percent or more of one merging party’s turnover. It would clearly be an awkward result to exempt from scrutiny the biggest transactions that may have an impact in many countries worldwide. Neither would it be appropriate to limit merger control to the countries where the highest turnover ratio in the typical multi-jurisdictional cases normally accrues; globally, these are usually the United States, but in cases with a more regional ambit, a relevant jurisdiction is often Germany. In line with these observations there are several factual elements with regard to the Phonak/ReSound case that should be noted: (i) the largest market for hearing aids after the United States was Germany; (ii) the merging parties achieved higher turnover in Germany than in the countries where they have their headquarters, i.e. Denmark and Switzerland; and (iii) 20 percent of European turnover was realized in Germany, which accounts for 70–100 million euros. The BHP/Rio Tinto case11 is also a good example to show that the effects doctrine is the right approach. Following intensive investigations and the expression of serious concerns by the Commission, Germany and Japan, the parties withdrew their notification. The merger was also notified in Australia, Korea, China and Taiwan. Essentially, it would have been a three-to-two and a two-to-one merger. The merged entity would have combined the numbers two and three in the production and sale of seaborne iron ore fines, rendering the merging parties the number one supplier. With regard to iron ore lumps the merger would have combined the only two significant suppliers. The markets are worldwide. Not included are regions that are close enough to mines so that they can be supplied by land transportation, for example the US. German customers are mainly supplied by BHP’s and Rio Tinto’s Brazilian rival Vale. Therefore, the vast majority of the parties’ turnover with the products concerned is achieved outside of Germany. In the BKartA’s opinion, it was in line with public international law and principles of comity to closely examine this merger because it had a sufficient impact on Germany. The merger would have had an effect on competition and on prices on the relevant world markets for seaborne iron ore fines and lumps, which included Germany. In addition, the transaction had a sufficient nexus to Germany, measured inter alia by turnover achieved by each of the merging parties in Germany, i.e. turnover which exceeded (in this case by far) the relevant notification thresholds of 25 million and 5 million euros. With regard to the BHP/Rio Tinto case one could ask the same questions that were raised by the merging parties in the Phonak/Resound case: (i) whether on the basis of international law and principles of comity it would have been preferable to leave the case to the jurisdiction where most of the assets of both companies 11 Bundeskartellamt, Case B1-10/10 BHP/Rio Tinto, Case summary of 24 January 2011, “Creation of a joint venture between BHP Billiton and Rio Tinto abandoned after concerns expressed by the Bundeskartellamt”, http://www.bundeskartellamt.de/wEnglisch/download/pdf/Fallberichte/B01-01010-english.pdf?navid=42.

Andreas Mundt and Andreas Bardong  217   are located and where their headquarters are, i.e. to Australia; and (ii) whether the case should have been dealt with in the jurisdiction where the major and most direct impact of the merger was expected, i.e. in China, the country that absorbs most of the exports of iron ore from Australia. Again, the correct answer to these questions is no. BHP/Rio Tinto shows clearly that a worldwide one-stop-shop for global mergers is unrealistic. All countries that are affected by a merger need to have a chance to review it, even where worldwide markets are concerned. If prices will rise after and because of the merger, this will affect the whole market. Where only national markets are affected, like in the Phonak/ Resound case, it is even more obvious that the review cannot be limited to one central authority. Competition authorities from third countries will normally not be in a position to conduct investigations and remedy competition concerns. They usually lack the power to investigate and intervene against competitive harm outside their own jurisdiction. What is equally or maybe even more important, they also lack the interest and incentives to do so. This view on the effects doctrine and merger control is shared by many jurisdictions worldwide. For example, a former Deputy Assistant Attorney General at the US DOJ phrased it as follows: “we have come to accept that in a global economy, conduct will often have effects in multiple jurisdictions and that each jurisdiction has not only the right, but the obligation to protect its own consumers”.12 Therefore, it is not surprising that the BHP/Rio Tinto merger triggered several review procedures in Asia, Australia and Europe. With respect to effects in the United States, the case is atypical. The merger did not raise any competition issues in the United States because seaborne iron ore is not consumed there given that this input can be sourced domestically from landborne sources. In conclusion, looking at multi-jurisdictional mergers, it seems unavoidable both from a policy and a legal perspective that many competition authorities sometimes take a look at the same case.

2. Prohibition of mergers and balancing of interests In the Phonak/ReSound case, the merging parties’ lawyers argued, inter alia, that Germany was barred from prohibiting the merger on the basis of international law because it had been cleared in other countries worldwide (in particular US, Brazil, Norway, Spain and Greece) and was not subject to merger control in Denmark and Switzerland, where the merging parties have their headquarters. They claimed that this amounted to an important interest that had be balanced against the interests of the jurisdiction that aims to intervene against a merger to prevent competitive harm in its territory.13 12 See William Kolasky, “International Comity in Antitrust: Advances and Challenges”, Legal Backgrounder, Washington Legal Foundation, 22 (2007), No. 16 (May 25), p. 4, http://www.wlf.org/ upload/05-25-07kolasky.pdf. 13 See also Andreas Weitbrecht, “Völkerrecht und Kollisionsrecht in der deutschen Fusionskontrolle,

218  Merger Control in European and Global Perspective It is recognized that public international law does not preclude foreign-toforeign transactions from being prohibited or cleared only subject to commitments, in order to protect competition and to prevent competitive harm within the State concerned.14 Provided there is a sufficient link of the merger to the intervening State, this also applies to cases where a merger cannot be divided into separate parts and the prohibition thus effectively blocks the merger worldwide and has an impact on other countries.15 Whether public international law limits the sovereign right of a state to protect its markets and consumers against competitive harm is debated,16 but in any case this is limited to extreme situations where an intervention against a merger would interfere with important interests of other states. Only in such cases, if any, does a balancing of interests seem to be required under public international law.17 Interests that could trigger the balancing test need to relate to vital interests of another State. Depending on the circumstances of the case, including the size of the economy and the relative size of the companies concerned, for example a massive loss of employment or the lapse of a major company are situations in which it cannot be excluded that the effect on such a State would be sufficiently severe to interfere with its vital interests. In this case, an important question would also be how the blocking of the merger versus the implementation of the merger furthers this interest, and whether any alternatives are available, for example alternative purchasers. Another scenario that appears to be possible, yet exceptional, relates to mergers that are necessary for national security reasons. For example, a particular merger that raises competition concerns in third countries may be necessary on the basis of national security interests if access to certain vital military technology could only be secured in the future for that State by way of an acquisition of a failing company by a competitor, assuming that the latter company is the only viable buyer within the State concerned. Zur Untersagung von Auslandszusammenschlüssen durch das Bundeskartellamt”, in Festschrift Birk (70. Geburtstag), (Tübingen, 2008), 977, 990. 14 See, e.g., OLG Düsseldorf, judgment of 26. November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477 – Phonak II Eckard Rehbinder in Immenga/Mestmäcker, eds., GWB, 4th edition (München, 2007), § 130, paras. 126 et seq. and paras. 134 et seq; Kurt Stockmann in Loewenheim/Meessen/ Riesenkampff, GWB, 2nd edition (München, 2009), paras. 47 et seq. 15 See OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 (V), WuW, DE-R 2477 – Phonak II; Rehbinder, cited previous footnote, § 130, paras. 126 et seq., 134 et seq., 279 et seq.; Stockmann, cited previous footnote, para. 47 et seq.; Burkhard Richter in Wiedemann, Handbuch des Kartellrechts, 2nd edition (München, 2008) § 19 para. 17, contra Henning Leupold and Sven Timmerbeil, “Die zivilrechtlichen Auswirkungen fusionskontrollrechtlicher Vollzugsverbote auf deutschem Recht unterliegende M&A Transaktionen”, (2007) EWS 155–159. 16 See, e.g., Stockmann cited previous footnote, para. 47; Thilo Reimers, Probleme und Perspektiven der Internationalisierung des Wettbewerbsrechts (Kiel, doctoral thesis; published by Nomos, 2007), 53 et seq. (only the non-binding principles of comity apply); Stephan Barthelmeß and Lars-Peter Rudolf, “Extraterritoriale Anwendbarkeit der deutschen Fusionskontrolle und das völkerrechtliche Abwägungsgebot”, (2003) WuW, 1176, 1178 et seq.; Stadler in Langen/Bunte GWB, 11th ed. (Munich, 2010), § 130 para.194; Weitbrecht, cited above note 13, 977–991. 17 See, e.g., Stockmann, cited above note 14, para. 47.

Andreas Mundt and Andreas Bardong  219   Clearances in other States will normally not carry too much weight in the balancing process, if any.18 The different outcome in the intervening state and the state(s) that have cleared the merger can have many reasons, in particular that the merger causes competition problems only in the intervening State, for example because the markets affected by it are national in scope and market conditions in the reviewing jurisdictions are very different. More importantly, the interests of the merging parties to consummate the merger cannot be equated with the interests of other States reviewing the merger or of the states where the merging parties are based. Moreover, it is clear that vital interests are not at stake when a merger simply does not meet the notification thresholds. In the rare situations where vital interests of other states are indeed at stake, the enforcement of competition law, and the competitive harm expected as a result of the merger in the intervening State would then have to be balanced against such interests. While it is not unusual for vital interests of foreign States to be claimed in merger proceedings and in challenges before courts, in practice it appears to be very exceptional that such overriding interests genuinely exist; more often, the interests claimed are in substance more closely related to the commercial interests of the merging parties. However, the examples mentioned above show that there may be such cases. For example, in the scenario of a merger in the defence industry, it may well be that national security interests, if indeed imperilled, could trump the competition interests in a third country also affected by such a merger.

III. Comity and local nexus in Germany – change in legislation (MEG III) Looking at multi-jurisdictional mergers from the perspective of comity,19 the requirements of the principle of non-intervention and the effects doctrine apply mutatis mutandis. However, what is required for a genuine link between the merger and the jurisdiction, i.e. for a local nexus, probably extends a little bit further on the basis of comity than on the basis of the aforementioned legally binding principles. It should be noted, though, that comity is not yet a binding and justiciable principle of law, but a customary principle of international cooperation. It describes expectations and best practices recognized by the wide majority of states, for example in the area of competition policy. A good proxy for the soft law rules of comity are probably the ICN Recommended Practices for Merger Notification Procedures.20 They provide that “Jurisdiction See, e.g., Rehbinder, cited above note 14, § 130 para. 280. See, e.g., Eva Buchmann, Positive Comity im internationalen Kartellrecht (Frankfurt, doctoral thesis; published by Sellier European Law Publications, 2004), 41 et. seq. 20  ICN Recommended Practices for Merger Notification Procedures, http://www. internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 18 19

220  Merger Control in European and Global Perspective should be asserted only over those transactions that have an appropriate nexus with the jurisdiction concerned.”21 The level of local nexus should be “material” and should be “based on activity within that jurisdiction, as measured by reference to the activities of at least two parties to the transaction in the local territory and/ or by reference to the acquired business in the local territory.”22 This was not the case in German merger control rules prior to March 2009, but the ICN’s Recommended Practices (“RPs”) also included exceptions that covered the situation in Germany. The RPs explained that jurisdictional thresholds that apply only to one party should be “set at a very high level” or, if not sufficient, “other objectively limited filters” should be adopted.23 In essence, these requirements translate the principle of comity into practical rules for the design of national jurisdiction in merger control. Although Germany was covered by the exceptions in the Merger RPs, the recent changes mean progress in respect of comity, because the local nexus requirement is now fulfilled more clearly. In March 2009, Germany introduced a second domestic turnover threshold.24 This has increased the emphasis on the local nexus of merger cases to Germany. In 2009, the number of mergers notifiable in Germany has decreased by about 40% in comparison to the previous year, also due to the financial crisis. Although it is difficult to accurately allocate this outcome to the change in legislation versus the downturn of the economy, it is probably fair to say that the effect of the second domestic turnover threshold is likely to account for a decrease of cases by at least one third.25 Possibly, the overlap between the effect of the financial crisis and the effect of the change of legislation is even broader. As a consequence of the new legislation, many foreign-to-foreign mergers now fall outside the scope of German merger control. The ICN Merger RPs26 were an important argument that helped to convince the German lawmaker that this change was the right step. After the changes it seems that Germany is in “compliance” with almost 100 percent of the RPs. This is remarkable, given that many countries retain market share thresholds as a notification threshold and that many others that employ a domestic turnover threshold do not require local turnover of more than one company. See ibid. I. A. Ibid. I. B, C. Ibid. I. C, Comment 4. 24 “Drittes Gesetz zum Abbau bürokratischer Hemmnisse insbesondere in der mittelständischen Wirtschaft (Drittes Mittelstandsentlastungsgesetz)”, dated 17 March 2009, BGBl. 2009, Part 1 No. 15, p. 550, http://www.bmwi.de/BMWi/Redaktion/PDF/Gesetz/meg-3,property=pdf,bereich=bmwi,sprac he=de,rwb=true.pdf (in German). 25 This is in line with the prognosticated impact described in the explanatory memorandum to the bill. See BT.-Drs. 16/10490, dated 7 October 2008, p. 15 (left column), http://dip21.bundestag.de/ dip21/btd/16/104/1610490.pdf (only available in German). See also Andreas Bardong, “Die zweite Inlandsumsatzschwelle, kein Änderungsbedarf!”, (2011) WuW, Heft 04,  pp. 350–359; Alexandra Georgi, Die doppelte Inlandsumsatzschwelle des § 35 I Nr. 2 GWB (Heidelberg, doctoral thesis) (publication forthcoming). 26 ICN, Implementation of the ICN Recommended Practices for Merger Notification and Review Procedures, http://www.internationalcompetitionnetwork.org/uploads/library/doc324.pdf, p. 2. 21 22 23

Andreas Mundt and Andreas Bardong  221   The law requires a turnover of 25 million euros for one company and, since 2009, 5 million euros for another company involved in the transaction. Sometimes it is asked why Germany did not adopt a higher second domestic turnover threshold.27 In this context it is helpful to draw on the experience within the ICN with merger thresholds. In 2008, the ICN issued a report on “Setting notification thresholds for merger review”,28 which outlines the experience of 21 countries in fine-tuning the jurisdictional thresholds. While the report focuses on the requirements of local nexus, it also contains some thoughts on general objectives. On the one hand, the thresholds should aim to ensure that merger regimes catch as many cases as possible and manageable that raise competition concerns.29 On the other hand, thresholds should be designed to minimize the number of transactions that must be notified and that are unlikely to raise competitive concerns.30 This means in practice that neither the authority nor the merging parties should be overburdened with unproblematic cases. As a further factor, the report recognizes that the tradeoff between casting the net wider or narrower also depends on the speed of proceedings and the amount of information required in standard cases.31 On this basis, Germany strikes the right balance. First, the information requirements for merging parties are extremely low. (This is recognized in the respective ICN reports on information requirements32 and notification thresholds.33) Second, the BKartA’s merger procedures are very fast. In most cases, there is no prenotification phase and filing of a draft is rarely required. In addition, the review is completed in many cases in less than the one month time limit for first phase cases. Therefore, the burden on business and on the authority’s administrative resources is manageable and reasonable despite the high volume of merger cases (currently, about 1,000 a year). A similar approach is taken by the US and Canada: low initial information requirements, but high filing numbers. This was also recognized in the ICN report on the setting of notification thresholds.34 Third, German law requires a 27 See, e.g., Dirk Schroeder and Andrea Edeler, “Reformbedarf in der deutschen Fusionskontrolle”, (2008) ZWeR, 364, 370 et seq. 28 ICN, Setting Notification Thresholds for Merger Review, Report to the ICN Annual Conference in Kyoto, Japan (2008), http://www.internationalcompetitionnetwork.org/uploads/library/doc326.pdf. 29 Ibid., p. 4: “Invariably, objective thresholds cast a very wide net to catch the few transactions that merit a closer review. Nonetheless, it is important that thresholds are set at a level calculated to minimize the number of transactions that must be notified that are unlikely to raise competitive concerns, without allowing transactions that do raise concerns to fall outside the notification requirement.” (emphasis added) 30 See ibid. 31 Ibid., pp. 4 et seq. 32 ICN, Information Requirements for Merger Notification, Report for the ICN Annual Conference in Zurich, Switzerland (2009), http://www.internationalcompetitionnetwork.org/uploads/library/ doc328.pdf, pp. 18–20. 33 ICN, Setting Notification Thresholds for Merger Review, cited above note 28, at p. 5. 34 Ibid. (“In jurisdictions that require minimal initial information from the parties, the impact of relatively low thresholds that may result in a larger number of notifications may be reduced because the total cost and burden imposed by such a system are relatively low. Canada, Germany, and the United States are well-known examples of such systems. Norway is another example of a jurisdiction that seeks to balance low notification thresholds that lead to a high number of notifications with very limited initial notification requirements.” “The United States and Germany are also examples of systems that can quickly and without additional information clear transactions that obviously do not

222  Merger Control in European and Global Perspective mandatory worldwide turnover threshold. Many other jurisdictions do not, or do not in all cases. Compared to other jurisdictions, the German threshold, which amounts to 500 million euros, is very high. Looking at the major jurisdictions worldwide, only the EU Merger Regulation foresees a higher mandatory worldwide threshold, for obvious reasons. In India,35 Switzerland,36 China37 and Poland38 the worldwide thresholds are higher than in the German system; however, in these jurisdictions the worldwide thresholds are not necessary requirements but only one alternative to trigger the notification obligation.39 Fourth, in German law, it is the worldwide threshold, not the domestic turnover threshold, that is used to filter out transactions with insufficient economic importance. Therefore, it is not surprising that the domestic turnover thresholds in some other countries without worldwide thresholds are much higher, for example in Italy, Spain, and Portugal.40 The function of Germany’s domestic turnover thresholds is to make sure that the cases have a sufficient local nexus. This is an important element of comity. Finally, the BKartA also evaluated the turnover data from past merger cases.41 The figures show that the size of “collateral damage”, i.e. the damage of failing to detect cases that raise significant competition issues, appears to be small after a second domestic turnover threshold of 5 million euros was introduced. Looking at second phase proceedings between 2003 and 2008, about 10% of cases would not have been reportable under the new regime. Raising this threshold would significantly increase the number of mergers that have a potential to cause competitive harm but fall outside the scope of merger control. Increasing the second domestic turnover threshold to 25 million euros would have missed about one third of second phase cases. This would clearly be too high a price to pay for a further reduction of notifications. Looking at the number of interventions by the BKartA (i.e., prohibitions or clearances with commitments) during the same period, which is probably an even better proxy for competitive harm, confirms this assessment. The threshold of 5 million euros would have excluded less than 5% of intervention cases from scrutiny, while a threshold of 25 million euros would have increased this figure to an unacceptable level of around 30%. raise competition concerns. This, again, will influence the perception of whether or not a notification system is considered unduly burdensome.”) 35 4.6 billion euros. See Merger Control 2011 (Getting the Deal through), Global Competition Review (London, 2010), p. 195. 36 1.6 billion euros. See ibid. 37 1.1 billion euros. See ibid., p. 93. 38 1 billion euros. See ibid., p. 298. 39 For a general introduction to the worldwide thresholds, see ibid. 40 Italy, 472 million euros; Spain, 240 million euros; Portugal, 150 million euros. See ibid., pp. 218, 357 and 303. 41 This was supported by an academic study that was prepared by a researcher from the University of Heidelberg on the basis of data accessed on site at the Bundeskartellamt. The full study will be available as Alexandra Georgi, Die doppelte Inlandsumsatzschwelle des § 35 I Nr. 2 GWB, cited above note 25. Some core data results of the study were analyzed by one of the authors of this paper in Bardong, cited above note 25.

Andreas Mundt and Andreas Bardong  223  

IV.

Cooperation and Convergence

If multi-jurisdictional merger review is inevitable in many transactions with a transnational dimension, it seems that close cooperation between the authorities, in particular in parallel in-depth investigations, is important to ensure fully informed decision making and to avoid inconsistent results. In the past, the BKartA has had very intensive and productive contacts in parallel in-depth investigations mainly with US authorities, in particular the FTC, for example in the GE/Invision case,42 which concerned X-Ray systems for nondestructive testing and was cleared subject to commitments by the FTC and the BKartA. Over the last couple of years, contacts with other authorities in parallel investigations have also intensified. For example, in the BHP/Rio Tinto case there were also close contacts with the Japanese FTC. Of course there was also close cooperation with the EU Commission, which investigated the case from an antitrust perspective. In both instances, collaboration was greatly facilitated by waivers provided by the merging parties. The timing of procedures at the EU level and in Germany was to a great extent synchronized, which helped the process a lot. For example, questionnaires by the two authorities were integrated and the Commission distributed them for both of them. Parties were asked to send their responses directly to both authorities. This also facilitated the investigation for all the addressees of the questionnaires. The synchronization did not extend to notifications in all countries concerned. For example, the notification in Japan was filed much later. This example shows very well that making the procedures compatible and enabling authorities to cooperate is crucial, and probably more significant than achieving procedural convergence. Procedures and institutions depend a lot on the local circumstances and conditions and have to fit into that context.43 Different models can be equally effective. In contrast to convergence on procedural matters, convergence on substantive issues is probably more important and also easier to achieve. It seems that in the area of merger control a high degree of convergence is already a reality in many jurisdictions worldwide and on many issues, in particular with respect to horizontal mergers. There is wide consensus on the factors for the analysis of markets. Some differences remain, such as for example whether an efficiency defence is recognized; but most of them are rarely decisive for the concrete outcome of a case. Achieving convergence should not mean that the further development of competition law is stopped or that the competition for good ideas should come to an end. Convergence does not amount to harmonization, and it is not an end in itself. 42 Bundeskartellamt, Decision of 17 August 2004, B7-65/04, WuW DE-V 965 - General Electric/ InVision. 43 See, e.g., Rosch, cited above note 1, pp. 7 et seq.

224  Merger Control in European and Global Perspective The last round of guideline revisions at the level of the Commission, in the UK,44 United States,45 and in France46 show a lot of commonality. There are many elements in those newly revised guidelines that would also accurately describe the BKartA’s approach to merger cases, and to the approaches of many other authorities. However, in several guidelines there are also elements that are not mainstream. To pick one prominent example, the new US guidelines have incorporated an upward pricing pressure test (UPP) in connection with a somewhat diminished role of market definition.47 This is a new tool that will have to be tested in case practice. Whether or when other competition authorities will include this instrument in their toolbox remains to be seen. In the ongoing revision of the Bundeskartellamt’s guidelines,48 the UPP is currently not included. The authority recognizes that market definition and the assessment of competitive effects are closely related. However, the BKartA will continue to work with market definitions as a conceptual framework in all merger cases, not least in order to be in a good position to convince the courts that all necessary analytical and investigative steps have been taken.49 Arguments and tools that do not depend on a prior definition of the market will probably also be used, but more as add-on arguments. The new draft text of the German merger guidelines50 takes account of the significant changes in the BKartA’s case practice over the last 10 years. The draft guidelines move away from the checklist approach and frame the analysis more in the context of the different theories of harm. In the field of coordinated effects, the guidelines can now refer to the E.on/Eschwege judgment of the German Federal Court of Justice,51 which has incorporated the Airtours criteria.52 Cases are analyzed according to the categories of horizontal, vertical and conglomerate mergers. For vertical mergers, the guidelines follow the approach of the EU Commission and 44 UK CC and OFT, Merger Assessment Guidelines, September 2010, http://www.competitioncommission.org.uk/about_us/our_organisation/workstreams/analysis/pdf/100916_merger_ assessment_guidelines.pdf. 45 U.S. FTC and DOJ, Horizontal Merger Guidelines, 19 August 2010, http://www.ftc.gov/ os/2010/08/100819hmg.pdf. 46 Lignes directrices relatives au contrôle des concentrations, 16 December 2009, http://www. autoritedelaconcurrence.fr/doc/ld_mergers_final.pdf (English-language version). 47 See U.S. Horizontal Merger Guidelines, 19 August 2010, p. 24, http://www.ftc.gov/ os/2010/08/100819hmg.pdf; Joseph Farrell and Carl Shapiro “Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition,” 10(1) B.E. Journal of Theoretical Economics, Article 9 (2010). SSRN version at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=1313782. 48 Draft guidelines on substantive merger analysis were published for comment on 21 July 2011. See http://www.bundeskartellamt.de/wEnglisch/download/pdf/20110721-Guidance_Merger_Control. pdf. 49 The approach of the US horizontal guidelines with regard to market definition has been criticized, for example by the District Court for the Southern District of California, Decision of 22 February 2011, FTC v. Laboratory Corporation of America, et al., C.D. Cal., No. SACV 10-1873 AG (denying preliminary injunction). The FTC initially challenged the District Court’s decision before the Ninth Circuit but later withdrew the appeal. 50 See above note 48. 51 BGH, Judgment of 11 November 2008, KVR 60/07, WuW, DE-R, 2451 – E.ON/Stadtwerke Eschwege. 52 Ibid., 2458.

Andreas Mundt and Andreas Bardong  225   of the UK authorities to focus on input and customer foreclosure, as well as on access to sensitive information of competitors that are customers upstream or downstream. In horizontal mergers the BKartA’s approach had already moved away from an emphasis on market shares a long time ago. The guidelines make clear once again that market shares are only the starting point for the analysis and that other important factors such as barriers to entry and countervailing buyer power play an important role. Substantive convergence must not be misunderstood as a straitjacket that hinders new ideas and techniques. Innovation and test cases with new concepts should always be encouraged. Sometimes it can also be a step forward to look at existing concepts that have not reached the level of mainstream thinking yet. For example, the acquisition of non-controlling minority interests53 seems to be a promising field where the longstanding practice in Germany may offer some ideas that could be ripe for a broader audience in the short to medium term. Noteworthy in that regard are the new US horizontal merger guidelines, which have dedicated a new chapter to the competitive effects of minority interests.54 With regard to the substantive test for merger control, Germany might also at some future point move from a dominance test to a Significant Impediment of Effective Competition (“SIEC”) test.55 As the Commission has applied the SIEC test, an SLC type test with dominance as one example, for more than 5 years, it becomes more likely now that Germany will adapt this test, too. However, there are important players outside the BKartA who are very sceptical, and who suggest that a change would entail too many risks for companies, in particular with regard to predictability, supply of data and transaction costs. A revision of German competition law is on the way, but it is not yet entirely certain whether or not a change of the substantive test will materialize. From the perspective of the BKartA there are pros and cons for both alternatives. For the BKartA, the issue that is more important for its case practice as a whole is to retain a procedural context, in which the test is applied, which ensures that in practice competitive harm can be proven in court proceedings.56 The procedural rules need to provide sufficient incentives for the merging parties to supply the 53 Mandatory ex ante merger control also covers the acquisition of non-controlling minority shareholdings (§ 37 section 1 No. 3 and No. 4 Act against Restraints of Competition (GWB)). Among the transactions that must be notified are acquisitions of 25 percent or more of the capital or voting rights in a company, or acquisitions of an interest below this level that enables the acquirer to exercise a “competitively significant influence” on another undertaking. This is a concept that is very close to the UK doctrine of “material influence”. 54 See U.S. Horizontal Merger Guidelines, 19 August 2010, pp. 33 et seq. (Section 13), http://www. ftc.gov/os/2010/08/100819hmg.pdf. 55 See German Coalition Agreement, “Wachstum, Bildung, Zusammenhalt, Koalitionsvertrag zwischen CDU, CSU und FDP”, 17. Legislaturperiode, 26 October 2009, p. 18 (“In addition, some elements of European Merger Control will be adopted.”–“Darüber hinaus werden Elemente der europäischen Fusionskontrolle übernommen.”), http://www.fdp-bundespartei.de/files/363/091024koalitionsvertrag-cducsu-fdp.pdf (in German). 56 Bundeskartellamt, “Marktbeherrschung und SIEC-Test, Eine Bestandsaufnahme” – Conference Arbeitskreis Kartellrecht 24 September 2009 (Discussion paper on Dominance and SIEC-Test), pp. 11–14 and 17, http://www.bundeskartellamt.de/wDeutsch/download/pdf/Diskussionsbeitraege/09_ Proftagungspapier.pdf (in German).

226  Merger Control in European and Global Perspective BKartA with all the necessary facts and documents within the short time periods in which merger control investigations have to be completed and without the strict requirements for a complete filing to the EU Commission outlined in its Form CO, which sets out an extensive amount of data that has to be provided to the authority upfront, in practice already during pre-notification contacts that can extend over a considerable time span.

V. Conclusion The introduction of a second domestic turnover threshold in Germany brings the merger regime even more closely in line with the ICN Merger RPs and therefore marks another step towards procedural convergence. More importantly, the legislative change defers intervention against many foreign-to-foreign mergers with limited impact on Germany, which can be seen as an application of comity principles. The increasing convergence in substantive matters facilitates cooperation. The revised merger guidelines in several countries show that market analysis is implemented today by competition authorities according to essentially equal standards, leaving non-horizontal cases aside. In non-horizontal mergers there are significant differences between European and US authorities, but one can detect some movements in the US practice on this front.57 A change in the substantive test in Germany to the SIEC test would also mean more convergence, but even if no change is made to the test, this would not have a negative impact on the already high level of convergence in the substantive assessment, which is in fact independent of the test. Furthermore, this would not influence the high level of cooperation with other competition authorities. In both scenarios, the BKartA is committed to maintaining and enhancing this level of cooperation.

57 See, e.g., Department of Justice, press release on settlement in the Ticketmaster/Live Nations case, which also had some vertical aspects: “Justice Department Requires Ticketmaster Entertainment Inc. to Make Significant Changes to Its Merger with Live Nation Inc.”, http://www.justice.gov/opa/ pr/2010/January/10-at-081.html. Statements by senior officials of the FTC and the DOJ also point in this direction. See J. Thomas Rosch, “Terra Incognita: Vertical and Conglomerate Merger and Interlocking Directorate Law Enforcement in the United States”, Remarks before the University of Hong Kong, 11 September 2009 (“we are about to witness a resurgence in non-horizontal merger enforcement in the United States“), http://www.ftc.gov/speeches/rosch/090911roschspeechunivhong kong.pdf (speech also provides further examples); Christine Varney, Answers to Questions for the Record from Senator Charles Schumer Before the Senate Judiciary Committee (10 March 2009), http://judiciary.senate.gov/nominations/111thCongressExecutiveNominations/upload/Varney-QFRs. pdf (“both horizontal and vertical mergers must be reviewed with rigor. […] I will not shy away from considering whether the vertical integration resulting from a merger or acquisition is likely to substantially lessen competition.”).

Thomas Deisenhofer1

International Cooperation in Merger Cases – An EU Practitioner’s Perspective

1. Introduction The trends are well known: an ever larger proportion of mergers are transnational and multi-jurisdictional essentially for two reasons. First, globalisation leads to more transnational mergers with effects in more than one jurisdiction. Second, the number of jurisdictions with merger control systems has grown exponentially. The ensuing multiplication of multi-jurisdictional mergers creates significant challenges for businesses and antitrust authorities. Over the last ten years merger control authorities around the world have therefore been investing significant efforts to increase international convergence and cooperation. To be effective, those efforts have to be made at several levels: o regular high level dialogue at political, senior management and academic level (bilateral, multilateral); o joint efforts to improve cooperation ground rules (bilateral agreements, multilateral principles for cooperation) and capacity building; o joint efforts to achieve a high degree of convergence on the substantive assessment of mergers and on the merger investigation process at bilateral and multilateral (ICN, OECD) level; and o day to day practical cooperation in cases. This paper focuses on practical day to day cooperation in cases, a significant part of which takes place at case-team level. It mainly discusses recent cooperation of the Commission with third country agencies and to a more limited extent cooperation of the Commission with the Member States of the EU (in particular as regards the allocation of jurisdiction through referrals). Section 2 recalls briefly why cooperation is so important. Section 3 provides and discusses some statistics on recent cooperation by the Commission in particular in the more complex cases. Section 4 describes some of the typical practical challenges. Section 5 presents selected recent cases.

1 Head of Unit, DG Competition, European Commission. The views expressed in this paper are my personal views and do not necessarily reflect the views of the Commission. I am grateful to Yannick Stomphorst, at the time of writing a stagiaire in my unit, for his valuable research assistance.

228  Merger Control in European and Global Perspective

2. Why close cooperation is important Close practical cooperation between merger control authorities can have significant benefits. Firstly, it can enhance the efficiency of fact-finding in a global business context, e.g. through joint collection or exchange of documents and data between authorities. Moreover, it can lead to more informed decision-making. Cooperation allows the authorities involved to: o bring together expertise from various geographies; o learn from each other (as Rachel Brandenburger has said, “nobody has a monopoly on good ideas”2); and o better understand the implications of a merger and of intervention in all of the affected jurisdictions. Close cooperation is also essential for achieving the comity objective of mutually taking into account the important interests of the authorities involved. Finally and importantly, cooperation can help to achieve the ultimate objective of preventing competitive harm in each affected jurisdiction, while preserving to the greatest extent the efficiencies of the merger. By contrast, where merger control authorities fail to cooperate or where they cooperate poorly, the potential costs and risks are also significant. Lack of cooperation or poor cooperation may: o delay the closing of deals and raise transaction costs for undertakings; o enable firms to engage in strategic behaviour such as forum-shopping or gaming different authorities against each other; o lead to inconsistent or even conflicting outcomes/remedies. Different outcomes or remedies are not per se a reason for concern, as they may reflect different effects of a merger in different jurisdictions (however, serious problems may arise where the parties are subject to conflicting legal obligations or unnecessary double regulatory burdens); o put the efficiencies of a merger at risk; or o cause divergences which lead to politicization and ultimately trade tensions. It is submitted that the costs and risks of not cooperating successfully set out above provide perhaps an even stronger reason for investing time and effort in close and convergence-oriented cooperation than the potential benefits. Close and effective cooperation is thus not just an option, but a “must”. VicePresident Almunia expressed it as follows: “It is clear that at the beginning of the 21st century we cannot afford to operate, to enforce our competition laws in national or regional silos.”3 2 Rachel Brandenburger, “Transatlantic Antitrust: Past and Present”, speech, St. Gallen, 21 May 2010, http://www.justice.gov/atr/public/speeches/260273.htm. 3 Joaquín Almunia, “Cooperation and Convergence: Competition Policy in the 21st Century”, SPEECH/10/183 of 27 April 2010.

Thomas Deisenhofer 

229  

3. Statistics For purposes of the present paper, two datasets have been examined. The first dataset comes from previously conducted internal research that examined information on all 926 merger decisions adopted between 1 January 2007 and 30 September 2009.4 The data was gathered from DG Competition’s internal case management system and complemented by data from an internal survey circulated in 2009. Based on this first dataset, the first part of this section will provide a broad picture of international cooperation in merger cases. The second dataset was created for the purposes of this paper and focuses on international cooperation in recent “significant” or at least “heavy” cases. The data was gathered from DG Competition’s internal case management system and covers a sample of 94 decisions taken in such cases between 1 November 2008 and 31 January 2011. Within DG Competition, cases are classified as “significant” when they require special attention from senior management. Cases are classified as “heavy” if they require significant resources. During the reference period, the significant and/or heavy cases corresponded to roughly 13% of all notified cases. The second dataset focuses on this specific sample because, in unproblematic cases, referrals and international cooperation are less necessary and play less of a role in practice. Building on this dataset, the second part of the section will provide a more focused insight with regard to international cooperation and referrals within the EU in the most important cases.

a. First dataset: a broad picture of international cooperation As shown in Table 1, DG Competition cooperated with third-country agencies in 61 (roughly 7%) of the 926 merger cases decided in the 1 January 2007 to 30 September 2009 timeframe. Table 1: Breakdown of cooperation cases by type of decision*  

Total decisions

Decisions involving international cooperation

Cooperation cases as a % of total merger cases

Phase I Decisions Simplified Procedures Non-Simplified Procedures o Decisions without Commitments

875 530 345 299

51 14 37 22

6 % 3 % 11 % 7 %

4 The internal research was conducted by DG Competition’s international relations unit (Unit A5), which kindly consented to the use of the data in an aggregate form.

230  Merger Control in European and Global Perspective o Decisions with Commitments

46

15

33 %

Phase II Decisions Decisions without Commitments Decisions with Commitments Article 8.3 Prohibition

27 14 12 1

8 5 3 0

30 % 33 % 25 % 0 %

Withdrawals or abandonments

24

2

8

Total

926

61

% 7 %

* Out of all merger decisions taken in the 1 January 2007 to 30 September 2009 timeframe.

A more detailed analysis of the data by type of decision shows that international cooperation was most prominent – in relative terms – in cases involving remedies (both Phase I and Phase II decisions) and in the other cases decided in Phase II. Whereas only 3% of the cases decided in a Simplified Procedure involved international cooperation, 11% of the cases decided on a non-Simplified Phase I procedure involved cooperation with agencies in third countries. Cooperation was most prevalent in cases involving commitments (32% of Phase I commitment decisions and 33% of Phase II commitment decisions) and in Phase II decisions in general (30%). Table 2 below shows that the two US agencies (the Federal Trade Commission and the Department of Justice) were involved in almost half the cases in which cooperation took place, with the FTC accounting for a greater number of cooperation cases than the DOJ. Although the frequency of cooperation with the agencies of other third countries was lower, cooperation took place with multiple other partners. Interestingly, there was also particularly frequent cooperation with the Canadian Competition Bureau (8 cases), which may be explained by (a) the longstanding nature of the relationship between DG Competition and the Competition Bureau (Canada was the first third country after the US with which a bilateral cooperation agreement was concluded) and (b) the close integration of the US and Canada economies. Table 2: Cooperation partners in merger decisions*  

US – FTC/DOJ Canada Mexico Switzerland

Phase I Simplified procedure

Phase I Phase II Non-Simplified Decisions procedure

Total

9 1 2 1

30 5

44 8 3 3

2

5 2 1

Thomas Deisenhofer  Australia Chile Japan South Africa Brazil Turkey Total cooperation partners Total cooperation cases

2

2

2 2

16

43 14

2 2 2 2 1 1

1

1

9 37

231  

68 8

74

* Out of all merger decisions taken in the 1 January 2007 to 30 September 2009 timeframe.

The internal research further suggests that the issues that were most frequently addressed in discussions with third country agencies were the definition of product and geographic markets, theories of harm and the timing and nature of remedies. Moreover, it appeared that cooperation in merger cases primarily took place at the case-team level. Other tendencies identified by the internal research can be described as follows: o in the majority of cases, it was DG Competition that took the initiative to engage in cooperation with the other agency; o discussions were most frequent before formal notification and in the first phase of a merger investigation; where the investigation would enter the second phase, cooperation continued; o in most cases, cooperation involved less than five contacts between DG Competition and the third country agency; the most common means of communication was by e-mail or telephone conversations; actual meetings and video conferences also occurred, but less frequently; o in some cases it was difficult to align the timing of investigations, as competition agencies are to a large extent dependent on the willingness of the parties to cooperate on this matter.

b. Second dataset: global cooperation and intra-EU referrals in heavy and significant cases As illustrated in Figure 1, of the sample of 94 significant or heavy cases in the 1 November 2008 to 31 January 2011 timeframe, 53 involved only parties from inside the EU (‘EU-mergers’), 18 had one party from the EU and one from a third country (‘mixed mergers’), while 23 included only parties from third countries (‘non-EU mergers’).

232  Merger Control in European and Global Perspective Figure 1: Merging parties and international cooperation* 60

53

50 40 30 20

23

19

18 6

10

2

0 non-EU

Mixed Total

EU

International co-operation

* Out of the selected 94 “significant” and/or “heavy” cases.

Global cooperation DG Competition cooperated with its non-European Economic Area (EEA)5 counterparts in a total of 27 cases of the sample of 94 cases (close to 30% of all significant or heavy cases). Not surprisingly it cooperated with third countries mainly in “non-EU” and “mixed mergers”. As Figure 1 indicates, international cooperation took place in 19 cases that are part of the 23 cases involving only firms from third countries (i.e. international cooperation in over 80% of the concentrations in the “nonEU” category). Moreover, such cooperation occurred in 6 cases that are part of the 18 cases involving one party from the EU and one from a third country (i.e. cooperation in one third of the concentrations in the “mixed mergers” category), while cooperation took place in 2 cases that are part of the 53 concentrations only involving parties from within the EU. The frequency of cooperation in cases in which the Commission “intervened”6 was similar to the frequency of cooperation in cases without such intervention. As Table 3 shows, out of the 41 “global” cases (i.e. the aggregate of mixed and non-EU cases), the Commission intervened in 19 cases. In 11 out of these 19 concentrations (corresponding to almost 60%), international cooperation took 5 “EEA” is short for European Economic Area and refers to the EU Member States plus Iceland, Liechtenstein, and Norway. 6 “Intervention” cases are cases in which (1) remedies were imposed (whether in Phase I or Phase II), (2) the transaction was prohibited, or (3) the transaction was abandoned in Phase II.

Thomas Deisenhofer 

233  

place. In the remaining 22 global, non-intervention cases, the Commission cooperated in 14 instances (corresponding to just over 60%). Table 3: International cooperation in global, intervention cases* Remedies in Phase I or II International cooperation No international cooperation

18 10 8

Withdrawals in Phase II International cooperation No international cooperation

1 1 0

Prohibitions

0

Overall interventions International cooperation No international cooperation

19 11 8

* Out of the sub-sample of 41 “global” cases.

Furthermore, the data show that international cooperation is most frequent in those merger units within DG Competition which are responsible for sectors characterized by global markets. As Figure 2 illustrates, almost 80% of the cooperation with third country authorities occurred within the sectors ‘Information, Communication and Media’, ‘Pharmaceuticals’, and ‘Basic Industries, Manufacturing and Agriculture’.

234  Merger Control in European and Global Perspective Figure 2: Sectoral overview of cooperation cases* 10 9

8

8

6

4

4

4

2 1

0

Energy and Environment

Information, Communication and Media

Financial Services

1

Pharmaceuticals

Basic Industries, Manufacturing and Agriculture

Transport, Post and other services

* Out of the selected 94 “significant” and/or “heavy” cases.

The dataset also suggests that in today’s globalized economy, cooperation in a given case often takes place with more than one jurisdiction. In 11 out of DG Competition’s 27 cooperation cases, more than one other jurisdiction was involved. In those cases DG Competition cooperated with up to 7 partner authorities. In the remaining 16 cooperation cases, DG Competition cooperated only with one third country regulator. Finally, the data indicate that the cooperation partners of DG Competition are diverse and represent all continents. While the US (either the Federal Trade Commission or the Department of Justice) remains the most frequent cooperation partner for DG Competition, with cooperation in 25 cases, other frequent partners include Canada (9), Australia, China, Japan and Switzerland (3), and South Africa (2). DG Competition also cooperated with regulators from Chile, Mexico, Russia, and South Korea.

Referrals within the EU In addition to cooperation with agencies from outside the EEA, DG Competition continued its close cooperation with the competition authorities of the Member States. It is useful to compare referral numbers to the data on international cooperation. The Merger Regulation foresees a number of different possibilities for case referrals between the Member States and the Commission:

Thomas Deisenhofer 

235  

o Article 4: pre-notification referrals – of a Community dimension concentration to a Member State (Article 4(4)) – of a concentration which does not have a Community dimension, but is notifiable in several Member States, to the Commission (Article 4(5)); o Article 9: post-notification referral of a Community dimension concentration to a Member State at the request of that Member State; and o Article 22: referral to the Commission of a concentration without EU dimension. Out of the sample of 94 cases, a case was referred either to Member State authorities or to DG Competition in 14 instances. In potentially problematic or heavy cases, referrals are therefore relatively frequent. Table 4 provides an overview of the distribution of referral decisions per category of concentrations. Referrals were mainly made in cases involving only parties from within the EU. However, there were also three referral cases which involved parties from outside the EU. Table 4: Overview of referrals* Referral mechanism

Referral cases Total cases

EU mergers Article 4(4), full referral to Member State authorities Article 9(3), full referral to Member State authorities Article 9(3), partial referral to Member State authorities Article 9(3), refusal to refer

11 3 2 4 2

53

Mixed mergers Article 22, full referral to the Commission

1 1

18

Non-EU mergers Article 22, full referral to the Commission

2 2

23

14

94

Total referral cases * Out of the selected 94 “significant” and/or “heavy” cases.

4. Challenges for successful international cooperation In DG Competition’s case practice, the typical challenges for effective international case cooperation include, first of all, the divergent and tight timetables. These are mainly due to the different procedures and administrative practices in different jurisdictions, and/or strategic behaviour by the parties as well as the usually tight deadlines typical for merger control procedures.

236  Merger Control in European and Global Perspective Other typical challenges are: (i) confidentiality requirements under national legislation which impede efficient sharing of information; (ii) sovereignty-related issues; and (iii) differences in legal and administrative cultures, and processes within partner authorities which make cooperation difficult (e.g., where the caseteam has no mandate to engage in proactive cooperation). Moreover, DG Competition, like other authorities, is faced with the general challenges of cooperating across the globe, challenges posed by language barriers, different time zones, and more ‘soft’ but important cultural factors. Finally, each merger case stands on its own, and high profile transnational merger cases are often highly complex; even where lawyers (or economists) from the same jurisdiction look at the same merger and the same facts, they do not necessarily agree on how to assess such complex cases. It follows, first, that successful international cooperation is hard work. It requires open minds on all sides, investment of significant efforts, some intercultural skills and strong support from the top of the respective authorities. Secondly, given the significant challenges identified above, it is actually surprising that international cooperation works so well and that so few cases lead to tensions. As regards the EU, of the 27 international cooperation cases included in the second dataset, only one case7 gave rise to some tensions during certain moments of the assessment, and the final outcome was identical. If we think of how many high-profile cases within each of our jurisdictions lead to divergent assessments by the national competition authority and the responsible courts in that jurisdiction (despite the fact that both dispose of the same information, consider the same effects and are steeped in the same legal culture), the very high degree of international agreement is truly remarkable and raises the interesting research question of why there is such a high incidence of agreement. It is submitted that the first and most obvious reason for this virtual absence of international tension and disagreement is the significant and exemplary effort invested by all major competition authorities around the world in international convergence and cooperation. There are few other areas of international economic governance where there has been so much progress over the last ten years. Another deeper reason might be that competition authorities around the world are very much aware of the potentially disruptive effects associated with international divergence. They seem to try to avoid the costs and risks associated with international ‘conflicts’ even more than a certain degree of risk of conflict with their own courts. They may regard the latter type of risk of conflict as a more acceptable, “normal” part of the national legal dialogue.

7

M.5529, Oracle/Sun (21 January 2010), discussed below.

Thomas Deisenhofer 

5.

Selected cooperation and referral cases

a.

Cooperation with third countries

237  

Panasonic/Sanyo8 Panasonic’s acquisition of Sanyo, both of Japan, raised horizontal concerns in a number of battery markets where the merged entity would have held significant market shares. In order to address these concerns, the parties gave divestiture commitments regarding certain production facilities. The Commission cleared the concentration – subject to commitments – in Phase I on 29 September 2009. The US Federal Trade Commission (FTC) and the Japanese Fair Trade Commission (JFTC) cleared the case subject to similar conditions. Throughout the case the Commission worked in close cooperation with the FTC and the JFTC. Cooperation started already prior to notification. Cooperation was challenging inter alia because of different timetables and language barriers (e.g., internal documents in Japanese). At the time of notification in the EU, the FTC had already issued a second request and the parties were already at an advanced stage of negotiation with the JFTC. Under the Japanese system the parties, anticipating possible concerns, had the possibility to engage in consultations on possible commitments well before the formal filing was due. The parties gave all the necessary waivers and invested significant resources in translating documents. At one stage it seems that they hired more than 1,000 translators to reply to the second request of the FTC in a timely manner. The contacts – which were facilitated by the bilateral cooperation agreements between the respective authorities – allowed for the extensive exchange of information, co-ordination of procedural timetables and discussions of the focus of the market investigations. DG Comp benefited, for example, from the information collected by and translated for the FTC for its second request. Most importantly, DG Competition worked successfully with its US and Japanese counterparts to avoid conflicting remedies in different regions and to ensure that, in the interest of consumers and the merging parties, the remedies were consistent and coherent.

BHP Billiton/Rio Tinto9 The Commission identified potential concerns in the markets for iron ore and for metallurgical coal and opened an in-depth (Phase II) investigation on 4 July 2008. The Commission closed its investigation after the transaction was abandoned on 25 November 2008. As a reason for the withdrawal, BHP stated that due to 8 9

M.5421, Panasonic Corporation/Sanyo Electric (29 September 2009). M.4985, BHP Billiton/Rio Tinto (26 November 2008).

238  Merger Control in European and Global Perspective the continued deterioration of near-term global economic conditions and due to regulatory concerns in Europe, the acquisition was no longer in the best interests of its shareholders. Prior to the withdrawal the Commission had issued a Statement of Objections. The transaction was under scrutiny of regulators in Asia, Australia and Europe. Throughout the process, DG Competition cooperated closely with the Japan, Korea, Australia and the US.

Oracle/Sun10 This case concerned Oracle’s (business software) acquisition of Sun Microsystems (hardware and software). After Phase I the Commission could exclude significant negative effects on Java and middleware and within the IT stack. Phase II focused on horizontal effects in the database market, where the world’s leading proprietary database vendor Oracle was acquiring, as part of Sun, the world’s leading open source database MySQL. Following an in-depth investigation (including a Statement of Objections and a hearing) as well as public pledges from Oracle, the Commission unconditionally cleared the case in January 2010. Timetables were different. Oracle notified the case to the Commission on 30 July 2009 several months after it had filed in the US and after the Department of Justice had issued a second request. The DOJ cleared the case at the end of August, several days before the Commission opened its Phase II investigation on 3 September 2009. The case attracted some controversy. Oracle triggered a letter campaign of 200 of its customers. Fifty-six US Senators sent a letter to the Commission requesting a fast process. On the other side, more than 1600 MySQL users sent e-mails to the Commission opposing the merger. An internet ‘Save MySQL’ petition attracted 40,000 signatures opposing the transaction. Throughout the Oracle/Sun review, DG Competition’s cooperation with the DOJ was close, frequent and based on full transparency, despite divergences in the assessment at certain moments. Without that close cooperation it would have been difficult, for example, for the Commission to exclude concerns as regards Java already in Phase I. The continuous dialogue started in pre-notification and included frequent exchanges at case-team level and phone and video conferences at Senior Management level at critical moments of the procedure. At the Commission’s invitation, a DOJ representative attended the Oral Hearing in Brussels. Divergences in the assessment at certain moments of the procedure may have stemmed partly from the fact that the competition issues surrounding the open source database MySQL were perhaps stronger in Europe than in the US, since MySQL was founded in Europe and open source software has greater acceptance in Europe. 10

Cited above note 7.

Thomas Deisenhofer 

239  

Microsoft/Yahoo! Search business11 In this case, Microsoft proposed to acquire Yahoo’s search business, composed of its search engine and the search advertising platform Panama. The transaction was unconditionally cleared on the same day by both the DOJ and the Commission in February 2010 following a second request investigation by the DOJ and an unusually intense Phase I investigation by the Commission. Cooperation with the DOJ started very early in pre-notification. This led to a common understanding of the facts of the case. It was characterized in practice by (i) at the case-team level, regular conference calls between the authorities, in particular regarding the economic analysis and the relevant data sets; (ii) each authority informed the other in advance of important steps in the procedure; and (iii) the DOJ allowed the Commission to participate in its informative conference calls with the parties. The clearance decisions were announced on the same day with aligned reasoning and press releases.

Cisco/Tandberg12 This case concerned the acquisition by Cisco (a US company active in the field of networking products including video communications solutions (‘VCS’), i.e. teleconferencing solutions) of Tandberg (Norway/US, a VCS vendor). The Commission identified horizontal competition concerns due to interoperability issues with competitors’ products. In order to solve this interoperability concern, Cisco committed, inter alia, to divest the rights attached to its proprietary Telepresence Interoperability Protocol to an independent industry body. The Commission cleared the case – subject to commitments – in Phase I on 29 March 2010. The DOJ also cleared the case due to, inter alia, the commitments made by the parties to the Commission. The close cooperation between DG Competition and the DOJ was highlighted by the title of the DOJ’s press release: “Justice Department will not challenge Cisco’s acquisition of Tandberg - Justice Department and European Commission cooperate closely”. Cooperation led to a common understanding of the facts of the case and alignment of the timing of the final decision (with both decisions adopted on the same day). It was characterized in practice by (i) at the case-team level, regular conference calls between the authorities and joint discussion with the parties, and (ii) at the senior management level, discussions held at crucial moments for the orientation of the case.

11 12

M.5727, Microsoft/Yahoo! Search business (18 February 2010). M.5669, Cisco/Tandberg (29 March 2010).

240  Merger Control in European and Global Perspective

b. Cooperation with Member States in particular on the allocation of jurisdiction Each of the referral mechanisms mentioned above is used on a frequent basis in practice. Examples of Article 4(4) referrals include the Spanish TV cases13 from the Commission to Spain in 2010, upon request of the notifying parties. The transactions had effects mainly in Spain and the Spanish Competition authority has considerable experience in this sector. An example of an Article 9 referral is the German internet Video on Demand case (ProSiebenSat.1/RTL)14 from the Commission to Germany and Austria at the request of the two Member States in question. On the basis of past precedents, markets were probably national. The Bundeskartellamt in particular knew the market and knew the parties very well. The Commission is also prepared to consider mixed solutions – such as partial Article 9 referrals, in which the Commission clears the rest of the case. In Veolia/ Transdev15 for example, a referral of national markets was made to France and the Netherlands and the Commission cleared on the EEA-wide markets. In CVC/ Univar/Eurochem,16 the Commission cleared the Belgian and Dutch aspects of the transaction, but referred the French parts to France because the market was different there. Furthermore, the Commission has dealt with a number of Article 22 referrals. Here the lesson is that, if the Commission is the best placed authority, it will often end up with the case, even if the parties did not wish to request an Article 4(5) referral. In SCJ/Sara Lee (Insecticides),17 the Commission received a referral request from six Member States. The transaction was abandoned in phase II. In P&G/Sara Lee (Air Care),18 the Commission received a request from five Member States but in the end the case was cleared unconditionally in Phase I. The latter case shows that Article 22 referrals do not prejudge the outcome. Finally, in Syngenta/Monsanto,19 the Commission received a request from two Member States and cleared the case in Phase II, subject to conditions. One first clear message which emerges as regards the allocation of jurisdiction within the EU is that forum shopping is a risky strategy. The Commission and the national competition authorities cooperate closely in order to make sure that the best placed authority deals with the case. The Commission is working to further strengthen this coordination and has set up a Mergers Working Group within the ECN. 13 M.5748, Prisa/Telefonica/Telecinco/Digital+ (11 March 2010); M.5776, Telecinco/Cuatro (11 March 2010). 14 M.5881, ProSiebenSat.1 Media/RTL Interactive/JV (24 September 2010). 15 M.5741, CDC/Veolia Environment/Transdev/Veolia Transport (12 August 2010). 16 M.5814, CVC/Univar Europe/Eurochem (16 July 2010). 17 M.5969, SCJ/Sara Lee (9 May 2011). 18 M.5828, Procter & Gamble/Sara Lee (Air Care) (17 June 2010). 19 M.5675, Syngenta/Monsanto (Sunflower Seeds Business) (17 November 2011).

Thomas Deisenhofer 

241  

A second message as regards the delays and transaction costs associated with the referral system is that a lot depends not only on well-functioning cooperation between the Commission and the Member State authorities concerned, but also on the efforts and strategies of the parties and their lawyers. A positive example of how win-win outcomes can be achieved is the Orange/TMobile case,20 in which the parties cooperated from the beginning openly and effectively with the Commission and with the UK authorities, the OFT and OFCOM. Due to the parties’ significant efforts and the close cooperation between the Commission and the OFT and OFCOM, it was possible to find effective and clear-cut remedies. On that basis, the UK could withdraw its Article 9 referral request and the Commission was able to clear the case in Phase I.

6. Conclusions and tentative outlook Close cooperation in important transnational merger cases can have significant benefits. Perhaps even more importantly, the absence of cooperation or poor cooperation may create significant risks or costs. From a risk-management point of view, investing resources in close and successful international cooperation is thus for a competition authority not just an option but a “must”. In the Commission’s case practice, close international cooperation is routine in all important international cases. Where appropriate, cooperation takes place with several authorities in parallel. Despite the significant practical challenges, cooperation works well and leads in the vast majority of cases to processes and outcomes which are satisfactory for the authorities and parties involved. Divergences or tensions with other authorities are remarkably rare. Successful cooperation depends to a considerable extent on the cooperation and goodwill of the merging parties. It is no coincidence that in complex cases where the parties have invested significant efforts to allow the Commission and its partner authorities to cooperate fully and along a mutually agreed timetable (e.g., in Panasonic/Sanyo, Microsoft/Yahoo! Search Business, Cisco/Tandberg or T-Mobile/Orange), the authorities were able to find an acceptable solution within a reasonable period of time. As regards the EU, all those cases could be solved in Phase I. In cases where the parties were less cooperative, the process was sometimes more confrontational or took more time. Given the progress which has already been made during the last ten years and given that, in recent years, there have been very few instances where problems arose, it is at first sight not easy to identify areas where significant further progress can be made. However, it is submitted that some progress could be made for example in the following areas: 20

M.5650, T-Mobile/Orange (1 March 2010).

242  Merger Control in European and Global Perspective o Stronger emphasis on the role and responsibility of the merging parties for allowing aligned procedural timetables; o Better coordination in the collection of information (joint investigative efforts) in order to save the resources of the authorities, of the parties and of third parties; o Devise legal means and waivers that would allow the authorities to better share information received from third parties; o Strengthened cooperation in the design and implementation of remedies: e.g., joint remedy packages, joint trustees and coordination of divestiture periods (as well as extensions) and of buyer approvals.

Rachel Brandenburger*

Promoting International Convergence: Substantive and Procedural Challenges – the Scope for Comity

[P]rinciples of comity provide a useful mechanism to avoid duplicative enforcement and to reduce instances of potentially conflicting decisions, thereby making antitrust enforcement more efficient and lessening costs on business and consumers.1

I. Introduction It is a great honor to be here today and to share this panel with my esteemed colleagues and friends in beautiful Florence. As we begin to close out the calendar year, it is hard for me to believe that I have already been working at the U.S. Department of Justice for nearly a year. I am proud of what the Antitrust Division has accomplished this year on the international front, feel very privileged to have been a part of it, and am looking forward to an even more productive 2011. Today I will address the themes of promoting international convergence, substantive and procedural challenges, and the scope for comity. I would like to share with you my perspectives on those issues in light of today’s global economy and multi-polar world. First, I will step back to the past and discuss how cooperation, convergence, and transparency came to be the working paradigm for U.S. international competition policy for the last decade. I will then address the progress we have made in merger enforcement in those three areas.2 Last, I will share with you some additional guiding principles, which I believe will help us meet the challenges ahead. As Special Advisor, International at the Antitrust Division, my overarching goal has been to further integrate the consideration of international issues into the day-to-day, practical work of the Antitrust Division – both with respect to our investigations and our policy work. My aim has been to intensify the Antitrust Division’s cooperative relationships with other competition agencies and to encourage our staffs to be mindful of the international implications of our actions right from the very start of an investigation through to the remedial phase. * Special Advisor, International, Antitrust Division, U.S. Department of Justice. 1 Antitrust Modernization Commission, Report and Recommendations (April 2007), http://govinfo. library.unt.edu/amc/report_recommendation/toc.htm, p. 221. 2 See also Rachel Brandenburger, “International Competition Policy and Practice: New Perspectives?”, speech, London, 29 October 2010, http://www.justice.gov/atr/public/speeches/270980.pdf.

244  Merger Control in European and Global Perspective We cannot afford to think of the international dimension of our competition work as an afterthought or a distraction – or just something interesting to talk about at conferences. We must infuse our daily competition enforcement work with the international dimension because, quite simply, the challenges presented by today’s global economy and multi-polar world demand it. Our collective challenge for the future will be adapting our existing cooperation protocols to a globalized economy and a world of over 100 enforcers. As former Assistant Attorney General Varney aptly and succinctly put it, “In today’s world, competition agencies can no longer cooperate on investigations with only one or two other jurisdictions and call it a day.”3 The world of multiple enforcers is no longer theoretical. As a number of this conference’s panelists have observed, it is very much today’s reality.4 I agree. Based on my recent experience in private practice and my work over the past year at the Antitrust Division, I can assure you that multiple competition agencies increasingly are reviewing the same transactions and conduct, and cooperation is taking place among more agencies than before. We are fortunate to have made good progress in international competition policy and practice over the last decade, and to have in place many of the building blocks we will need for the future. In this regard, I wish to pay tribute to Jim Rill, former Assistant Attorney General of the Antitrust Division. He has been instrumental in a number of the important international initiatives that have led us to where we are today. For example, as many of you know, Jim Rill was co-chair of the U.S. International Competition Policy Advisory Committee (ICPAC) – a blueribbon advisory committee established by the U.S. Department of Justice in 1997 to consider international competition issues. ICPAC’s groundbreaking report in 2000 identified the most pressing international competition policy issues facing the United States, as well as principles to guide the U.S. Department of Justice, and the U.S. government in general. In particular, ICPAC recommended (1) increased transparency and accountability of government actions; (2) expanded and deeper cooperation between U.S. and non-U.S. competition enforcement authorities; and (3) greater soft harmonization and convergence of systems.5 The principles of transparency, cooperation, and convergence have been the core of our international competition policy efforts over the past 10 years. As I will discuss in a few moments, while we may wish to add to them and refine them for the future, they certainly will remain important in the years to come. Let us look now at what we have achieved to date with respect to these three principles in the area of merger enforcement – the focus of this workshop. 3 Christine Varney, then-Assistant Attorney General, U.S. Department of Justice, “International Cooperation: Preparing for the Future”, speech, Washington, D.C., 21 September 2010, http://www. justice.gov/atr/public/speeches/262606.pdf, at p. 16. 4 For example, Nicholas Levy, in his oral presentation at this Workshop, referred to charts showing the proliferation of merger control regimes and summarizing some of the challenges regarding multijurisdictional merger review. 5 International Competition Policy Advisory Committee, Final Report to the Attorney General and Assistant Attorney General for Antitrust (28 February 2000) (“ICPAC Report”), http://www.justice. gov/atr/icpac/finalreport.htm.

Rachel Brandenburger  245  

II. Convergence, cooperation and transparency in merger enforcement To date, we have made a great deal of progress on convergence, cooperation, and transparency in international merger enforcement. As for convergence, we now have agreement on many of the fundamentals of merger review. One need only compare the recently revised U.S. Merger Guidelines6 with other merger guidelines around the world – or look to the ICN’s recommended practices on merger procedures7 and substantive merger review,8 and the scores of jurisdictions that have changed their laws and policies and now conform with them – to see how far the convergence of merger analysis has come in the last decade. Convergence and cooperation have tended to go hand in hand. Greater convergence has been facilitated by greater cooperation between competition agencies – in particular, through their interactions with each other in the course of their investigations, but also through technical assistance and international bodies like the Organisation for Economic Cooperation and Development (OECD) and the International Competition Network (ICN). And, as agencies’ policy views on mergers have converged over the years, the robustness of cooperation among competition agencies on actual investigations has correspondingly increased. It is important to keep in mind, however, that achieving agreement on general merger pronouncements on procedures and substantive review, while reducing the likelihood of divergent analyses and outcomes, does not necessarily guarantee that we will arrive at the same results in the context of a particular case. Thomas Deisenhofer points out in his paper for this workshop that very few cases have led to divergence or tensions between the reviewing competition agencies.9 This has been my experience at the Antitrust Division as well; when our staffs are fully engaged with another competition agency and when all involved are working from a common set of facts on a particular transaction, there is 6 United States Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, (19 August 2010), http://www.justice.gov/atr/public/guidelines/hmg-2010.html. See Rachel Brandenburger and Joseph Matelis, “The 2010 U.S. Horizontal Merger Guidelines: An Historical and International Perspective”, 25 Antitrust 48 (Summer 2011). 7 International Competition Network, Recommended Practices for Merger Notification and Review Procedures, http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. The ICN recently reported that more than half of ICN’s member jurisdictions with merger laws have implemented, or were planning revisions to bring their merger regimes into greater conformity with, these Recommended Practices. See ICN, A Statement of Achievements Through April 2010, http:// www.icn-istanbul.org/Upload/Materials/Others/StatementOfAchievements.pdf, at p. 3. 8 ICN, Recommended Practices for Merger Analysis, http://www.internationalcompetitionnetwork. org/uploads/library/doc316.pdf. Consistent with these Recommended Practices, while different jurisdictions may use different wording to describe their substantive standards, the overwhelming majority focus on whether a merger will substantially lessen competition. The Recommended Practices contain consensus principles reached in the analysis of competitive effects, market definition, use of market shares, entry and expansion, failing firm/exiting assets, and the legal framework for competition merger analysis. 9 See Thomas Deisenhofer, “International Cooperation in Merger Cases – An EU Practitioner’s View”, this volume, pp. 227 et seq.

246  Merger Control in European and Global Perspective rarely divergence. However, we should not become complacent. In the future, we may be faced with more challenging circumstances, as more jurisdictions, more agencies, and different market conditions and economies become involved in the same merger investigations. Today, we see agencies reviewing the same or substantially similar merger transactions increasingly cooperating and coordinating their investigations. Two recent examples of particularly effective cooperation in merger review come immediately to mind. The first is the review by the Antitrust Division and the Canadian Competition Bureau of the merger between Ticketmaster and Live Nation. The Division coordinated closely with the Bureau at the investigative stage, and the two agencies worked closely together to obtain a remedy, announced the same day, that preserved competition across North America.10 A second example of effective international merger cooperation is the review of the Cisco/Tandberg merger by the Department of Justice and the European Commission. Cisco, a U.S. firm, was the leading provider of high-end telepresence videoconferencing products, while Tandberg, headquartered in both Norway and New York, was the leading provider in the broader videoconferencing products space, with a strong and growing role in telepresence. With waivers and cooperation from the merging parties and third-party industry participants in place, the Antitrust Division and the European Commission were able to work closely together from the opening to the conclusion of their investigations. This cooperation included numerous contacts between the investigative staffs, the exchange of documents, sharing one another’s competitive effects analyses, and also conducting joint meetings and interviews with the parties and third parties. It also included active involvement and discussions among both agencies’ senior management. In deciding to conclude our investigation, the Antitrust Division took into account the commitments that the parties were giving to the European Commission to facilitate interoperability and determined that the proposed merger was not likely to be anticompetitive. The announcement that the Department of Justice had closed its investigation was made on the same day that the European Commission announced its clearance decision. Then-Assistant Attorney General Varney characterized the investigation as “a model of international cooperation between the United States and the European Commission,” and she commended the parties for “making every effort to facilitate the close working relationship” between the two agencies.11 VicePresident Almunia similarly expressed satisfaction with “the overall review process that was carried out in close co-operation with the U.S. Department of Justice.”12 10 Press Release, U.S. Dep’t of Justice, Justice Department Requires Ticketmaster Entertainment Inc. to Make Significant Changes to Its Merger with Live Nation Inc. (25 January 2010), http://www. justice.gov/atr/public/press_releases/2010/254540.pdf; Press Release, Canadian Competition Bureau, Competition Bureau Requires Divestitures by Ticketmaster-Live Nation to Promote Competition (25 January 2010), http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03191.html. 11 Press Release, U.S. Dep’t of Justice, Justice Department Will Not Challenge Cisco’s Acquisition of Tandberg (29 March 2010), http://www.justice.gov/atr/public/press_releases/2010/257173.htm. 12 Press Release, European Commission, Commission Clears Cisco’s Proposed Acquisition of Tandberg Subject to Conditions (29 March 2010), http://europa.eu/rapid/pressReleasesAction. do?reference=IP/10/377.

Rachel Brandenburger  247   This is the type of extensive day-to-day cooperation and coordination among competition agencies that we must strive for in the years to come – including recognition that the need for, or nature of, a remedy may differ because of the markets at issue in each investigation. Adam Fanaki’s paper does a fine job summarizing the costs associated with multi-jurisdictional merger review, including the costs associated with inconsistent results, delay and uncertainty.13 For these and other reasons, we must be vigilant in ensuring that global merger review is as efficient and effective as possible. Cooperation is an important vehicle toward that end. Thomas Deisenhofer puts it well in his paper when he states, “[c]lose and effective cooperation is thus not just an ‘option’, but a ‘must’.”14 Turning to transparency, there seems to be more transparency in merger review process and analysis than in other areas of enforcement. One early effort that significantly advanced transparency in the merger area was the development by the ICN of a comprehensive online collection of links to member agencies’ merger laws and other information, as well as members’ responses to a uniform template of questions on key features of their merger review systems.15 In addition, numerous jurisdictions have introduced, or reviewed and updated, merger guidelines to provide businesses, the public, and other agencies around the world with greater insight into their merger procedures and analyses.16

III. Complements: mindfulness, respect, trust and dialogue As I have noted, the three goals of transparency, cooperation and convergence identified in the ICPAC report have served us well over the past 10 years, and they will continue to be important in the years to come. Nevertheless, the world does not stand still, and the challenges faced by competition policymakers and enforcers continue to evolve and change. These new challenges offer new opportunities for us to think about our approaches. As a start, I would like to propose that we should look at transparency, cooperation, and transparency in a new light, complemented by the related principles of mindfulness, respect, trust, and dialogue. All of these principles are, of course, related to, and indeed facilitate, the principle of comity 13 See Adam Fanaki, “Convergence in Multi-jurisdictional Merger Reviews: A Canadian Perspective”, this volume, pp. 261 et seq. 14 See Deisenhofer, cited above note 9, at p. 228. 15 See http://www.internationalcompetitionnetwork.org/working-groups/current/merger/templates. aspx. 16 Notably, the recent revision and update of the U.S. horizontal merger guidelines was primarily an exercise in transparency – closing gaps that had grown between the Guidelines and actual agency practice since their last significant revision in 1992 – 18 years ago. See Christine Varney, then-Assistant Attorney General, U.S. Dept. of Justice, “An Update on the Review of the Horizontal Merger Guidelines”, Remarks prepared for the Horizontal Merger Guidelines Review Project’s Final Workshop, 26 January 2010, http://www.justice.gov/atr/public/speeches/254577.pdf, pp. 1–5; Carl Shapiro, then-Deputy Assistant Attorney General, “The 2010 Horizontal Merger Guidelines: From Hedgehog to Fox in Forty Years”, 77 Antitrust Law Journal 49, 49–50 (2010); Brandenburger and Matelis, cited above note 6.

248  Merger Control in European and Global Perspective (as this panel’s title suggests). None of these principles is new; but all will become increasingly important in a globalized, multipolar world.

A. Mindfulness By mindfulness, I mean that competition agencies must be mindful on a day-to-day basis of the impact of their actions and approaches outside their own jurisdictions, and of the effects that other agencies’ actions and approaches may have on them. Mindfulness of other competition authorities’ jurisdiction, practices, and traditions allows agencies to work together to minimize inconsistent or conflicting approaches. As Christine Varney has observed, “divergent outcomes should occur, if they do, for well-founded reasons, and not arbitrarily or unexpectedly.”17 Mindfulness is especially important with respect to remedies. While in some cases, different remedies may be appropriate for differing market conditions, the potential for multiple remedies carries a serious risk that remedies imposed by one agency will undermine the enforcement options available to other agencies, complicate or frustrate the ability of the parties to comply with multiple forms of ordered relief, and potentially even harm consumers. In our multi-polar world, agencies need to remain mindful of the international impact of their remedial options, as well as the effect that remedies being considered by other agencies will have on them. Mindfulness is not a new concept to international competition enforcement and, in some respects, is very much related to comity. Indeed, the 1995 Department of Justice (DOJ) and Federal Trade Commission (FTC) Antitrust Enforcement Guidelines for International Operations state that the DOJ and FTC consider international comity in enforcing the antitrust laws, “including the extent to which the enforcement activities of another country with respect to the same persons, including remedies resulting from those activities, may be affected.”18 The United States’ bilateral cooperation agreements contain comity-related provisions, as well.19 For example, the cooperation agreement with the European Union requires each party to consider the “important interests” of the other in decisions as to whether or not to initiate an investigation, the scope of investigations and proceedings, and the nature of remedies.20 The agreement further states that 17 Christine Varney, then-Assistant Attorney Gen., U.S. Dep’t of Justice, “Coordinated Remedies: Convergence, Cooperation, and the Role of Transparency”, speech, Paris, 15 February 2010, http:// www.justice.gov/atr/public/speeches/255189.htm. 18 United States Department of Justice and Federal Trade Commission, Antitrust Enforcement Guidelines for International Operations (April 1995), http://www.justice.gov/atr/public/guidelines/ internat.htm. 19 The U.S. has nine cooperation agreements, with the following jurisdictions: Australia, Brazil, Canada, Chile, European Union, Germany, Israel, Japan, and Mexico. These cooperation agreements are available at http://www.justice.gov/atr/public/international/int_arrangements.htm. 20 Agreement Between the Government of the United States of America and the Commission of the European Communities Regarding the Application of their Competition laws (September 1991), http:// www.justice.gov/atr/public/international/docs/0525.htm.

Rachel Brandenburger  249   the potential for adverse impact on one party’s important interests is less at the investigative stage and greater at the stage at which conduct is prohibited, penalized, or at which other forms of remedial orders are imposed.21 Similarly, with respect to merger enforcement, the ICN has adopted a recommended practice which provides that “[r]eviewing agencies should seek remedies tailored to cure domestic competitive concerns and endeavor to avoid inconsistency with remedies in other reviewing jurisdictions.”22 The OECD’s Council Recommendation on Merger Review contains a similar provision,23 and the OECD Council Recommendation on Cooperation requires members to give “full and sympathetic consideration” to the “important interests” of other members.24 We have seen some recent examples of mindfulness in practice. The first (although not a merger case) involved the European Commission’s Microsoft browser case in 2009, where the European Commission limited the remedy to Microsoft products sold in the European Economic Area.25 Another example of mindfulness is when competition agencies work together to develop unified remedies that address competition concerns across jurisdictions, as was done by the Antitrust Division and the Canadian Competition Bureau in the Live Nation/ Ticketmaster merger,26 where the two agencies worked closely together to obtain a remedy that preserved competition across North America. A third example involves competition agencies taking account of whether the remedies imposed by other agencies may resolve part or all of the competition issues in their own jurisdiction, as the Antitrust Division did in relation to the commitments that the parties offered to the European Commission in the Cisco/Tandberg merger.27 Looking to the future, competition agencies will need increasingly to explore these and other ways to work together in a way that is sensitive to the steps another competition agency has taken or has yet to take.28

Ibid. See International Competition Network, Recommended Practices for Merger Notification and Review Procedures, http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 23 Organisation for Economic Co-Operation and Development, Recommendation of the Council on Merger Review (Mar. 23, 2005), http://www.oecd.org/dataoecd/3/41/40537528.pdf. 24 Organisation for Economic Co-Operation and Development, Recommendation of the Council Concerning Co-Operation Between Member countries on Anticompetitive Practices Affecting International trade (July, 27, 1995), http://www.oecd.org/dataoecd/60/42/21570317.pdf. 25 Press Release, European Commission, Commission accepts Microsoft Commitments to Give Users Browser Choice (16 December 2009), http://europa.eu/rapid/pressReleasesAction. do?reference=IP/09/1941. 26 Press Release, U.S. Dept. of Justice, Justice Dept. Requires Ticketmaster Entertainment Inc. to Make Significant Changes to Its Merger with Live Nation, Inc. (25 January 2010), http://www.justice. gov/atr/public/press_releases/2010/254540.htm. 27 Press Release, Justice Department Will Not Challenge Cisco’s Acquisition of Tandberg, cited above note 11. 28 Varney, cited above note 17 (noting that we must be “sensitive to what our international counterparts may yet do in the future – to the choices that they may be considering – and attempt to avoid surprising one another”). 21 22

250  Merger Control in European and Global Perspective

B. Respect and trust In addition to mindfulness, respect should also play an important role. Respect involves: (1) openness to the ideas of others, and (2) respect for our differences. Greater cooperation and convergence will not be possible if any of us comes to the table with the notion that our agency has all the right answers and other jurisdictions must therefore adopt our standards or processes wholesale. In areas where we cannot yet agree, we must also learn to respect our differences. Trust, of course, is also a fundamental element of effective cooperation. As with any sound relationship, trust should grow over time. Trust among competition agencies arises most often out of mutual respect born of working productively together, whether in negotiating consensus documents such as ICN Recommended Practices, having detailed discussions on particular substantive or procedural issues, working together on investigations, or providing technical assistance. The more familiar competition agencies become with one another, the more we can appreciate that we share a common purpose and the more we can realize just how similar our approaches and challenges often are. Building trust between competition agencies and the business community is also important. Effective cooperation among competition agencies is often not solely in the agencies’ own hands. Businesses need to be certain that confidentiality waivers will not result in compromised corporate information. By the same token, competition agencies need to have confidence that parties are not seeking to game the multi-jurisdictional system or play one agency off or against another.

C. Dialogue Finally, ensuring an ongoing dialogue will also be essential. Whether with regard to specific enforcement matters or in a broader policy context, discussion with other competition agencies allows us to exchange insights regarding our experiences in competition law enforcement. Dialogue includes a willingness of competition agencies to revisit their own policies and practices over time to reflect new learning and the experiences of others. For example, a number of agencies around the world are currently revising, or have recently revised, their merger guidelines.29 Indeed, the Competition Law and Policy Workshop held every year at the European University Institute in Florence is another example of promoting constructive dialogue both among agencies and between the agencies and the competition bar. Ultimately, ensuring an ongoing, deep, and meaningful dialogue may be one of the most important things we can do in those areas where we have not yet converged, where convergence is difficult, or where convergence may not 29

For details, see Brandenburger and Matelis, cited above note 6.

Rachel Brandenburger  251   be achieved. Much benefit can come from competition agencies discussing an issue thoroughly, thoughtfully, and respectfully on a bilateral or multilateral basis. Officials involved in such discussions emerge better informed about what other agencies are doing, including what has worked well and not so well in the past; and they are better equipped to think about their own practices with greater perspective.

IV. Conclusion These are some of my thoughts on international convergence and the challenges facing the international competition community. In recognition of these challenges, as well as the importance of getting our answers right, then-Assistant Attorney General Varney, when speaking in September 2010 at Georgetown University Law School, called for a dialogue about what international competition policy and cooperation – and thus the scope for comity concepts – should look like for the future.30 We in the Antitrust Division are very much interested in your ideas, and our door is always open.

30

Varney, cited above note 3, at p. 16.

William Kovacic*

International Convergence: Assessing the Quality of Horizontal Merger Enforcement

Introduction Horizontal merger policy is an important focus of contemporary discussions about the quality of competition policy.1 It should be. Horizontal merger policy attempts to forestall combinations that otherwise would permit the merged entities to exercise substantial market power, and it tries to curb the creation of market environments that encourage coordination by rival firms through tacit coordination or the formation of express agreements. Because society also has a major stake in allowing restructurings that improve economic performance, both in individual transactions and in the preservation of a robust market for corporate control, merger control ought to go about these tasks without blocking combinations that are benign or procompetitive. The fulfillment of these objectives has important links to other areas of competition law.2 If competition policy misses the dominance issue, mergers can create durable market power with consequent adverse effects on prices, quality, and innovation. If it overlooks a transaction’s contribution to oligopolistic interdependence, a merger can build a market configuration in which the surviving firms either find it easier to establish effective cartels by a direct exchange of assurances or to use indirect means to realize results equivalent to those that an express agreement would yield. Because competition law has never addressed dominance or tacit collusion with great success, it is important to design merger policy so that proper choices are made as to when authorities intervene.3 * Professor, George Washington University Law School (At the time of writing: Commissioner, US Federal Trade Commission). The views expressed here are the author’s alone. 1 For a representative discussion of current issues, see “Roundtable Discussion on Developments – and Divergence – in Merger Enforcement”, 23 Antitrust 9 (Fall 2008). 2 See William Kovacic, Robert Marshall, Leslie Marx and Steven Schulenberg, “Quantitative Analysis of Coordinated Effects”, 76 Antitrust Law Journal 397 (2009) (discussing connections between merger policy and learning from antitrust enforcement against cartels). 3 In an unpublished lecture at the Federal Trade Commission in the early 1980s, I recall Philip Areeda borrowing a Cold War metaphor from George Kennan to describe merger control. Areeda said merger policy was antitrust law’s program of “containment” because it sought to avoid the expansion of dominance and the growth of oligopolistic market structures which invited tacit coordination that yielded cartel-like results but generally evaded effective intervention by competition bodies.

254  Merger Control in European and Global Perspective In most jurisdictions, the competition agencies evaluate transactions before the parties complete them.4 This process is unavoidably predictive and, in some instances, speculative. In a wide range of matters, no analytical calculus provides a sure way to distinguish transactions that pose anticompetitive dangers from those which promise to be benign or procompetitive. The examination of a proposed transaction often involves difficult, probabilistic assessments of future commercial developments. This is especially true in markets that display high levels of dynamism owing to technological or organizational innovation, or to developments in trade, transport, and communications that link previously isolated geographic regions into unified commercial markets. Each possible course of action by a competition agency poses risks. Block the deal improvidently, and valuable economic benefits from consolidation are lost. Accept the wrong divestitures or conduct-related undertakings as a condition of allowing the deal to proceed, and the agency creates an illusion of effective control that masks future anticompetitive results. Unwisely allow the deal to proceed as proposed, and consumers suffer the costs of diminished economic performance. The public statements of agencies concerning specific decisions to intervene or take no action ordinarily either acknowledge no risks associated with the choice taken or assert that all risks were thoughtfully and correctly weighed. It takes a high and unusual level of institutional self-assurance to state that the chosen course of action could be wrong. Seen in the aggregate, public enforcement policy over time reflects more humility about the analytical quandaries and difficult judgments associated with merger control than do the agencies’ portrayal of individual episodes of review. The history of merger review has featured what best can be seen as a series of experiments through which the public agencies have used various analytical and procedural measures to improve the accuracy of the predictive process. Modern commentary tends to accept the view that contemporary analytical methods are superior to predecessor techniques, but that may be because many contemporary commentators played some part in creating the modern techniques. Much remains to be done to improve procedures and substantive analysis, particularly for what might generally be labeled as the hard cases. So how are we to tell if a competition system is doing a good job of the important, forward-looking exercise of merger control? This question ought to be an important focus of attention in discussions about international competition law. Countries with merger control regimes should see themselves as having a common, abiding interest in the identification of substantive standards and procedural mechanisms that best promote improvements in economic performance. To determine whether we are making progress toward a better international framework for merger review, it is necessary to have some means for assessing the quality of the status quo. 4 Many competition policy regimes oblige the parties to notify the public enforcement agency of a proposed transaction. In these systems, the parties may not complete the consolidation until the agency has had a period of time to analyze the likely competitive effects. In a number of other systems, premerger notification and review are optional, but many companies choose to report proposed mergers in advance and allow the authority to review them before the integration of assets takes place.

William Kovacic 

255  

A popular and seemingly irresistible technique for measuring the quality of a competition policy system and its merger control regime is to study how often it blocks deals, allows deals, or subjects proposed transactions to elaborate analysis.5 Commentators lean on this method so often and so heavily that they forget its frailties. To say that an agency is doing a lot of things or only a few things does not tell us whether it is doing the right things. In sport, coaches admonish athletes not to equate activity with accomplishment.6 So it should be for merger control. There is a debate worth having, and that is whether antitrust oversight of mergers is improving or retarding economic performance. Answers to questions about actual economic effects will not emerge from the study of activity levels, unless we bravely (and dubiously) assume that specific levels of enforcement activity invariably or typically beget good results. Especially amid larger contemporary debates about the correct form of government intervention in the economy, we cannot rely on these feeble proxies to assess effectiveness. When competition policy agencies ask external audiences to accept the value of antitrust intervention on faith, they are likely to hear variants of the aphorism: “In God we trust; all others provide data.”7 The relevant data cannot be found in simple counts of merger reviews and challenges. This paper suggests how a jurisdiction might best go about evaluating the quality of its competition policy system. It urges competition agencies and collateral institutions to strive to improve our ability to measure the effects of merger control and to verify the consequences of different approaches to enforcement.8 The paper seeks to encourage the recent trend within the global competition policy community of accepting a norm that focuses greater attention on the evaluation of the effects of enforcement decisions – especially by developing better quantitative measures of actual economic effects9 – and the assessment of the processes by which competition agencies examine individual transactions.10 Among other sources, it draws upon the results of a self-study of the U.S. Federal Trade Commission (FTC) completed early in 2009.11

In US parlance this is the “second request.” In the European Union, it is the second phase inquiry. This advice seems to have untraceable antecedents. 7 I thank David Hyman for bringing this caution to my attention. 8 For an earlier treatment of this theme, see William Kovacic, “Evaluating Antitrust Experiments: Using Ex Post Assessments of Government Enforcement Decisions to Inform Competition Policy,” 9 George Mason Law Review 843 (2001). 9 The case for increased efforts to conduct quantitative studies of the effects of merger policy is presented in Dennis Carlton, “The Need to Measure the Effect of Merger Policy and How to Do It”, 22 Antitrust 39 (Summer 2008). 10 These trends are reviewed in William Kovacic, “Using Ex Post Evaluation to Improve the Performance of Competition Policy Authorities”, 31 Journal of Corporation Law 503 (2006). 11 See The Federal Trade Commission at 100: Into Our 2d Century (January 2009), http://www.ftc. gov/os/2009/01/ftc100rpt.pdf. 5 6

256  Merger Control in European and Global Perspective

What is “good” merger policy? Three suggested criteria Discussions about competition policy tend in a colloquial way to ask whether public enforcement agencies are doing a “good” job of carrying out their responsibilities. This form of discourse seldom involves a careful specification of what constitutes “good” performance. Expressly or implicitly, levels of enforcement activity are the foundation for judgments. In the case of merger policy, an appropriate assessment of the quality of merger policy should focus on three criteria. First, has merger policy improved economic performance by reducing the price or improving the quality of goods or services? This is the essential question about the effectiveness of merger policy. It is worth asking and debating regularly. A merger review system accomplishes this result by intervening to correct or preclude transactions that pose serious competitive dangers and by allowing combinations that promise to have benign or procompetitive effects. The second criterion is whether individual competition systems minimize unnecessary implementation costs within and across jurisdictions. Enforcement agencies should seek to achieve a given level of monitoring and enforcement at the lowest possible cost to society.12 Among other means, a jurisdiction can eliminate unnecessary burdens associated with its own notification procedures and investigations, promote international standardization based on superior techniques, and raise levels of interoperability across competition systems. The third criterion is whether a competition system has committed itself to a process of continuous reassessment and improvement.13 This has two dimensions. The first deals with the testing and improvement of methods to assess the economic consequences of individual decisions to intervene or not to intervene, and the aggregate effects of a system of merger review. The second involves an examination of the procedures for merger review and an analysis of whether the jurisdiction can achieve a given level of oversight at lower cost. Improvements in both dimensions require competition authorities to make meaningful disclosures about decisions to prosecute or not prosecute, to maintain and reveal informative data sets about activity levels, and to refine techniques – with the agency’s resources and in cooperative ventures with external bodies such as research institutions – for measuring actual economic effects of intervention decisions. Discussions about the quality of merger policy ought to dwell upon whether a system of competition law satisfies the criteria sketched above. Such discourse frequently does not. In many instances, assessments of merger policy do not define normative criteria clearly or apply them systematically. In other cases, problems 12 For statements of this normative aim and critical assessments of the efforts of the U.S. enforcement agencies to achieve this goal, see Joe Sims et al., “Merger Process Reform: A Sisyphean Journey?”, 23 Antitrust 60 (Spring 2009); Joe Sims and Deborah Platt Majoras, “The Effect of Twenty Years of HartScott-Rodino on Merger Practice: A Case Study of Unintended Consequences Applied to Antitrust Legislation”, 65 Antitrust Law Journal 865 (1997). 13 See The Federal Trade Commission at 100, cited above note 11, at pp. 22–23.

William Kovacic 

257  

associated with the measurement of merger enforcement consequences cause commentators to run away from the issue of actual economic effects. The means for determining the economic effects of merger policy are not ideal.14 In practice, it can be difficult to determine how merger control, in individual cases or across a range of intervention opportunities, affects economic performance. Owing to problems of measurement, the antitrust community ordinarily succumbs to the temptation to duck the ultimate question of economic effects.15 Discussions about the quality of merger enforcement instead use a variety of effectiveness proxies. Three stand out. The foremost fallback is to trace and analyze levels of activity, such as the total number of government interventions over a period of time or the percentage of all transactions in which the competition agency conducts an elaborate inquiry or takes action to modify or block a deal. By this measure, enforcement quality is inferred from rates of action or inaction. A second popular evaluation technique is to seek out the opinions of practitioners about the quality of the competition authority’s performance. Is it challenging too many deals, or too few? Are remedies too weak or too strong? Does the agency have sound processes in place for sorting out the good and the bad? Compared to other eras of competition policy, is it easier, or more difficult, to get a merger approved under the current enforcement agency leadership? In principle, practitioner views are a valuable source of information, and commentators and competition authorities ought to seek them out. In practice, the published reports of practitioner views that appear in the literature about merger review tend to be qualitative, unsystematic, and unverifiable. As a group, the reports of practitioner views generally provide a haze of unattributed impressions that no outsider can test rigorously. Some commentary offers the vastness of the narrator’s own experience as authority that an asserted proposition captures a broad, important reality.16 Other articles and press reports quote unidentified individuals with the suggestion that the speaker has revealed a universal, fundamental truth.17 There have been limited, somewhat broader efforts to conduct practitioner surveys, but these provide little discussion about the specific transactions that provide the basis for the participants’ qualitative views, and the identities of the participants invariably are anonymous. The anonymity may be necessary to avoid retribution by a competition agency that dislikes the speaker’s opinion, but anonymity also relaxes the speaker’s incentives to portray events fully and accurately. 14 See Carlton, cited above note 9, at p. 39 (noting that the dearth of quantitative studies and measures of effectiveness “means that there is no reliable guide for determining whether our antitrust policy is too lax in some areas and too stringent in others”). 15 Dennis Carlton has observed: “Antitrust analysis of proposed mergers has become increasingly sophisticated. Evaluation of antitrust policy has not.” Ibid. at 42. 16 See, e.g., Thomas Krattenmaker and Robert Pitofsky, “Antitrust Merger Policy and the Reagan Administration”, 33 Antitrust Bulletin 211, 228 (1988) (“Our experience has been that the U.S. business community has read the enforcement actions of the Reagan administration as an invitation to everyone to merger with anyone.”). 17 See, e.g., Thomas Greaney, “Merger Mania Has Gone Too Far”, St. Louis Post-Dispatch (27 February 1991), at 3B (“At the height of the Reagan administration’s permissiveness toward corporate mergers, a former assistant attorney general with the Carter administration summarized the advice he was giving clients: ‘I simply tell them that there’s no merger not worth trying.’”).

258  Merger Control in European and Global Perspective The third approach is to present specific enforcement episodes as exemplars of the competition agency’s philosophy about merger control. By offering an exemplar, the commentator asks the reader to draw broader conclusions about whether the competition agency’s analytical methods and ultimate conclusions are sound.18 Case studies can be extremely informative in what they say about the agency’s philosophy, analytical perspectives, and methodology. Yet individual enforcement episodes too often are analyzed in isolation. To be reliable as a way to make larger judgments about the quality of merger enforcement quality, one needs a sufficiently large number of case study observations to know how the agency is performing in any single period or across periods. For example, comparisons of enforcement choices in specific sectors over time can help illuminate adjustments in policy and technique, and to offer insights about how a collection of consolidation events affected sectoral performance.

An improved approach for evaluation Activity levels, practitioner perspectives, and the occasional case study provide the main ingredients for discussions of merger policy. The development of a superior performance evaluation methodology for horizontal merger enforcement and other forms of competition policy can take advantage of a growing body of experience and scholarship with the subject. Improvements in existing evaluation programs and extensions of the methodological state of the art might proceed along several paths. One is to engage competition authorities and researchers in more extensive collaborative discussions about existing projects and in explorations of evaluation techniques. This can take place in a variety of multinational and regional forums such as the International Competition Network and the Organization for Economic Cooperation and Development. In recent years, these and other organizations have shown an increased interest in operational issues, including performance management. Another way is for competition agencies to form partnerships with major research institutions. One can imagine the formation of collaboration projects by which competition authorities engage in intensive side-by-side discussions of commonly reviewed completed matters. Focal points for these assessments would be the types of evidence that each agency relied upon to formulate its decisions and consideration of how alternative procedures – for example, different methods for testing evidence – might affect substantive outcomes. These collaborations also could study the degree to which post-merger experience was consistent with the assumptions that led to agency action or inaction. 18 See, e.g., Jonathan Baker and Carl Shapiro, “Reinvigorating Horizontal Merger Enforcement”, in Robert Pitofsky, ed., How Chicago Overshot the Mark, OUP, 2008, pp. 235 et seq., at pp. 248–251 (discussing DOJ decision not to oppose the merger of Whirlpool and Maytag).

William Kovacic 

259  

A second element is for competition authorities to expand resources devoted to performance measurement. Agencies can ensure that, in every budget cycle, there are outlays for evaluation. The performance measure exercises can be carried out by agency insiders, external consultants, or some combination of the two. Competition authorities with common interests and common investigations usefully could cooperate to do relevant research. Focal points for collaboration would include the assessment of economic effects and of the processes for merger control. In making budget outlays, agencies should view performance measurement as an integral element of the policy making life cycle and not simply as a luxury. Performance measurement investments are part of the policy R&D by which a public competition authority grows smarter. A third element is to continue and extend the trend of publishing fuller data sets on merger enforcement activity. For the US federal agencies, for example, this means an acceleration of the recent trend to publish accounts of decisions not to prosecute and to issue reports on major variables affecting the decision to prosecute. These disclosure measures could be coupled with workshops and seminars that rely on these and other materials to discuss enforcement trends and effects. These techniques will help to build and reinforce an ethic of self-assessment and continuous improvement. They underscore the importance of institutional improvement as a necessary complement to advances in doctrine or theory. Good policy runs on an infrastructure of institutions, and broadband-quality policy cannot be delivered on dial-up quality institutions. If one asks whether the U.S. antitrust agencies have got things just right today, the answer yesterday and today is no. If one asks whether there are measures in progress to get there, the answer is emphatically yes. Better answers to the question of how to assess actual economic effects of enforcement will be key ingredients for reaching that destination.

Adam Fanaki*

Convergence in Multi-Jurisdictional Merger Reviews: A Canadian Perspective

[I]t must be recognized that the proliferation of merger control regimes is imposing significant – and unnecessary – transaction costs on virtually all international transactions, and in particular on those transactions which do not raise any significant competitive concerns whatsoever.1

I. Introduction The transnational nature of mergers in a globalized economy and the marked increase in the number of jurisdictions with merger control laws raise unique challenges for both competition agencies and merging parties. There are now at least 90 jurisdictions with a merger control regime, a significant increase from the handful of jurisdictions reviewing mergers only 15 years ago. The proliferation of national merger control laws presents a number of benefits, such as protecting markets from mergers that could substantially lessen or prevent competition and allowing for reviews that account for the unique legal and economic context of each jurisdiction. However, subjecting the same transaction to scrutiny by multiple national competition agencies may also result in greater costs of compliance, an unnecessary duplication of resources, increased risks of inconsistent remedies, uncertainty over legal treatment and jurisdictional disputes. Ultimately, a fragmented approach to the enforcement of competition laws has the potential to chill pro-competitive activity. Although inconsistent enforcement decisions in merger reviews are rare, such conflicts pose considerable challenges for agencies and merging parties. The most prominent examples involve divergent outcomes in the review of mergers by the European Commission and U.S. antitrust agencies, such as the GE/Honeywell merger. However, with the recent proliferation of national merger control laws, the potential for conflicting decisions is no longer confined to the transatlantic context, and now includes concerns that merger reviews in China, India or other * Adam Fanaki is a partner in the Competition & Foreign Investment Review and Litigation Groups of Davies Ward Phillips & Vineberg LLP. Prior to joining Davies, he acted as Senior Deputy Commissioner of Competition, the head of the Mergers Branch at the Canadian Competition Bureau. 1 Business and Industry Advisory Committee to the OECD, “Recommended Framework for Best Practices in International Merger Control Procedures” (4 October 2001) at 2, http://www.biac.org/ statements/comp/BIAC-ICCMergerPaper.pdf.

262  Merger Control in European and Global Perspective relatively new merger control regimes “could delay or thwart high-profile crossborder mergers and acquisitions”.2 To date, the majority of the efforts by competition authorities to address the challenges of multi-jurisdictional merger reviews have focused upon increasing coordination, such as through greater information-sharing between reviewing jurisdictions, and harmonizing remedies. However, in 2009, the Canadian government took a significant step towards procedural convergence by implementing sweeping reforms that align Canada’s competition laws more closely with those of the United States. The amendments to the Canadian Competition Act (the “Act”) enacted a new two-stage, merger review process that is very similar to the U.S. Hart-Scott-Rodino process, including an initial 30-day waiting period that may be extended by the issuance of an information request similar to a “Second Request” by the Commissioner of Competition. Although many supported the concept of procedural convergence as a means of reducing the cost of multi-jurisdictional merger reviews, the reforms to the Act’s merger review process met with significant opposition within business and legal communities in Canada. Specifically, a number of commentators expressed concerns about the “Americanization of Canadian competition law” and the implementation in Canada of more controversial aspects of the U.S. merger regime, such as the “Second Request” process for gathering information. As one commentator stated: The U.S. system has been criticized around the world for the long delays and hefty costs it imposed on companies facing merger reviews…Indeed dozens of other countries have revised or adopted new competition rules over recent decades, and none have copied the U.S. model.3

This paper discusses reforms to the Canadian merger review process as an illustration of the potential benefits and challenges associated with procedural convergence in merger control laws. In addition, this paper considers whether in the limited period since its enactment, the revised Canadian merger review process has increased opportunities for coordination and cooperation with respect to mergers subject to review in both Canada and the United States.

II. The challenges of multi-jurisdictional merger reviews The costs associated with multi-jurisdictional merger reviews have been the subject of consistent commentary and study over the past decade. For example, Konrad von Finckenstein, a former Canadian Commissioner of Competition, 2 Sundeep Tucker and Patti Waldmeir, “Asian Antitrust Laws Threaten Deals”, Financial Times (27 July 2008). 3 Janet McFarland, “The Americanization of Canada’s Competition Act”, Globe and Mail (10 February 2009).

Adam Fanaki 

263  

recognized in 2001 that the “growing multiplicity of antitrust laws means that some large mergers…are subject to review by numerous jurisdictions” and that “such multiple reviews may impose real costs on the parties, costs that might sometimes function as a tax on efficient transactions”.4 Indeed, where market participants are faced with uncertainty as to the outcome of a merger review, they may ultimately decide to abandon a proposed transaction. This is particularly true where the effort and expense required to address the concerns of multiple authorities are viewed as being disproportionate to the advantages that would have been realized by the transactions in question. In 2002, the International Bar Association and American Bar Association commissioned PricewaterhouseCoopers LLP (“PWC”) to conduct a survey of the time and cost associated with multi-jurisdictional merger reviews. The results were published in a 2003 study which found that the average duration of a review of a transaction was approximately seven months and the average external cost (including legal fees, merger filing fees and the fees for other advisors) to the merging parties was approximately €3.28 million.5 Although PWC did not measure the internal costs of the merging parties in financial terms, the estimated time spent by company personnel on merger review processes ranged from an average of 28 person-weeks for all transactions considered in the study to 120 personweeks for the transactions subject to in-depth merger review. While constituting only a small proportion of the overall transaction value, the costs of merger review were estimated by PWC to amount to approximately 42% of the overall cost of executing the transaction. It is likely that the external costs of modern merger reviews are even higher today than those identified in the PWC study. At least part of these higher costs stem from the fact that it is increasingly common for mergers of any significance to be reviewed by multiple competition authorities. For example, the AlcanPechiney-Alusuisse merger was subject to review in 16 jurisdictions and required notifications in eight languages, with 35 law firms providing advice on the transaction.6 More recently, the merger of Merck & Co. and Schering-Plough Corporation in 2009 was notified in Argentina, Australia, Brazil, Canada, China, Colombia, European Community, Israel, Japan, Mexico, New Zealand, Pakistan, Russia, Serbia, South Africa, South Korea, Switzerland, Taiwan, Turkey, Ukraine and the United States.7 4 Konrad von Finckenstein, “International Antitrust Cooperation: Bilateralism or Multilateralism?,” Address to joint meeting of the American Bar Association Section of Antitrust Law and Canadian Bar Association National Competition Law Section (May 31, 2001), http://cb-bc.gc.ca/epic/internet/incbbc.nsf/en/ct02240e.html. 5 PricewaterhouseCoopers, “A tax on mergers? Surveying the costs to business of multi-jurisdictional merger review,” (June 2003), http://www.pwcglobal.com/uk/eng/about/svcs/vs/pwc_mergers.pdf. 6 See Jacques Bougie, “Reflections on the Merger Task Force at the Turn of the Millennium: the Requirement for Convergence of Multijurisdictional Merger Review Systems,” European Commission Merger Task Force’s 10th Anniversary Conference (September 2000). 7 See, e.g., M.5502, Merck/Schering-Plough (22 October 2009); “Merck and Schering-Plough Receive Approvals From U.S. Federal Trade Commission, Swiss Competition Commission and Canadian Competition Bureau for Merger” (29 October 2009), http://www.merck.com/newsroom/ news-release-archive/corporate/2009_1029.html.

264  Merger Control in European and Global Perspective The costs associated with multi-jurisdictional merger reviews include costs of ascertaining whether notification and filing is required in a given jurisdiction, preparing and submitting notifications, paying sometimes significant filing fees, complying with multiple information requests, making submissions to various competition authorities, participating in witness interviews or depositions, and negotiating remedies. In addition to these external costs, there are also costs associated with postponing the consummation of the transaction, such as delaying the realization of efficiencies that may be generated from the integration. Multiple merger reviews may also result in increased uncertainty, which provides greater opportunities for rivals to take customers or staff from the target firm. In addition to the costs to merging parties, overlapping reviews may also result in an unnecessary duplication of resources between reviewing agencies and other costs to the public. Finally, there is the prospect of inconsistent or contradictory results of merger reviews, along with the possibility that when faced with significant uncertainty, parties may elect to abandon a potentially efficiency-enhancing integration. Recognizing the potential costs of multi-jurisdictional merger reviews, competition authorities, academics and practitioners have devoted significant efforts to the search for methods of minimizing the costs of transnational merger control and reducing the risk of inconsistent outcomes or incompatible remedies. Such efforts include the continuing work of the Merger Working Group of the International Competition Network, the OECD and other organizations which are studying and developing best practices for merger review and procedure. The efforts of these organizations and participating competition agencies have led to increased bilateral cooperation, as well as heightened coordination and consistency between national merger regimes. Nevertheless, despite this progress, parties continue to face significant costs, uncertainty and duplication in multijurisdictional merger reviews. There have been a number of different proposals designed to bring about greater convergence in merger reviews. These proposals range from the implementation of a supranational merger control agency to greater procedural convergence, such as the adoption of common merger notification forms, coordinated merger review processes and the application of similar notification thresholds. Proponents of procedural convergence note that having a similar merger review process may substantially reduce the costs associated with multi-jurisdictional merger reviews and also provide greater opportunities for coordination between reviewing jurisdictions. As discussed in the next section of this paper, the 2009 reforms to the Canadian Act implement a significant step towards procedural convergence with the U.S. merger review process.

Adam Fanaki 

265  

III. Procedural convergence in Canada-U.S. merger reviews The recent reforms to the Canadian merger review process followed several decades of largely unsuccessful efforts to modernize the Act. In 2007, the Canadian Ministers of Finance and Industry announced the creation of the Competition Policy Review Panel, comprised of senior business leaders charged with reviewing Canadian competition and foreign investment policies, with a view to making Canada more competitive globally. In its report, titled “Compete to Win”,8 the Panel found that innovation “underpins the fastest growing industries and high-wage jobs, provides the tools needed to compete in every business today and drives growth in all major countries and in every sector”.9 However, the Panel also found that a major cause – if not the cause – of a growing productivity gap between Canada and other industrialized nations was a weak record of innovation. In this regard, the Panel stated: Much of Canada’s poor productivity performance can be attributed to the comparatively poor performance of Canadian firms with respect to innovation. We rank poorly across almost all aspects of innovation: the creation of knowledge, the diffusion of knowledge, the transformation of knowledge and the use of knowledge through commercialization. This is seen by the Conference Board of Canada as “a serious weakness in Canada’s overall performance and [an] alarming portent for the future”. Other research also indicates that Canadian firms lag behind firms in other major industrialized countries on a number of measures of innovation.10

In the Panel’s view, bridging this productivity gap required reforms to Canada’s competition and foreign investment laws. The Panel found that effective competition laws “are key elements in ensuring the competitiveness and efficiency of the Canadian economy”11 and noted that “long-term improvements to Canada’s productivity could be achieved by amending certain outmoded or ineffective provisions of Canada’s competition laws”.12 Ultimately, the Panel recommended significant amendments to the Act that were aimed at increasing its predictability and effectiveness, as well as providing greater protection for Canadian consumers. With respect to merger review, the Panel identified a number of shortcomings in the Canadian process. To address these issues, the Panel concluded that, where practicable, Canada’s competition laws should be harmonized with the laws of the United States, “with a view to minimizing unnecessary procedural or substantive differences, given the high level of integration of business operations in the two countries”.13 The Panel explicitly identified benefits of procedural and substantive 8 Competition Policy Review Panel, Compete to Win: Final Report – June 2008 at 53, http:// www.ic.gc.ca/epic/site/cprp-gepmc.nsf/vwapj/Compete_to_Win.pdf/$FILE/Compete_to_Win.pdf (“Compete to Win”). 9 Compete to Win at 92. 10 Ibid. at 18. 11 Ibid. at 53. 12 Ibid. 13 Ibid.

266  Merger Control in European and Global Perspective harmonization with other jurisdictions, including greater confidence in the Canadian merger review process by foreign investors: Merger review is a feature of every modern economy. Increasingly, the most significant mergers are international in scope. It is important for Canada to have a voice along with the competition agencies of other countries that are engaged in the review of mergers affecting Canada’s economic interests. Consequently, using an analytical approach and regulatory process that is convergent with our major trading partners should not only help the Competition Bureau conduct its work but also reassure international investors that Canadian competition laws in respect of mergers are modern and transparent.14

Consistent with the Panel’s recommendations, the federal Government enacted reforms to the merger regime to create a two-stage review process, very similar to the merger review process in the United States. Under the amended process, the Bureau will review the vast majority of transactions within an initial 30-day waiting period during which the parties may not proceed to close the transaction. For transactions that raise material competition issues, and for which additional information is required, the Commissioner may issue a Supplementary Information Request, or “SIR”, to the merging parties seeking additional information relevant to the Bureau’s assessment of the proposed transaction. As with the Second Request process in the United States, the issuance of an SIR triggers a second waiting period that expires 30 days after the parties have supplied the information requested in the SIR. Although many supported the concept of procedural harmonization between merger control regimes, the adoption in Canada of an information-gathering process similar to the U.S. “Second Request” prompted significant concerns. Much has been written regarding the potential burdens of the Second Request process in the United States, including the potentially high costs of compliance and possible delays.15 Similar concerns were echoed strongly by Canadian businesses and their advisors regarding the SIR process in the periods leading up to and following the amendments.16 For example, the Merger Streamlining Group, an organization whose mandate includes advocating that “merger review regimes in over 70 countries operate effectively and efficiently and do not impose undue compliance burdens”, made the following submission regarding the amendments to the Canadian merger review process: Ibid. at 55. See, e.g., comments by the American Bar Association Section of Antitrust Law, Submission Regarding the Hart-Scott-Rodino Second Request Process (7 December 2005), http://govinfo.library. unt.edu/amc/public_studies_fr28902/merger_pdf/051207_ABA_Mergers_ HSR.pdf; International Chamber of Commerce, Comments on selected issues for study by the US Antitrust Modernization Commission (5 September 2005), http://govinfo.library.unt.edu/amc/public_studies_fr28902/ enforcement_pdf/050905_ICC_revised.pdf. 16 See, e.g., J. William Rowley, Letter to the Commissioner of Competition from the Merger Streamlining Group (29 May 2009), http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/vwapj/ Submission%20re.%20Draft%20Merger%20Review%20Process%20Guidelines-McMillan. PDF/$FILE/Submission%20re.%20Draft%20Merger%20Review%20Process%20GuidelinesMcMillan.PDF; Canadian Chamber of Commerce, Submission to the Competition Bureau (26 May 2009),http://competitionbureau.gc.ca/eic/site/cb-bc.nsf/vwapj/mergers%20review%20processCCC-%20May-09%28FINAL%29.pdf/$FILE/mergers%20review%20process-CCC-%20May09%28FINAL%29.pdf. 14 15

Adam Fanaki 

267  

The merger review process described in the Draft Guidelines is very similar to the process followed by the United States Federal Trade Commission (“FTC”) and Department of Justice (“DOJ”) under the Hart-Scott-Rodino Antitrust Improvements Act of 1976 (the “HSR Act”). The second request process under the HSR Act has raised significant concerns due to the costs and burdens it imposes on both merging parties and the reviewing agencies, and the extended period of time that is required to complete a merger investigation. For example, a 2003 survey by the American Bar Association Section of Antitrust Law found that the average costs of complying with a second request were US$3.3 million. Responding to second requests in more complex investigations can cost substantially more than the US$3.3 million average. A 2003 survey by PricewaterhouseCoopers found that the costs to merging parties of a merger review in the United States were at least double the costs in any other jurisdiction. In addition, the average length of an HSR investigation is six months or more. The Group believes that the United States second request process is not in compliance with either ICN Recommended Practice VI-E or the fifth of the ICN Guiding Principles for Merger Notification and Review (“Guiding Principles”) (“Efficient, timely and effective review”).17

Generally, those expressing concerns advocated that certain measures be taken to differentiate the Canadian merger review process. Ultimately, although the Canadian merger review process is very similar to the process applied in the U.S., it has also retained a number of uniquely Canadian features. As the Canadian Commissioner of Competition remarked when discussing the implementation of the revised merger review process: I will readily admit that we borrowed heavily from our U.S. counterparts but, at the same time, we were careful to design a “made in Canada” approach. By that, I mean an approach that is sensitive to our context and tradition, where flexibility and creativity (necessities under our former regime, which didn’t fit the demands of the process for the parties or the Bureau) have played a significant role in merger review.18

Examples of the “made in Canada” approach include a limitation which only permits the Commissioner to challenge a transaction within one year following closing.19 No equivalent limitation period exists in the United States.20 Additionally, in Canada, parties are able to apply to the Bureau for approval of a transaction through a summary process called an application for an Advanced Ruling Certificate (“ARC”), which may be issued in situations where the Commissioner is satisfied that the proposed transaction does not raise competition concerns sufficient to warrant a challenge before the Competition Tribunal.21 The issuance of an ARC prevents the Commissioner from bringing a future application to challenge the merger on the basis of information that is the same, or substantially the same, as the information used for the granting of the certificate. The American process does not provide for an equivalent pre-notification merger approval and J. William Rowley, cited previous footnote, at 2–3. Melanie Aitken, “Remarks to ABA/US Chamber Event”, 1 February 2010, http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03216.html. 19 Competition Act, section 97. 20 See Federal Trade Commission, What is the Premerger Notification Program: An Overview (March 2009) at 2, http://www.ftc.gov/bc/hsr/introguides/guide1.pdf. 21 Competition Act, section 102(1). 17 18

268  Merger Control in European and Global Perspective filing exemption; however, parties are permitted to request “early termination” of the initial 30-day waiting period where the proposed transaction is not overly complex or is not likely to result in substantial harm to competition. On September 18, 2009, the Competition Bureau released its Merger Review Process Guidelines22 which also incorporated a number of additional distinguishing features. For example, the Merger Guidelines state that the Bureau will generally share a draft of the SIR in advance of issuance and engage the parties in a dialogue to: ensure that the party understands the information requests; discuss whether the party maintains data in the form requested by the Bureau ... identify confidentiality concerns; determine whether there are sources and forms of information that may be more directly responsive to the Bureau’s request; and ascertain whether there are any other issues that might impair the ability of the party to comply with the SIR...23

In addition, the scope of questions and document requests found in a SIR has, at least thus far, been narrower than those found in the model Second Request.

IV. Experience under the revised merger review process Over a year-and-a-half has passed since the enactment of the revised merger review process in Canada and so it seems appropriate to begin to consider the actual impact of the recent reforms on multi-jurisdictional merger reviews involving Canada. It is often difficult to compare merger enforcement over time, given differences in overall economic conditions. This is especially true during the atypical period for merger review that began in late 2008 with the financial crisis and resulting economic downturn. Although the number of transactions notified in the past 18 months is lower than in previous years, the Bureau has received an unprecedented number of highly substantive and complex transactions, which provide a meaningful basis for assessing the new regime. In the Merger Review Performance Report published on May 31, 2010,24 the Bureau disclosed statistics relating to merger reviews conducted during the first year of the amended merger regime. Not surprisingly, given the prevailing economic conditions, these statistics show a significant decline in the number of transactions notified to the Competition Bureau. Specifically, for the period April 1, 2009 to March 31, 2010, the Bureau reviewed 216 transactions, which is approximately 36% less than during the equivalent period in 2008 to 2009. 22 Canadian Competition Bureau, Merger Review Process Guidelines (18 September 2009), http:// www.bureaudelaconcurrence.gc.ca/eic/site/cb-bc.nsf/vwapj/merger_review_process-e.pdf/$FILE/ merger_review_process-e.pdf (“Merger Guidelines”). 23 Ibid. at 8. 24 Canadian Competition Bureau, “Merger Review Performance Report” (May 2010), http:// www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/vwapj/Merger-Performance-Report-May-2010-e. pdf/$FILE/Merger-Performance-Report-May-2010-e.pdf.

Adam Fanaki 

269  

What is perhaps more surprising is the number of consensual remedies in mergers cases, including six Consent Agreements registered during the first year of the new merger regime. As a point of comparison, the Bureau secured the same number of merger remedies in its first year under the new regime as it did in all of 2006, 2007 and 2008 combined. These cases ranged from multinational pharmaceutical mergers (Pfizer/Wyeth and Merck/Schering) to mergers involving local waste disposal sites (Clean Harbors/Eveready). This trend has continued into the 2010 fiscal year with the registration of four additional Consent Agreements (Novartis/Alcon, IESI-BFC Ltd. (BFI)/Waste Services Inc. (WSI), Teva Pharmaceutical Industries Ltd./Ratiopharm (Merckle Group)) and The Coca-Cola Company. As of September 2010, the Bureau has issued SIRs with respect to ten mergers, or approximately 3% of all mergers reviewed during the first 18-months of the amended merger regime.25 Among the benefits anticipated to arise from the recent amendments to the Canadian merger review regime was “more productive and effective”26 collaboration between U.S. and Canadian agencies with respect to international merger reviews. Even prior to the recent amendments, the Canadian Competition Bureau and U.S. agencies have had a long history of cooperation and collaboration. While such collaboration was possible under the predecessor merger regime, it was more challenging given the differences in the Canadian and U.S. merger review processes. For example, the timing of the expiry of the waiting period between Canada and the United States was not aligned. Under the prior regime, for the most complex transactions, the waiting period would expire 42 days after initial notification of the transaction, irrespective of whether the merging parties had supplied the information requested by the Competition Bureau, unless the Bureau was successful in securing an order to extend this period. As such, the Competition Bureau was occasionally under pressure to reach substantive determinations prior to expiry of the Canadian waiting period, while the transaction remained at the initial review stage in the United States, Europe and elsewhere. As a result of the recent amendments, the timing of various stages of the merger review processes in Canada and the United States are often synchronized. Under the revised merger review process, merging parties generally file at or about the same time in Canada and the United States. This facilitates information exchanges on a number of common issues (such as market definition, entry conditions and case theories), joint interviews of witnesses, joint discussions with merging parties, and collaboration with respect to the design and implementation of remedies. In addition, aligned reviews facilitate other measures designed to allow parties to avoid unnecessary costs associated with potentially duplicative processes in responding to Second Requests in the United States and Supplementary Information Requests in Canada, such as: (i) using common lists of custodians and search terms for gathering documents; (ii) using common forms of specifications 25 Melanie Aitken, “Remarks to CBA Fall Competition Law Conference”, 30 September 2010, Gatineau, Quebec, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03306.html. 26 Melanie Aitken, “Remarks to ABA/US Chamber Event”, cited above note 18.

270  Merger Control in European and Global Perspective requesting the production of records or data and common relevant time periods, so as to reduce the need for multiple searches for relevant records or separate data runs; and (iii) accepting the response to a Second Request in the United States as a full or partial response to the Supplementary Information Request issued by the Canadian Competition Bureau. The degree of alignment between the Canadian and U.S. agencies has been particularly evident with respect to the design of remedies in recent merger cases. In a number of mergers, the Competition Bureau has adopted remedies that are closely aligned with remedies obtained by the U.S. DOJ or FTC, even where such remedies include behavioural commitments generally avoided by the Bureau. Examples of close alignment in merger remedies include the Consent Agreements registered in Ticketmaster/Live Nation, Pfizer/Wyeth, Merck/Schering and The Coca-Cola Company. Also, in a number of cases, the Canadian Competition Bureau either took limited or no action in light of remedies obtained by other competition authorities. Recent examples include the Dow Chemical/Rohm and Haas, Thomson/Reuters and Danaher Corporation/MDS Analytical Technologies transactions where the Bureau accepted remedies negotiated in the United States as the basis for remedies to address competition concerns in Canada. It is important to recognize that in these cases, the Bureau is not presumptively deferring to the findings of another agency. Rather, in each of these cases, the Bureau took steps to assess the likely competitive effects in Canada, for example by making initial market contacts or otherwise conducting its own research. As a result of this initial examination, the Bureau was satisfied that that it was unnecessary to implement a set of Canada-specific remedies. Although this paper has focused on convergence between merger reviews in Canada and the United States, there are also a number of examples of coordination with respect to remedies between the Canadian Competition Bureau and other reviewing agencies, such as the European Commission. For example, in the Aventis/Sanofi-Sythélabo transaction, worldwide divestitures accepted by the European Commission adequately addressed the Bureau’s concerns resulting from this merger. The high degree of coordination with respect to the design of remedies in multijurisdictional merger reviews has a number of benefits for both the Competition Bureau and merging parties. From the Competition Bureau’s perspective, being able to rely upon coordinated remedies in multi-jurisdictional mergers allows the Bureau to focus more of its limited enforcement resources on domestic matters. As the Commissioner of Competition recently explained: To be frank, we are the net beneficiaries in this arrangement. This cooperation and collaboration allows us to do far more with less. But what is most important from your perspective is the fact that we work closely. Ultimately, this is good for businesses that do not seek to evade honest competition. You should be aware of this for your merger work; and you need to recognize this in respect of your criminal files.27 27

Melanie Aitken, “Remarks to ABA/US Chamber Event”, cited above note 18.

Adam Fanaki 

271  

For merging parties, having a high degree of coordination between reviewing agencies with respect to remedies avoids the potential of conflicting remedies, allows for consistency in timing and permits merging parties to assess the totality of remedies required to secure global approval for a transaction, as opposed to dealing with each jurisdiction on a piece-meal basis.

V. Substantive convergence in merger reviews In recent years, there has been significant convergence among national competition authorities with respect to the substantive standards to be applied in examining mergers. Yet, there remain differences in certain aspects of the analysis, even among jurisdictions with long-standing merger control laws. For example, in Canada, unlike the United States or Europe, parties are entitled to defend a merger on the basis that the efficiencies resulting from the proposed transaction outweigh its likely anti-competitive effects, irrespective of whether those efficiencies can be shown to benefit consumers. While the recent amendments to the Act significantly altered the procedures applicable to merger reviews, they did not change the substantive approach the Bureau uses in reviewing mergers. However, on September 7, 2010, the Bureau announced that consultations would be held to consider potential revisions to the Merger Enforcement Guidelines (the “MEGs”).28 The MEGs were last updated in 2004 and describe the principal analytical techniques and evidence on which the Bureau relies to review mergers. The announcement was not unexpected, following the recent completion of a similar consultation process in the United States and the issuance of the revised U.S. Horizontal Merger Guidelines in August 2010.29 The analytical approach set out in the MEGs was clearly influenced by the 1992 version of the U.S. Horizontal Merger Guidelines, as is apparent from their close alignment on issues such as the use of the hypothetical monopolist test for defining markets, the emphasis placed on evaluating barriers to entry and a number of other common analytical factors. Nevertheless, there remain points of divergence in the analysis applied in Canada and the U.S., as reflected in the current version of the MEGs, which are largely due to differences in the governing legislation and jurisprudence. Most recently, on February 25, 2011, the Bureau announced that it will undertake “moderate” revisions to the MEGs.30 Among the proposed revisions 28 See Canadian Competition Bureau, (7 September 2010), http://www.competitionbureau.gc.ca/ eic/site/cb-bc.nsf/eng/03284.html. 29 See U.S. Department of Justice and the Federal Trade Commission, Horizontal Merger Guidelines (19 August 2010), http://ftc.gov/os/2010/08/100819hmg.pdf. 30 See Canadian Competition Bureau, “Bureau Announces Plans to Revise Merger Enforcement Guidelines” (25 February 2011), http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03350.html.

272  Merger Control in European and Global Perspective that the Bureau has identified are a clarification that merger review “is not a linear process that must start with market definition, but rather an iterative process in which evidence of market concentration is considered alongside other evidence of competitive effects” and “more detailed guidance on how the Bureau assesses the unilateral effects of a merger, particularly in light of current economic thinking”.31 It therefore seems likely that the consultative process will consider whether the Canadian MEGs should be revised to reflect changes in the analytical approach made to the U.S. Horizontal Merger Guidelines, such as a more fact-specific and less formulaic approach to reviewing mergers and explicit recognition of a number of analytical tools and economic concepts that may be applied to review a merger, including diversion ratios, critical loss analysis and an analysis of margins.

VI. Are multi-jurisdictional reviews now less costly? As discussed above, convergence between reviewing jurisdictions carries a number of benefits for both agencies and merging parties, including more coordinated remedies and the avoidance (or at least minimization) of inconsistent outcomes. However, it is difficult to determine whether greater convergence between the Canadian and U.S. merger review processes has, in fact, reduced the burden on merging parties in multi-jurisdictional merger reviews. It would appear that the recent reforms have not reduced the costs to merging parties of complying with the Canadian component of a multi-jurisdictional merger review. Although evidence on this point is not currently available, it is likely that the costs to merging parties of the merger review process in Canada have continued to increase following the implementation of the amendments, and in particular, merging parties likely face significant costs in responding to Supplementary Information Requests under the revised process. However, the increasing cost of merger review is also due, at least in part, to a number of different factors unrelated to the amendments, such as: a movement from structural presumptions to a more nuanced, facts-based analysis of competitive effects; growing reliance on more sophisticated methods of analysis, such as merger simulations; and, changes in the nature of products and services themselves. The type of analysis employed by the Bureau in recent merger reviews requires more significant amounts of information. For example, in recent years, the Bureau has used simulations and other analytical tools as methods of evaluating the competitive effects of mergers. However, use of simulations is wholly dependent on data availability and on the mathematical characterization of the market or markets at issue. Insufficient or inaccurate information requires enforcement agencies to apply a larger number of assumptions, thereby decreasing 31

Ibid.

Adam Fanaki 

273  

the simulation’s overall accuracy and effectiveness, and reducing the confidence with which the Bureau and parties may rely on the results. The analysis of mergers has also shifted from simply proving market concentration to establishing that the merger will likely lead to a reduction in competition: This change towards an effects-based approach profoundly affected the review process. While the market structure is still an important starting point, the real focus is now on facts and data, which suggest the future action of the merged firm. This has led to changes in the tools used in the analysis and to changes in the fact finding process as well.32

Movement away from structural presumptions, use of merger simulations and changes in the nature of products and services have all contributed to a more complex merger analysis which requires information that goes beyond simply defining the basic boundaries of relevant markets or levels of concentration to the more data-intensive exercise of predicting anti-competitive effects. The movement away from structural presumptions regarding the competitive impact of a merger towards a less formulaic approach is evident in the recent revisions to the U.S. Horizontal Merger Guidelines. Other factors that may increase the cost to merging parties of merger reviews, such as the need to respond to Supplementary Information Requests under tight deadlines in some cases and an increased use by the Bureau of Supplementary Information Requests as compared with investigatory orders (as had been the practice under the predecessor regime), are more directly attributable to the revised merger review process.

VII. Conclusion In recent years, efforts have intensified to promote convergence in competition laws – a process driven in large part by an increasingly global market, a growing number of multi-jurisdictional transactions, and a greater awareness of the negative effects on businesses that result from inconsistent outcomes and friction between the laws of national competition regimes. Consistent with this objective, the 2009 reforms to the Competition Act have aligned Canada’s competition regime more closely with that of the United States, yet the Act retains a number of uniquely Canadian aspects. Although convergence clearly provides a number of benefits, it is difficult to determine whether the recent reforms have, in fact, reduced the cost of multi-jurisdictional merger reviews involving Canada. Further study of this issue would be beneficial. While even greater procedural convergence in some 32 Sean Boland et al., “Remarks on Mergers, Cartels, and Single Firm Conduct – Panel Discussion, George Mason Law Review Antitrust Symposium, Washington, D.C., September 13, 2006”, 14 George Mason Law Review 879, 880 (2007).

274  Merger Control in European and Global Perspective areas (such as the use of a common merger notification form) could lead to lower costs of compliance for merging parties subject to notification in both jurisdictions, continued divergence between Canadian and U.S. competition laws also carries certain advantages. Among other things, divergence allows for experimentation across jurisdictions on the ideal manner in which to handle difficult competition issues and also reflects the unique legal context and traditions in each jurisdiction.

Tadashi Shiraishi*

Effects on Domestic Purchasers: A Descriptive Theory for Competition Law in Cross-Border Cases

I. Introduction The “effects doctrine” has been accepted in almost all jurisdictions as the ruling principle in the application of competition law to cross-border cases.1 However, little effort has been made to define the meaning of “effects” in a concrete manner. I propose to define it descriptively as “effects on domestic purchasers” (eDP). This article will show that the eDP theory can explain most issues in substantive law, and that this analysis will bring us to another frontier, the issue of borderless purchasers.2

II. The effects doctrine The effects doctrine now prevails in issues related to the application of competition law to cross-border cases. In the United States, the landmark decision of Alcoa3 effectively replaced territoriality principles, which historically were influential.4 Furthermore, in the European Union (EU), where a similar doctrine of “implementation” exists,5 it seems that the effects doctrine has been accepted in legal practice, especially in the area of merger control.6 * Professor at The University of Tokyo Graduate Schools for Law and Politics. 1 Applications of competition law to cross-border cases have usually been discussed using the term “extraterritorial application”. However, as this chapter shows, relevant topics include purchasers located abroad. In other words, the alleged violators may be domestically located. This is one of the main reasons why I decline to use the word “extraterritoriality”. 2 Although the topic of the 15th Annual EU Competition Law and Policy Workshop focused on merger control, it seems that the same substantive norm applies both to merger control and to the antitrust rules in cross-border cases. This is why I discuss not only mergers in this chapter but also cartels. 3 United States v. Aluminum Co. of America, 148 F.2d 416, 443–444 (2d Cir. 1945) (“We shall assume that the Act does not cover agreements, even though intended to affect imports or exports, unless its performance is shown actually to have had some effect upon them.”). 4 American Banana Co. v. United Fruit Co., 213 U.S. 347, 356 (1909) (“But the general and almost universal rule is that the character of an act as lawful or unlawful must be determined wholly by the law of the country where the act is done.”). 5 Joined Cases 89, 104, 114, 116, 117 and 125 to 129/85, A. Ahlstrom v. Commission [1988] ECR 5193, para.16 (“The decisive factor is therefore the place where it is implemented.”). 6 Case T-102/96, Gencor Ltd. v. Commission [1999] ECR II-753, paras. 90–92 (“Application of the Regulation is justified under public international law when it is foreseeable that a proposed

276  Merger Control in European and Global Perspective Taking another example, in Japan, the Japan Fair Trade Commission (JFTC) has been no exception; indeed, it has adopted a number of decisions endorsing the effects doctrine.7

III. The eDP theory 1. Introduction

As we have seen above, “effects” is a term that is accepted almost universally; however, a concrete definition has never been established. My proposal is to define “effects” as “effects on domestic purchasers” (eDP). In other words, “a market of our jurisdiction” 8 should be a market with its purchasers located in the area of the jurisdiction concerned. This is a proposal for a descriptive theory: it can explain almost all cross-border cases handled by most jurisdictions. To understand the eDP theory, it should be underscored that every competition law market consists of suppliers and purchasers. Most debates on market definition are on the side of defining relevant suppliers. In general, theories and disputes related to “market definition” have focused on definitions of the relevant suppliers, which are mostly based on “demand substitutability”. However, demand substitutability is based on the premise that the relevant purchasers have already been defined. The eDP theory concerns the definition of purchasers and their location. The location of suppliers is irrelevant here: all, some, or none of the suppliers may be domestically located. concentration will have an immediate and substantial effect in the Community.” “It is therefore necessary to verify whether the three criteria of immediate, substantial and foreseeable effect are satisfied in this case.”). 7 Examples include Nordion (Decision of 3 September 1998, Shinketsushu vol.45, p. 148), Microsoft (Decision of 16 September 2008, Shinketsushu vol.55, p.380), and Qualcomm (Order of 28 September 2009). For information in English about Microsoft, see Japan Fair Trade Commission, “Hearing Decision against Microsoft Corporation (Trading on Restrictive Terms Relating to Windows OEM Sales Agreements)” (18 September 2008), http://www.jftc.go.jp/e-page/pressreleases/2008/ September/080918.pdf. For information in English about Qualcomm, see Japan Fair Trade Commission, “Cease and Desist Order against QUALCOMM Incorporated” (30 September 2009), http://www.jftc. go.jp/e-page/pressreleases/2009/September/090930.pdf. With respect to Marine Hoses and CRT, see below notes 16–17 and 26–27 and accompanying text. For merger control examples, see Akira Goto, “Cross-Border Merger Control in Japan” (prepared remarks, OECD Global Forum on Competition on 17 February 2011), http://www.jftc.go.jp/e-page/policyupdates/speeches/Cross_border_merger_ control_in_Japan.pdf. The date and the forum can be checked using his slides, available at http://www. jftc.go.jp/e-page/policyupdates/speeches/Cross_border_merger_control_in_Japan(slide).pdf. 8 The JFTC tends to use this phrase to justify its international applications. See the 1992 announcement cited below at note 12, and accompanying text.

Tadashi Shiraishi  277   Figure 1: Basic Structure of the eDP Theory relevant market

P

Q

R

suppliers

A

B

C

purchasers Jurisdiction X

Figure 1: Basic Structure of the eDP Theory The illustrations below only show the tip of the iceberg of the eDP theory. There have been many ordinary cases where domestic purchasers were adversely affected, but these are too ordinary to avoid omission here.

2. Illustrative phenomena supporting the eDP theory a. Reactions to the Department of Justice’s 1992 announcement In 1988, the U.S. Department of Justice (DOJ) clarified its enforcement policy on cross-border cases by its famous footnote 159, which said that “the Department is concerned only with adverse effects on competition that would harm U.S. consumers by reducing output or raising prices”.9 This is a core example of the eDP theory.10 Superseding this footnote, the DOJ announced a change in 1992 which would permit the DOJ “to challenge foreign business conduct that harms American exports when the conduct would have violated U.S. antitrust laws if it occurred in the United States”.11 9 The U.S. Department of Justice, “Antitrust Enforcement Guidelines for International Operations” (November 1988), reprinted in: Antitrust & Trade Regulation Report (BNA) vol.55, no.1391. The full text of footnote 159 is as follows: “Although the FTAIA extends jurisdiction under the Sherman Act to conduct that has a direct, substantial, and reasonably foreseeable effect on the export trade or export commerce of a person engaged in such commerce in the United States, the Department is concerned only with adverse effects on competition that would harm U.S. consumers by reducing output or raising prices. See Section 3.0, supra.” 10 It seems that the term “consumers” in footnote 159 included corporate purchasers. Otherwise, it would have too narrowly confined the application of the U.S. antitrust law. 11 The Department of Justice, “Justice Department will Challenge Foreign Restraints on U.S. Exports under Antitrust Laws” (3 April 1992), http://www.justice.gov/atr/public/press_releases/1992/211137. pdf.

278  Merger Control in European and Global Perspective This meant that the jurisdiction in which the affected suppliers were located could apply its competition law even when the relevant purchasers were located in another jurisdiction altogether. The eDP doctrine cannot explain this announcement. Figure 2: U.S. DOJ Announcement in 1992

P

Q

R

A

B

Figure 2: U.S. DOJ Announcement in 1992 The week after the announcement was made, the JFTC reacted critically to the DOJ’s policy change.12 It claimed that conduct that affected “a market of our jurisdiction” should be remedied by the competition law of “our” jurisdiction. It mentioned the risk of “protecting not competition but competitors” by the competition authority of the jurisdiction where the allegedly excluded exporters were located. The JFTC did not define, and has never defined, the key concept of “a market of our jurisdiction”. However, the context of this dispute shows that it is a market with its purchasers located in the geographic area of the jurisdiction. As the history after 1992 shows, it seems that the DOJ has declined to actively enforce the changed policy.13 The eDP theory explains the absence of enough cases following the 1992 announcement by the DOJ.

b. Empagran In Empagran, a global cartel adversely affected purchasers all over the world, including those located in the United States. It was natural even from the perspective of the eDP theory that the cartel would be considered illegal in light of the U.S. antitrust law. 12 Kosei Torihiki Iinkai (Japan Fair Trade Commission), “Beikoku no Yushutsu wo Seigen suru Hankyosoteki Koi ni taisuru Beikoku Shihosho no Hantorasutoho no Shikko Hoshin no Henko ni tsuite (On the Change by the U.S. DOJ of the Antitrust Enforcement Guidelines for Anticompetitive Conduct Which Restrains U.S. Exports)” (9 April 1992), reprinted in: Kosei Torihiki no. 500, p. 53 (in Japanese). 13 For example, the 1995 international guidelines by the DOJ did not mention the policy announced in 1992. The U.S. Department of Justice and the Federal Trade Commission, “Antitrust Enforcement Guidelines for International Operations” (April 1995), http://www.justice.gov/atr/public/guidelines/ internat.htm.

Tadashi Shiraishi  279   Then, purchasers located outside the United States filed lawsuits at a federal court in the U.S. to obtain treble damage recovery. Figure 3: Empagran

P

Q

A

R

Empagran

Figure 3: Empagran The U.S. Supreme Court made clear that such claims would be denied if the adverse foreign effect was independent of any adverse domestic effect.14 Later, for the remanded proceedings, the U.S. Court of Appeals for the District of Columbia Circuit ruled against the foreign purchasers, saying that there should be “a direct causal relationship, that is, proximate causation” between the adverse domestic effect and the adverse foreign effect to justify an application of the U.S. antitrust law for the foreign purchasers.15 These judgments by the U.S. federal courts can be explained by the eDP theory.

c. International market sharing cases In extreme circumstances where only domestic suppliers were allowed to sell to domestic purchasers due to international market sharing arrangements, interesting phenomena can be observed from the perspective of the eDP theory. In Marine Hoses, the JFTC defined the relevant market as including only domestic purchasers.16 It followed that the JFTC did not impose administrative surcharges on European violators at all, but only on a Japanese supplier, Bridgestone.17 The Antimonopoly Act (AMA) calculates surcharges based on each sales amount within the relevant market. Zero multiplied by any percentage would still be zero. F. Hoffman-La Roche Ltd. v. Empagran S.A., 542 U.S. 155 (2004). Empagran S.A. v. F. Hoffman-La Roche Ltd., 417 F. 3d 1267, 1270–1271 (D.C. Cir. 2005), cert. denied, 546 U.S. 1092 (2006). 16 Orders of February 20, 2008, Shinketsushu vol.54, pp. 512 and 623. For information in English, see Japan Fair Trade Commission, “Cease and Desist Order and Surcharge Payment Order against Marine Hose Manufacturers” (22 February 2008), http://www.jftc.go.jp/e-page/pressreleases/2008/ February/080222.pdf. 17 Yokohama was granted full leniency immunity. See ibid. 14 15

280  Merger Control in European and Global Perspective Figure 4: Marine Hoses (JFTC)

Bridgestone Q

A

R

S

B

C

In contrast, the European Commission imposed a heavy fine on Bridgestone, which Figure 4: Marine Hoses (JFTC) did not have very many sales, if any, in the European Economic Area (EEA). 18 This does not necessarily mean that EU law is incompatible with the eDP theory. The Fining Guidelines of the Commission explain that, in such extreme circumstances, the value of sales (i.e., small or zero) may not properly reflect the contribution of the foreign violator to the infringement, and that “the Commission may assess the total value of the sales of goods or services to which the infringement relates in the relevant geographic area (wider than the EEA), may determine the share of the sales of each undertaking party to the infringement on that market and may apply this share to the aggregate sales within the EEA of the undertakings concerned”. 19 It may seem to be defining a worldwide geographic market that is “wider than the EEA”, but a logical reading of the paragraph above would show that the real target is the protection of purchasers located in the EEA. Worldwide presences just show the proportionate contribution by the violator to the adverse effects imposed on purchasers located in the EEA. It follows that EU law is still compatible with the eDP theory.

Commission Decision of 28 January 2009 in Case COMP/39.406 – Marine Hoses. Commission, Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003, 2006 OJ C210/02, para. 18. For the application in Marine Hoses, see Commission Decision, cited previous footnote, paras. 429–436. 18 19

Tadashi Shiraishi  281   Figure 5: Marine Hoses (European Commission)

Bridgestone Q

A

R

S

B

C

Figure 5: Marine HosesJapanese (European In other words, the difference between lawCommission) and EU law is not in whether it is compatible with the eDP theory, but rather in whether the amount of the fine (or in Japan, the surcharge) is based on real sales or “but-for” sales within the relevant market.

d. A famous hypothetical merger in Japan There have been disputes in Japan between the business community and the JFTC in relation to the following hypothetical merger: 1) The only domestic suppliers P and Q are planning to merge. 2) Domestic purchasers A and B prefer domestic suppliers, i.e., P and Q, over foreign ones as they wish to ensure the timely acquisition of input. 3) Foreign purchasers C, D, and E do not care about the timing of input acquisition. It is believed that a merger between P and Q is indispensable for them to survive competition from foreign suppliers R and S.

282  Merger Control in European and Global Perspective Figure 6: A Hypothetical Merger (1) “global market”

P

Q

A

B

R C

S D

E

“Japanese market” Figure 6: A Hypothetical Merger (1)

The Japanese business community has been lobbying the JFTC to define a “global market”, which would make it possible for P and Q to obtain legal clearance to merge. The JFTC opposes this idea, claiming that if Japanese purchasers choose only Japanese suppliers, the JFTC should define a “Japanese market” in order to protect these domestic purchasers. Under political pressure, the JFTC changed the language in its merger guidelines slightly in 2007. However, the JFTC defended its baseline, as shown above.20 This can be explained by the eDP theory in the sense that the purchasers in the “Japanese market” are located in Japan.21

e. Definition of a global market Even the JFTC sometimes defines a “global market” that includes foreign purchasers. This usually happens when each supplier sets its prices without regard to destination, so that purchasers in each country pay the same price. The chain of logic is as follows: 20 Japan Fair Trade Commission, “Guidelines to Application of the Antimonopoly Act Concerning Review of Business Combination” (as of 28 March 2007), Part II 3 (underscoring that the relevant purchasers should be “users in Japan”), http://www.jftc.go.jp/e-page/pressreleases/2007/March/ RevisedMergerGuidelines.pdf. For the proposed modification unveiled on 4 March 2011, see below note 24. 21 The compatibility between the eDP theory and Japanese merger control was reinforced by the 2009 amendment of Article 10(2) of the AMA, which adopted domestic sales instead of assets as notification threshold criteria. It requires both the acquiring group and the acquired company (including its subsidiaries) to exceed respective domestic sales thresholds. For the text of Article 10(2), see the tentative translation by the JFTC, http://www.jftc.go.jp/e-page/legislation/ama/amended_ama09.pdf.

Tadashi Shiraishi  283   1) The ultimate relevant market is still consistent with the eDP theory: The market only has domestic purchasers, no matter where the suppliers are located. 2) When each supplier sets unified pricing, whether the proposed merger enhances market power in the ultimate relevant market always corresponds with whether it does so in the global proxy market. 3) If the global proxy market will still be competitive post-merger, it follows that the ultimate relevant market will also remain competitive post-merger. Figure 7: Global Market Admitted even by the JFTC proxy market

P

Q

A

B

R

S

Each supplier sets unified pricing worldwide.

C

D

E

ultimate relevant market

Figure 7: Global Market Admitted even by the JFTC

This logic was clearly set out in the JFTC’s analysis of Sony/NEC, a decision adopted in 2006.22 Recently, the JFTC has tended to hide the logic above, stressing instead that it has defined a global market, probably to defend itself from the political accusation that it is too domestically oriented in its analyses.23 Examples include the JFTC’s proposal of a slight modification of its merger guidelines.24 The eDP theory, with the chain of logic above, provides a framework for consistently understanding the JFTC’s approach in relation to the hypothetical merger discussed (supra section e) and this global market definition. 22 Kosei Torihiki Iinkai (Japan Fair Trade Commission), “Heisei 17 Nendo ni okeru Shuyo na Kigyo Ketsugo Jirei” (Selected Merger Cases Reviewed during FY 2005), Case 8 (7 June 2006), http://www. jftc.go.jp/ma/jirei2/H17jirei8.html (in Japanese). 23 See above note 20 and accompanying text. 24 A proposal unveiled on 4 March 2011 to modify the merger guidelines (Japan Fair Trade Commission, “Guidelines to Application of The Antimonopoly Act Concerning Review of Business Combination”, http://www.jftc.go.jp/e-page/legislation/ama/RevisedMergerGuidelines.pdf). The proposed modification will just add an example of global market definition. The added sentence would be as follows: “For example, a global market can be defined if each of leading suppliers in the world sets unified pricing for purchasers located worldwide, and if purchasers can choose their main suppliers globally” (original text in Japanese).

284  Merger Control in European and Global Perspective

IV. Borderless Purchasers 1. A core premise of the eDP theory An important premise of the eDP theory is that the location of purchasers can be identified. This means that the eDP theory naturally has a controversial frontier, especially when purchasers move across borders and/or are engaged in borderless business.

2. CRT as a typical illustration The JFTC’s ruling in a case involving the price fixing of cathode ray tubes (CRTs) provides a good example for analysis.25 The price-fixing conduct took place across several South East Asian countries. Most of the cartel members were local subsidiaries (from Ps to Ss) of parent companies (from Pp to Sp) headquartered in Japan, South Korea, and Taiwan. More importantly from the perspective of the eDP theory, the purchasers (from As to Es) were also spread across South East Asia, with parent companies in Japan (from Ap to Ep). These purchasers procured CRTs and manufactured CRT TVs. Supplier parent companies and purchaser parent companies sometimes negotiated trade conditions and prices in Japan. Purchaser parent companies directed their subsidiaries to purchase at the prices promised in Japan, which had been secretly fixed among the suppliers. Figure 8: CRT

Pp Qp Rp Sp Ps Qs Rs

Ss

T

As Bs Cs Ds Es Ap Bp Cp Dp Ep Figure 8: CRT 25

For citations, see below notes 26–27.

Tadashi Shiraishi  285   The JFTC issued a cease and desist order and imposed an administrative surcharge in 2009,26 and the South Korean companies and their subsidiaries, which refused to receive notification documents, were deemed to have received the orders by public notification in 2010.27 The Taiwanese group was reportedly granted full leniency immunity.

3. Conclusion Along with CRT, we can consider a similar hypothetical merger. Suppose companies P and Q are both located outside jurisdiction X. They serve a limited number of purchasers, companies As and Bs, also located outside jurisdiction X. Companies Ap and Bp, located in jurisdiction X, are the parent companies of companies As and Bs, respectively. If companies P and Q are now planning to merge, can the competition authority of jurisdiction X apply its own merger control? Figure 9: A Hypothetical Merger (2)

Jurisdiction X

P

Q

As

Bs

Ap

Bp

Figure 9: A Hypothetical Merger (2) These cases may be limitations of the eDP theory, or one may wish to develop the theory to fit these difficulties. In any event, this frontier has been identified only after we described most incumbent phenomena using the eDP theory. Otherwise, this frontier would have been hidden by confused disputes. 26 Orders of 7 October 2009. For information in English, see Japan Fair Trade Commission, “Ceaseand-Desist Order and Surcharge Payment Orders against Manufacturers of Cathode Ray Tubes for Televisions” (7 October 2009), http://www.jftc.go.jp/e-page/pressreleases/2009/October/091007.pdf. 27 Orders of 7 October 2009, 5 February 2010 and 27 March 2010. For information in English, see Japan Fair Trade Commission, “Cease-and-Desist Order and Surcharge Payment Orders against Manufacturers of Cathode Ray Tubes for Televisions (follow-up report)” (29 March 2010), http:// www.jftc.go.jp/e-page/pressreleases/2010/March/100329a.pdf.

Etsuko Kameoka and Mel Marquis*

Recent Developments in Japan’s Merger Control System

Introduction In 2011, M&A activity in Japan increased by 28% compared to 2010, while the number of transactions outside Japan but involving Japanese companies more than doubled.1 This was motivated not only by a strong yen but also by the survival strategy of Japanese companies operating in the reality of a global business world. In general, these companies are concerned that protracted economic stagnation and a declining population will lead to a shrinking Japanese market. The economic outcomes of certain mergers are thus perceived to be increasingly important to the Japanese economy. In addition to such “endogenous” considerations, Japan has been, for over 20 years (and often notoriously), subject to pressures from the outside world.2 The most well-known source of external pressure, often linked to conflicts in the sphere of international trade, has been the United States. But the EC/EU has exerted some influence as well. In the context of competition policy and enforcement, and in addition to contacts between case handlers on specific cases, the EU and Japan have organized information exchange meetings since 1980 for the purpose of enhancing cooperation.3 In this forum the two jurisdictions have emphasized the importance of a harmonized framework to enforce competition policy efficiently, and they have expressed their commitment to setting a good example of cooperation and coordination between developed jurisdictions for the potential inspiration of countries (or regional groupings) with emerging competition law systems.4 * Etsuko Kameoka is Counsel at Van Bael & Bellis, Brussels. Mel Marquis is Part-time Professor of Law at the European University Institute in Florence and Professore a contratto at the University of Verona. We are grateful to Tadashi Shiraishi, who commented on an earlier, more abbreviated draft. 1 The Wall Street Journal Europe (30 December 2011 – 2 January 2012), p. 32. 2 Recently, for example, in November 2010 and February 2011, the US announced the US-Japan Economic Harmonization Initiative. In the context of the Initiative, the US called for increased efficiency and transparency of merger review in Japan. See http://www.mac.doc.gov/japan-korea/EHI/ EHI%20USG%20Agenda%20Items%202-11-11%20FINAL.pdf, In particular at p. 7. 3 The meeting is currently held annually. The latest meeting, as of this writing, took place on 14 July 2011. For the announcement of the meeting of 2011, see http://www.jftc.go.jp/pressrelease/11. july/110713.pdf (in Japanese). 4 Kazuhiko Takeshima, “Cooperation and Alliance between Japan and EU in competition policy”, Europe 257 (Spring 2009), p. 6; Neelie Kroes, “Competition Policy bringing consumers benefits”, Europe 257 (Spring 2009), p. 7; Delegation of the European Commission to Japan, “Press, Public and Cultural Affairs” (in Japanese).

288  Merger Control in European and Global Perspective A significant by-product of this cooperative framework and dialogue concerns a package of merger control reforms that seek to better accommodate the advancing process of globalization. These reforms entered into force on 1 July 2011 and the Japan Fair Trade Commission (JFTC) has begun to examine mergers and acquisitions in accordance with the revised system. The new elements took the form of an amendment to the JFTC’s notification rules and a revised set of merger guidelines.5 In this brief contribution we propose to comment on some key aspects of the reforms just mentioned. We will touch on the following developments in connection with the revised system: (i) the formal (but not total) abolition of the “prior consultation” system; (ii) enhanced transparency in merger clearance procedures; (iii) confirmation of the JFTC’s greater openness to accepting wider geographic markets, in particular worldwide markets; and (iv) early signs, drawn from three cases, of how the reforms seem to be influencing the JFTC’s merger practice. In a final section we add a few short concluding lines.

Goodbye to prior consultation? Prior to the 2011 merger control reforms, parties to a merger notifiable in Japan had the option of consulting with the JFTC before filing the transaction under the Prior Notification Guidelines.6 Such informal consultations presented advantages such as confidentiality pending discussions with case handlers. As a result of the reforms, this system has been abolished. From the JFTC’s perspective, the main benefit of prior consultation was the possibility to obtain information quickly and to head off problems at an early stage (thereby reducing the need to resort later to more drastic measures); this might also have seemed attractive to the parties to the extent that it led to quicker approval.7 In reality, according to the business community, prior consultation sometimes bogged down the procedure and allowed for a de facto circumvention of statutory timelines.8 5 For translations of the Rules on Applications for Approval, Reporting and Notification (“Notification Rules”), the Guidelines on the application of the Policies Concerning Procedures of Review of Business Combination (“Procedural Guidelines”), and the revised Guidelines on the Application of Antimonopoly Act Concerning Review of Business Combination (“Revised Guidelines”), see http://www.jftc.go.jp/en/policy_enforcement/mergers/pdf/110713.2.pdf; http:// www.jftc.go.jp/en/policy_enforcement/mergers/pdf/rules_on_applications.pdf; and http://www.jftc. go.jp/en/pressreleases/uploads/110620attach2.pdf. 6 See http://www.jftc.go.jp/ma/jizen-gl.pdf (in Japanese). 7 From the 1970s to the 1990s, the JFTC resolved competition problems informally, and merging parties were quite content to have matters handled on a confidential basis. See, e.g., Toshiaki Takigawa, “The Prospect of Antitrust Law and Policy in the Twenty-First Century: In Reference to the Japanese Antimonopoly Law and Japan Fair Trade Commission”, 1 Washington University Global Legal Studies Law Review 275, 288 (2002). However, following changes made in 2002, general satisfaction with prior consultation seems to have declined. 8 In theory, time should have been running according to “phase 1” and “phase 2” of the prior consultation phase (note: these periods bore no correspondence to the familiar Phase I and Phase II

Etsuko Kameoka and Mel Marquis  289   This probably explains why, in recent years, the prior consultation system appears to have been used only infrequently. In 2008, 1008 notifications were filed with the JFTC, yet prior consultation was used in just 28 of those cases.9 In 2009, prior consultation was used only 24 times as compared to 985 notifications. In 2010, when only 265 notifications were filed, prior consultation was used 13 times. The figures just recited (which do not take account of aborted deals that were never filed) imply that, at least with respect to garden-variety notifications, the impact of the reform may be limited. On the other hand, inasmuch as merging parties generally did avail themselves of prior consultation in large and complex cases, the statistics to some extent belie the qualitative importance of having an additional informal procedural tool.10 In the absence of prior consultation, the JFTC will have fewer opportunities to engage in informal, possibly opaque manoeuvring. Moreover, by bringing the procedure more squarely under the more formal framework, the JFTC may be more often obliged to introduce detailed economic analysis when presenting its concerns to the parties. Under the previous system, in cases where competitive concerns were discussed off the record, it was less imperative for the JFTC to build up the kind of sophisticated economic evidence that may be necessary to convince the parties and the public. Notwithstanding the abolition of prior consultation, if the parties so request they can still consult with the JFTC before the official notification even under the new system. The difference between the abolished prior consultation and the consultation under the new system basically lies in the fact that, under the previous framework, it was often possible for the JFTC to reach its final decision, including the imposition of any commitment requirements, by the time the consultation was concluded.11 In the EU, pre-notification contacts are considered to be important, and the European Commission actively encourages such contacts.12 However, in light of the fact that, as just described, voluntary prior consultations with the JFTC remain possible, the Japanese reform does not, in practical terms, seem to create a significant distinction between the Japanese and EU procedures. Indeed, our expectations are that, to the contrary, the revised system will contribute to greater convergence, in terms of procedures, because the system promotes a greater degree of procedural efficiency and transparency. in other jurisdictions). However, the JFTC was criticized for holding up the process, in particular by insisting that submitted documents were incomplete and imposing vast data requirements. See, e.g., Kaori Nakano, “The Recent Developments in Merger Investigation”, 320 Rippo to Chosa [Legislation and Investigation] 79, 81 (2011) (in Japanese). Source: JFTC, www.jftc.go.jp/kenkyukai/dk-kondan/110404/188-1.pdf (in Japanese). For the impact of the end of prior consultation, see, e.g., Masahiro Murakami, “Review of the Merger Procedure of 2011”, 1357 Hanrei Times 36, 44 (2011) (in Japanese). 11 See e.g., Tadashi Shiraishi, Antitrust Law Lecture, 5th edition, Yuhikaku Publishing House, 2010, p. 211 (in Japanese). 12 See Commission Regulation (EC) No 802/2004 of 7 April 2004 implementing Council Regulation 139/2004, 2004 L133/1, recital 11; DG Competition, Best Practices on the conduct of EC merger control proceedings (2004), http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf, p. 2. See also, e.g., Van Bael & Bellis: Competition Law of the European Community, 5th edition, Wolters Kluwer, 2010, p. 751. 9

10

290  Merger Control in European and Global Perspective In practice, we expect that in complex merger cases involving a wide range of competition concerns, many merging parties will still tend to enter into early consultations with the JFTC. Where the stakes are significant there may be a strong incentive to have early access to the regulator, and to have an early possibility of reaching a common understanding regarding the scope of the case and the nature of the information to be collected and disclosed. The marginal benefit of prior consultations may thus be decisive.

Confirmed recognition of worldwide markets by the JFTC The “old” Merger Guidelines (2007) already envisaged relevant geographic markets wider than the domestic market.13 Furthermore, the JFTC had in fact already begun to accept international markets, as it did, for example, in Sony/NEC (2005), a merger of optical disc drive businesses.14 In the EU, by comparison, although the parties to that merger argued for a worldwide market (given low transportation costs, the absence of trade barriers, important trade flows, global product standards and globally active suppliers and customers), the European Commission left the issue of the geographic market open because it would not have affected the Commission’s competitive assessment.15 Nevertheless, responding to business concerns and political pressure to quell lingering doubts, the revised Merger Guidelines add specificity as to when the JFTC will recognize a worldwide or regional (i.e., East Asian) market. According to Section 2.3(2) of the Guidelines, such instances include those where domestic or overseas suppliers are selling products in the worldwide (or East Asian) market for almost same price, and buyers are purchasing mainly from worldwide (or East Asian) sellers. This approach may permit greater compatibility between the JFTC’s investigations and international standards while facilitating merger parties’ discussions with case handlers as to how the geographic market should be defined.

13 See http://www.jftc.go.jp/en/pressreleases/uploads/2007-Mar-28_1_1.pdf. See also Tadashi Shiraishi, “Effects on Domestic Purchasers: A Descriptive Theory for Competition Law in CrossBorder Cases”, this volume, pp. 275 et seq. 14 See http://www.jftc.go.jp/ma/jirei2/H17jirei8.html (in Japanese). See also NEC Electronics/ Renesas Technology (2009), where the relevant market was found to be global in scope. See http://www. jftc.go.jp/ma/jirei2/H21jirei6.html (in Japanese). In the EU, the Commission mentioned in its decision that the geographic scope of the markets was at least EEA-wide, if not worldwide (for the semiconductor market, the relevant market could be left open, as the deal did not raise any competition concerns in any of the markets concerned). M.5535, Renesas Technology/NEC Electronics (2 December 2009). 15 M.4139, Sony/NEC/JV (31 March 2006).

Etsuko Kameoka and Mel Marquis  291  

Other procedural changes to enhance transparency The merger control reforms of 2011 were also designed to achieve more transparent procedures, a seemingly perpetual concern with regard to Japanese competition law procedures (and Japanese administrative law procedures more generally). Under the new framework, when the JFTC requests a report from the parties it should indicate the purpose of its request in order to ensure that the parties have a good understanding of what precisely is needed.16 At the parties’ request, the JFTC should also explain the nature of its competition concerns.17 This was already a matter of informal practice but it is now made explicit. Also under the new system, when the JFTC decides to initiate a second stage investigation, it will publish the announcement and ask the public for its views.18 Furthermore, an approval decision (which previously was not announced in written form) should now be provided to the parties in writing.19 Yet another helpful feature, found in the revised Guidelines,20 is a description of cases falling outside the JFTC’s jurisdiction. As concerns substantive analysis, the revised Guidelines also provide more detailed explanations of the elements that determine whether a merger will substantially restrain competition. These include, among other things, the supply posture of other producers, competitive pressure from neighboring markets, and countervailing pressure exerted by customers.21

Some practical applications of the JFTC’s revised framework 1) The NSC/Sumitomo Metal case The NSC/Sumitomo Metal merger was notified on 31 May 2011. The first stage of the investigation was completed on 30 June 2011 and, following an in-depth second stage investigation the merger was conditionally approved on 14 December 2011.22 This was the first notified transaction that the JFTC reviewed under the revised merger control framework. The merger between NSC and Sumitomo Metal presented a fairly typical large and complex case. If it had been cleared unconditionally it would have resulted in the world’s second largest steel producer (behind ArcelorMittal), and the parties’ business activities overlapped in approximately thirty areas. The JFTC’s investigation focused on Notification Rules, Article 8; Procedural Guidelines, Section 6(1). Procedural Guidelines, Section 4. 18 Ibid., Section 6(1), (2). 19 Notification Rules, Article 9; Procedural Guidelines, Sections 3, 5 and 6. 20 Revised Guidelines, Section 1.1 (1). 21 Ibid., Section 4 (2). 22 See http://www.jftc.go.jp/pressrelease/11.december/11121403.pdf (in Japanese). For an Englishlanguage summary, see http://www.jftc.go.jp/en/pressreleases/archives/individual-000457.html. 16 17

292  Merger Control in European and Global Perspective the six most sensitive markets. Although the deadlines were initially expected in February 2012, the JFTC processed the case with exceptional alacrity, granting its conditional approval two months early, on 14 December 2011. Under the law, the JFTC is expected to take a decision within 90 days from the date of submission of the required documents. Under the new procedure, when the parties submit all the required documents, the JFTC can reach its decision. As regards market definition in this case, the JFTC dismissed the proposition, insofar as “non-oriented electrical steel sheets” were concerned, that the merged entity would face competitive pressure from across East Asia. Distinct distribution patterns and a “completely different” distribution of market shares led the JFTC to define a market limited to Japan. With regard to “hot-rolled steel sheets” and “H-section steel”, the JFTC acknowledged the importance of inflows from South Korea and China but, in the absence of more specific data, Japan was again found to be the relevant geographic market. NSC/Sumitomo Metal is an example of a speedy procedure and a flexible examination under Japan’s new merger rules. If the parties of multi-jurisdictional mergers manage to obtain a clearance decision quickly from the JFTC, they can focus their energies on other jurisdictions,23 although care should be taken so as not to raise doubts on the part of other agencies about perceived attempts to “game” diverse timetables. As explained earlier, the parties in NSC/Sumitomo Metal were also entitled to written notice from the JFTC when it terminated the investigation.24 This more formal document will be useful not only to make the outcome of the investigation clear to the parties but will also serve, if necessary, as authentic proof of the JFTC’s investigation and, more generally, it will facilitate public awareness about the decision. Overall, the first investigation under the new system seems successful from the perspectives of international harmonization, transparency and prompt procedures. Furthermore, under the new rules, the JFTC can achieve a decision which fits the current needs of the Japanese industry more easily. The restructuring of mature industries is essential as a way to galvanize the stagnant Japanese economy. The procedure under the new rules can be expected to facilitate the process of securing the JFTC’s approval to create large competitive firms.25 The tension between industrial policy and competition policy, in sectors such as notably the steel and paper industries, has been a recurring theme for many decades in Japan. Finding the right balance and means of coordination between these policies has not been easy, to put it mildly. The latest reform represents another attempt – not the last, certainly – to achieve a workable equilibrium in this regard. 23 In this case, the parties notified their transaction in more than ten jurisdictions, including EU Member States such as Germany. In a press release of 19 March 2012, NSC indicated that the parties had secured approvals from the relevant jurisdictions. See http://www.nsc.co.jp/CGI/news/whatsnew_ detail.cgi?section=11&seq=00021166. 24 In a press release of 14 December 2011, NSC indicated that the parties had received written notice from the JFTC to the effect that it would not issue a cease and desist order. See http://www.nsc. co.jp/CGI/news/whatsnew_detail.cgi?section=0&seq=00021108 (in Japanese). 25 See, e.g., Masahiro Murakami, “JFTC Considered Companies’ Competitiveness: Merger Investigation Reached International Standard”, Economist (31 January 2012), p. 44 (in Japanese).

Etsuko Kameoka and Mel Marquis  293   The NSC/Sumitomo Metal investigation was also notable because it was the first case in which another revised piece of legislation was applied. In May of 2011, Japan also modified its “Law on Special Measures for Industrial Revitalization”, and the amendments entered into force in July 2011.26 Under this law, the ministry in charge of the industrial area concerned is entitled to provide comments to the JFTC regarding a pending investigation. The Japanese business community expects or hopes that where such consultation occurs, arguments to the effect that a merger will strengthen the position of the company in global markets will be more easily accepted.27 Thus, like the merger reform, this provision is intended to accelerate merger investigations whose outcome will affect important Japanese markets and may enhance the competitiveness of Japanese industry in global markets. Under this procedure, for example, the ministry concerned can explain the need for the creation of a larger company to face global competition in the market concerned, especially if it is expected that the JFTC will hesitate to approve the deal on account of competition problems. In NSC/Sumitomo Metal, the relevant ministry (as it often will be) was the Ministry of Economy, Trade and Industry (METI). In the ongoing Tokyo/Osaka Stock Exchange merger case (which, if approved by the JFTC, would create Japan’s largest exchange group, and the world’s third largest), the parties have requested that Japan’s Financial Services Agency intervene.28 We would note that, in the EU, it is not unusual for a Directorate General of the Commission other than DG Competition to become involved in the investigation, sometimes as a result of the parties’ lobbying activities, even if in the end the decisive criteria are quasi-exclusively competition criteria. Of course, the Commission’s merger decision is adopted not by DG Competition but by the College of the Commissioners. Although the institutional frameworks differ (dramatically) in the EU and in Japan, a parallel nevertheless arguably emerges here, with “triangular” dialogues in each jurisdiction among the competition authority, the parties to a merger, and relevant organs of the executive.

2) The Western Digital/Viviti Technologies (/Hitachi Global Technology) case and the Seagate Technology International/Samsung case The second and the third cases after the new system was introduced concerned two transactions in the hard disc drive (HDD) market.29 In these cases, the largest 26 Article 13 of Law on Special Measures for Industrial Revitalization. See http://www.meti.go.jp/ sankatsuhou/outline/data/etc-2/sankatsuhou.pdf (in Japanese). 27 See, e.g., Information on Minutes of Meeting of House of Representatives, No.177 Parliament, Committee on Economy and Industry, No. 7, See http://kokkai.ndl.go.jp/SENTAKU/syugiin/177/00 98/17704270098007c.html (in Japanese). However, a consultation does not necessarily mean that the JFTC and the ministry in charge will discuss the existence of illegal conduct during the investigation of a pending case. According to the Secretary-General of the JFTC, the JFTC’s views on breaches of AMA are submitted to the ministry in charge only after completing the investigation. See http://www. jftc.go.jp/teirei/h23/kaikenkiroku110518.html (in Japanese). 28 See, e.g., The Asahi Shinbun (4 January 2012). http://www.asahi.com/business/jiji/ JJT201201040060.html (in Japanese). 29 See http://www.jftc.go.jp/pressrelease/11.december/111228.pdf (in Japanese). For an Englishlanguage summary, see http://www.jftc.go.jp/en/pressreleases/archives/individual-000460.html.

294  Merger Control in European and Global Perspective company in the 3.5 inch HDD market planned to acquire the number 3 company in the market, while the second largest company sought to acquire the fourth company in the same market. The combined implication of these cases was thus a reduction of the number of major firms in the market from four to two. Western Digital Ireland (WDI) notified its acquisition of shares in Hitachi Global Technology (renamed Viviti Technologies) on 10 June 2011. A secondstage investigation was initiated on 4 July 2011, and the authority received all necessary information from the parties on 26 August 2011. WDI then offered remedies to remove the possible competition concerns of the JFTC, which conditionally approved the transaction, under Article 10 of the Antimonopoly Act (AMA) regarding share acquisition, on 24 November 2011. Meanwhile, on 19 May 2011, Seagate Technology International notified to the JFTC its planned acquisition of Samsung’s HDD business. The JFTC started its second-stage investigation on 17 June 2011 and received all the required documents from the parties on 27 October 2011. According to the JFTC, in light of the remedies proposed by WDI in the parallel case described above, this transaction would not substantially restrict the competition in the HDD market. The JFTC therefore approved the transaction without remedies, under Article 16 AMA (business transfers), on 15 December 2011. With regard to the geographic market definition, the JFTC concluded that all five HDD markets in issue were global markets. These cases were also investigated in the EU in 2011, and the European Commission approved both transactions. Under the EU Merger Regulation, the Commission started an in-depth investigation on 30 May 2011, inviting interested third parties to submit their observations.30 In its analysis the Commission adopted the same geographic market definition as the JFTC, finding worldwide markets for HDDs based on their form factor (3.5-inch or 2.5-inch) and their end use (in desktop computers, mobile computers, consumer electronics devices and enterprise business critical and mission critical applications).31  However, the issue of “priority among notifications”, which became an important issue in the EU, was not discussed in the procedure before the JFTC. In the EU, the fact that the notifications were filed only one day apart may have led to unfair outcomes.32 While the Seagate/Samsung merger was notified to the 2001 OJ C165/3. M.6203, Western Digital Ireland/Viviti Technologies (23 November 2011) and M.6214, Seagate Technology/The HDD business of Samsung Electronics (19 November 2011). See also Commission Press Release IP/11/1395 of 23 November 2011. 32 When two transactions are notified to the European Commission in close proximity, under the so-called “priority rule” the first notification will be reviewed as if the other transaction did not exist, whereas the second notification will be reviewed as if the first deal had already been implemented. In the Seagate/Samsung case, the Commission applied a “priority rule” whereby Seagate’s proposed acquisition was examined on the basis of the actual market structure absent the Western Digital/ Hitachi merger, while the Western Digital/Hitachi deal was examined taking into account the impact of the Seagate/Samsung notification. Thus, a one-day difference in notification dates may create an unexpectedly heavy burden on the merging parties of the second notification in the same sector. This somewhat awkward procedure has been criticized in the EU as being inequitable. By contrast, in Japan the JFTC reviewed both transactions cumulatively, adopting a more holistic analysis. 30 31

Etsuko Kameoka and Mel Marquis  295   Commission on 19 April 2011, the WDI transaction was notified on 20 April 2011. After both notifications were assessed separately based on the priority rule, the former was approved without commitments, while the latter was cleared subject to commitments more than a month later.

Conclusions It is no secret that for about 20 years, and putting natural disasters aside, the fortunes of the Japanese economy have ranged from bleak to dismal, with patches of dreary and forlorn. Adding to Japan’s woes, unfortunately, is an outlook of sputtering global demand – not much help to an economy traditionally dependent on the country’s export activities. A more streamlined and transparent merger clearance procedure cannot singlehandedly reverse those fortunes, but it does have the potential to contribute to Japan’s gradual economic recovery. Ideally, efficient merger control can facilitate a restructuring and rejuvenation of those Japanese industries whose competitive advantages have subsided or been displaced. However, a good deal of caution is warranted, as a host of different conditions must coalesce if Japan is to have any hope of a robust and sustained turnaround.33 As a tentative conclusion we would add that the 2011 reforms we have discussed may also be understood as a turn away from Koizumi-style independent competition law enforcement.34 What we may be witnessing is an effort to mediate between the need for effective competition policy (which seems to be the global trend) and other impulses well-known in Japan, such as an active industrial policy. It is worth noting that the current merger policy has been harmonized with METI’s industrial policy agenda, and it arose out of the spirit of the “New Growth Strategy” formulated under the Kan administration.35 On the other hand, the Japanese system of competition law more generally serves the Japanese economy not by means of the overbearing bureaucratic approach of the past, but rather by a rule-based approach grounded in law. In this respect, and taking account of the reforms we have outlined in this chapter, the general line of development 33 According to Noland and Pack, although countries that have experienced slower economic growth than expected despite relatively good macroeconomic policies may be tempted to pursue an active industrial policy, in most cases this strategy has turned out to be unsuccessful. See Marcus Noland and Howard Pack, Industrial Policy in an Era of Globalization: Lessons from Asia, Institute for International Economics, 2003, chapter 5, at p. 100. 34 In May of 2001, Prime Minister Koizumi announced to the Japanese parliament that he would reinforce the independence and stature of the JFTC. See, e.g., Harry First and Tadashi Shiraishi, “Concentrated Power: The Paradox of Antitrust in Japan”, in Daniel Foote, ed., Law in Japan. A Turning Point, University of Washington Press, 2007, chapter 21, at p. 522. See also, e.g., Sadaaki Suwazono, “Explanation of reform of Antimonopoly Act”, 1733 Shojihomu 4, 4 (2005) (in Japanese). 35 The New Growth Strategy was adopted by the cabinet meeting under the Kan administration on 18 June 2010. The objective is to create a strong Japanese economy by, among other things, harmonizing the merger procedure with the global standard in terms of rapidity, transparency and predictability in merger investigations. See JFTC’s press release of 14 June 2011, http://www.jftc. go.jp/pressrelease/11.june/110614kiketsu.pdf (in Japanese).

296  Merger Control in European and Global Perspective in Japanese merger procedure appears to be a path of convergence towards current global standards. More specifically (since “global standards” should not be assumed to be monolithic), and following in the direction of the last reform in 2009, the further refinements made in 2011 bring the JFTC’s merger control system generally in line with the EU model. The JFTC’s deepening impulse to align itself on global convergence patterns is not surprising given, for example, its significant participation in the ICN’s Merger Working Group, where it recently promoted, and secured the ICN’s approval for, a voluntary Framework for Merger Review Cooperation.36

36 The basic thrust of the Framework is to promote a growing collaborative network of liaisons, in the context of multijurisdictional merger review, so as to facilitate information exchange among agencies. The Framework, meant to be a complement and supplement to existing bilateral cooperation agreements, was approved at the ICN’s Annual Conference in Rio de Janeiro (20 April 2012). See http://www.jftc.go.jp/en/policy_enforcement/speeches/pdf/1Framework%20for%20Merger%20 Review%20Cooperation.pdf. It is perhaps in the nature of things that the modest ambitions embodied in the Framework are seen as “not an end, but only the beginning”. Kazuhiko Takeshima, “Toward more effective and efficient cooperation in Merger Review”, speech, Rio de Janeiro, 19 April 2012, http://www.jftc.go.jp/en/policy_enforcement/speeches/pdf/2MWG_Plenary_Script.pdf, at p. 6. For a summary of the approved Framework, see also http://www.jftc.go.jp/pressrelease/12. april/120425betten1.pdf.

Seonghoon Jeon *

International Convergence and Recent Korean Experiences in Merger Control

1. Introduction There are two aspects of international convergence in merger control: emphasis on economic analysis, and international coordination in multi-jurisdictional cases. Firstly, most competition authorities in developed countries exercise more rational discretion based on economic analyses, employing more economic tools in defining relevant markets and assessing adverse competitive effects, such as critical loss, diversion ratio, merger simulation, etc. Secondly, they exert more coordination efforts in multi-jurisdictional merger cases, which results in more consistent conclusions across nations. In this short paper I will explain the key features of the merger control system in Korea, and then I’ll introduce three recent merger cases that might shed some light on the two aspects of international convergence in merger control mentioned above.

2. Merger control in Korea A special feature of the Korean merger control system is its two-tier structure. At a notification level, the guidelines for M&A review specify the conditions for when a reported merger belongs to the safe harbour. The safe harbour conditions are similar to those in the EU and other jurisdictions, even though the threshold levels of HHI and changes in HHI might be slightly different: (i) HHI < 1,200, (ii) 1,200 ≤ HHI < 2,500 and ∆HHI < 250, or (iii) HHI ≥ 2,500 and ∆HHI < 150.1 But a distinctive feature of the Korean system is that at the law level, the socalled Monopoly Regulation and Fair Trade Act specifies the conditions for when a merger should be presumed to be anticompetitive. The conditions are basically as follows: firstly, the combined market share is greater than 50%, or it is the largest share in the market and the CR-3 is greater than 75%, and secondly, the * Professor of Economics at Sogang University and Non-standing Commissioner of Korea Fair Trade Commission. This work was supported by the National Research Foundation of Korea Grant funded by the Korean Government (NRF-2010-330-B00091). 1 See KFTC notification 2009-39, “Guideline for M&A Review”, 20 August 2009.

298  Merger Control in European and Global Perspective difference between the combined entity’s share and the next largest firm’s share is greater than 25%.2 Even though the presumption is rebuttable in principle, it is not so easy to rebut it in practice since the conditions are specified at the law level. However, as economic analysis is gaining more importance, KFTC (Korea Fair Trade Commission) has recently allowed some mergers which were presumed to be anticompetitive according to the law under the prescription of behavioural remedies. The most prominent example is eBay-G market merger in 2009, which I will introduce shortly. Other features such as market definition and considerations in assessing anticompetitive effects in Korea are similar to those in the EU and other advanced countries. One minor caveat is that, in Korea, the relevant market is defined from the representative consumer’s perspective as the range where he/she can switch his/ her purchases in response to a SSNIP, while in the EU and in the US, it is defined from the producer’s perspective as the smallest range where he/she can make profits by adopting a SSNIP.3 However, there are no practical differences between the two approaches in that KFTC adopts various methodological tools, such as critical loss analysis, based on the hypothetical monopolist version of the test. Considerations in assessing anticompetitive effects such as product differentiation, potential competition and entry barriers, and defenses based on efficiency and failing firms, are similar to those in the EU and in the US.

3. Recent merger cases in Korea In this section I will introduce three recent merger cases which shed some light on the two aspects of international convergence that I mentioned before: emphasis on economic analysis, and international coordination in multi-jurisdictional cases.4 2 More specifically, Korea Monopoly Regulation and Fair Trade Act presumes “a combination of enterprises” as “practically suppressing competition in any particular business area” if all of the following conditions in Article 7.4 are met: (a) The combined company is in a dominant position in the relevant market; i.e., its market share is 50% or more, or CR-3 is 75% or more unless the combined market share is less than 10%; (b) The combined market share is the largest in the relevant market; and (c) The difference between the combined market share and the next largest market share is 25% or more. 3 The representative consumer version may imply a larger market than the hypothetical monopolist version. The reason is as follows: the representative consumer version considers only the substitution effect, and includes products or regions which are close substitutes in the relevant market. On the other hand, the hypothetical monopolist version considers the price effect, i.e., both substitution and income effects. Even if all close substitutes are included, a hypothetical monopolist may not be able to increase price profitably because of the negative income effect in the case of normal goods. Hence, the range of the relevant market should be enlarged further under the hypothetical monopolist version. The consumer version that includes close substitutes may be fine purely in terms of economics. But from an antimonopoly perspective, the market for which there are concerns that a monopolist can increase price seems to be more appropriate. 4 I was involved in the first two cases as an expert economist for defendants, and in the third case as a commissioner of the KFTC.

Seonghoon Jeon 

299  

3.1. TDK-Alps (2007) The first case is a merger between TDK and Alps in 2007. Both firms are based in Japan. But the merger would have had competitive effects on a Korean buyer, Samsung Electronics, and hence on Korean consumers. Therefore, KFTC reviewed the possible anticompetitive effects of the merger. Here the main economic issue was upward transmission of competitive pressures from the downstream market. As shown in Graph 1, TDK and Alps are producers of Magnetic Head for Hard Disk Drive in personal computers. They sell magnetic heads to hard disk drive makers, such as Samsung, Toshiba, and Fujitsu. On the other hand, firms such as Seagate, WD (Western Digital) and HGST (Hitachi Global Storage Technologies) are Hard Disk Drive makers producing magnetic heads in house. Graph 1: Vertical Relationships in the TDK-Alps case

If the relevant rule is applied simply, the merger is presumed to restrain competition according to Korean law since the combination results in a market share of more than 50% in some market segments on magnetic heads, even if the internal transactions of the integrated firms are included in the market sales volume. However, the KFTC allowed the merger without any restriction. My speculation is that the decision was based on the economic reasoning that competitive pressures from the downstream HHD market were sufficient to address the concerns raised by the proposed merger. This case hints at a close international coordination between the KFTC and the JFTC (Japan Fair Trade Commission). The JFTC approved the deal in November 2007 without finding anticompetitive effects, and the following month the KFTC likewise concluded that the merger would not have anticompetitive effects.

300  Merger Control in European and Global Perspective

3.2. eBay-G market (2009) The second case is the eBay-G market merger of 2009. The key issue here was dynamic competition in an Internet-related industry. The market environments were as follows. eBay, which owned Auction, tried to combine with G market through a stock purchase agreement. G market and Auction are two leading “open market” operators in Korea. Open market operators provide an online marketplace where sellers directly sell their products to consumers, and play a minimum platform role by providing site management, transaction settlements, etc. On the other hand, general on-line shopping malls buy products from suppliers, keeping them in stock, and sell to consumers through Internet sites at their own risk. Graph 2 shows the two-sided open market structure. Graph 2: Relevant Markets in the eBay-G market case

The KFTC defined the consumer-side relevant market broadly as the Internet shopping market, since consumers are very price-sensitive and can easily search low price sellers and buy from them. The combined market share of Auction and G market in the Internet shopping market in 2008 was 37% in terms of transaction volume. The KFTC concluded that there were no anticompetitive concerns in the consumer-side market. However, the KFTC defined the seller-side relevant market narrowly as the open market, since open markets and general on-line shopping malls are distinguished as explained before. The combined market share of Auction and G market in the open market in 2008 was 87.5% in terms of transaction volume. The KFTC concluded that there might be some competition concerns in the seller’s side of the market, at least in the short-run.

Seonghoon Jeon 

301  

However, reasoning that the prospective dynamic competition from new entry and repositioning in adjacent markets would relieve the concerns in the intermediate run or in the long run, the KFTC allowed the merger with some temporary behavioural remedies in the seller-side market.5 This case is significant in that, on the basis of economic analysis, the KFTC allowed a merger which was to be presumed anticompetitive according to Korean law.6

3.3. BHP Billiton-Rio Tinto (2010) The last case to be mentioned here was a proposed common-production joint venture of two Australian iron ore producers, BHP Billiton and Rio Tinto in 2010. Table 1 shows the distribution of market shares of iron ore producers in the worldwide seaborne market. Table 1: Market Shares of Iron Ore Producers in Worldwide Seaborne Market in 2008 Fine Firm BHPB Rio

BHPB + Rio

Lump Share

15.8%

Firm BHPB

38.1% 22.3%

Rio

BHPB + Rio

Pellet Share

23.0%

Firm BHPB

55.2% 32.2%

Rio

BHPB + Rio

Share 6.4% 12.8% 6.4%

Vale

30.2%

Vale

10.4%

Vale

46.9%

FMG

2.5%

Kumba

9.4%

LKAB

9.9%

Others

29.2%

Others

25.0%

Others

30.4%

Anticompetitiveness could be presumed for the lump iron ore market according to the Korean law, since the combined market share was greater than 50%, and since the difference with the second largest share was more than 40%. Anticompetitive concerns might be raised for the fine market as well. The KFTC found adverse competitive effects in the proposed deal. First of all, the common-production joint venture would have led to the same anticompetitive 5 See KFTC Decision 2009-145, 25 June 2009. The behavioural remedies included restrictions on the increase of various service fees which the combination might levy on sellers in the open market. 6 In the past, presumptively anticompetitive mergers were allowed only very rarely. Indeed, such transactions were cleared only in the period of economic crises. Examples include the mergers of Hyundai-Kia motors and SKT-Shinseghi mobile telecommunications in 1999, where the justifications were practical considerations of industry restructuring, rather than economic reasoning.

302  Merger Control in European and Global Perspective effects as a full merger since it would be difficult to separate it from the marketing departments of the two companies. In terms of unilateral effects, the joint venture might have reduced production in the short-run, or capacity investment in the long-run, resulting in price increases. The KFTC’s economic analyses, such as critical slope and merger simulation, confirmed these concerns. In terms of coordinated effects, only two major producers would have remained after the joint venture, which would have made collusion easier.7 In this case, international coordination played a very important role. The KFTC exchanged information and coordinated comprehensively on possible remedies with other enforcement agencies, especially with the JFTC. Just after the Examination Report was submitted to the full committee of the KFTC on 1 October 2010, the two companies announced their agreement to drop the joint venture plan on 18 October. It was therefore unnecessary for the full committee to take further action.

4. Concluding remarks I will conclude this brief and descriptive paper by remarking that recent Korean experiences confirm two aspects of international convergence in merger control: emphasis on economic analysis and the importance of international coordination in multi-jurisdictional cases. Recently, the KFTC has been exercising more rational discretion in evaluating merger cases on the basis of economic analysis, rather than applying the simple conditions for presuming anticompetitiveness specified in the law. The cases of TDK-Alps and eBay-G market confirm this trend. Moreover, in regulating multijurisdictional merger cases, the KFTC exchanges information and coordinates comprehensively with respect to possible remedies with the enforcement agencies of other jurisdictions. Such efforts lead to internationally consistent conclusions. The cases of TDK-Alps and BHP Billiton-Rio Tinto confirm the KFTC’s further efforts to promote international coordination in multi-jurisdictional cases.

7 See KFTC Examination Report on the BHP Billiton-Rio Tinto case, which was submitted to the full committee of the KFTC on 1 October 2010.

Xinzhu Zhang* and Vanessa Yanhua Zhang**

China’s Merger Control Policy: A Three-Year Milestone

I. Introduction After thirteen years of incubation, the Chinese government finally in 2007 adopted the Anti-Monopoly Law (“AML”), with China’s economy still in transition from a centrally-planned economy to a market economy. Sustained discussion of China’s experience in antitrust is important in at least two respects. On the one hand, as a country with a newly established antitrust regime, its experience will certainly contribute to the understanding of antitrust issues among antitrust agencies in other jurisdictions all over the world. On the other hand, and perhaps more importantly, given that China has achieved remarkable economic successes in the absence of a comprehensive competition law, the Chinese experience will contribute to both policy and academic debates clarifying whether antitrust has any role to play in promoting economic development.1 Similar to the competition laws in other jurisdictions, the AML establishes three bodies of substantive legal rules against “monopolistic” or collusive agreements, the abuse of dominance, and anticompetitive mergers and acquisitions. However, while it has been three years since the AML came into effect on 1 August 2008, its enforcement has developed slowly and unevenly. Merger control is probably the most advanced enforcement area. Several hundreds of cases have been completed by the Anti-Monopoly Bureau of the Ministry of Commerce (MOFCOM), China’s merger review agency, and several high-profile cases have been decided. As the body of merger cases has grown significantly, it’s a good time to take stock and revisit the question: where are we going after the three-year milestone? In a previous paper published in 2010,2 we discussed the patterns of China’s merger control policy and tried to provide some implications as to what we could * Jiangxi University of Finance and Economics, Nanchang 310033, China; Research Center for Regulation and Competition, Chinese Academy of Social Sciences, Beijing 100732, China. ** Renmin University of China, Beijing, 100872, China; Global Economics Group, Beijing and New York. The views expressed in this paper are exclusively those of the authors and do not necessarily reflect the views of their affiliated institutes. The authors would like to thank Russell Pittman, Ping Lin and all of the participants at the International Conference on Regulation and Competition Policy: New Developments and Empirical Evidence at Jiangxi University of Finance and Economics, China (3–4 December 2011) for their helpful comments. All errors are of course our own. 1 See Thomas Ulen, “The Uneasy Case for Competition Law and Regulation as Decisive Factors in Development: Some Lessons from China”, in Michael Faure and Xinzhu Zhang, eds., Competition Policy and Regulation: Recent Developments in China, the US and Europe, Edward Elgar Publishing, 2011. 2 See Xinzhu Zhang and Vanessa Yanhua Zhang, “Chinese Merger Control: Patterns and Implications,” 6 Journal of Competition Law and Economics 477 (2010).

304  Merger Control in European and Global Perspective look forward to. In the meantime there have been new developments in Chinese merger control policy. To begin with, many new transactions have been reviewed. According to MOFCOM,3 from 1 August 2008 to the end of December 2011, it has completed merger reviews in 382 cases, with 17 cases in 2008, 80 in 2009, 117 in 2010 and 168 in 2011. Among the 382 completed cases, 371 were cleared by MOFCOM without any condition; this accounts for 97 percent of the total closed cases. The other 3 percent consists of ten mergers approved subject to conditions and one prohibition. Analyzing those case decisions, we find that although the rules and provisions of China’s merger control policy is largely consistent with those of other jurisdictions such as the US and the EU, MOFCOM takes its own approach to reviewing mergers and making decisions. The rest of this chapter is organized as follows: Section 2 analyzes the major merger review provisions. Section 3 provides a brief discussion and analysis of the eleven decisions described above where mergers were either approved conditionally or blocked. 4 In Section 4 we try to shed some light on the patterns and implications of China’s merger control policy, and Section 5 concludes.

II. Recent provisions relating to merger review in China The Market Definition Guidelines In terms of regulations and rules adopted in order to implement the AML, the Anti-Monopoly Commission of the Chinese State Council issued the first antitrust guidelines on 7 July 2009, i.e. The Guidelines on the Definition of Relevant Market (“the Market Definition Guidelines”).5 The Guidelines incorporate and integrate some concepts and principles from the horizontal and non-horizontal merger guidelines issued in other jurisdictions, such as the US and the EU. As usual, they also contain certain Chinese characteristics. Compared to the US Horizontal Merger Guidelines which were released by the US Department of Justice and 3 See Xiaosong Zhang, “Antitrust Cases Grow in China since the Anti-Monopoly Law Took Effect,” Xinhua News, 21 September 2011, http://www.gov.cn/jrzg/2011-09/21/content_1953353. htm. See also MOFCOM Press Release, “MOFCOM Held a Press Conference on 2011 Review of Antitrust Enforcement,” 31 December 2011, http://english.mofcom.gov.cn/aarticle/subject/ cv/lanmua/201112/20111207909371.html. MOFCOM Press Release, “MOFCOM Held a Press Conference on the Interim Measures for Investigating and Publishing of Concentration of Undertakings Failing to File Notification,” 10 January 2012, http://www.gov.cn/jrzg/2012-01/10/ content_2041384.htm. 4 For a more detailed discussion of cases during the period of August 2008 and August 2009, see Zhang and Zhang, “Chinese Merger Control”, cited above note 2. For discussion of cases between August 2009 and December 2011, see Xinzhu Zhang and Vanessa Yanhua Zhang, “Revisiting China’s Merger Control: Where Are We Going after Three-Year Milestone?” CASS Working Paper, 2012. 5 See the State Council of People’s Republic of China Press Release, The Guidelines on the Definition of Relevant Market, July 7, 2009, http://www.gov.cn/zwhd/2009-07/07/content_1355288. htm.

Xinzhu Zhang and Vanessa Yanhua Zhang 

305  

Federal Trade Commission on 19 August 2010 (“the 2010 US Horizontal Merger Guidelines”)6 and the EU Horizontal Merger Guidelines,7 there is some similarity in China’s Guidelines. The guidelines were issued by the AntiMonopoly Commission of the State Council rather than any of the three antitrust enforcement agencies. This implies that these guidelines are applicable not only to merger review but also to the investigation of “monopolistic” agreements and the abuse of dominance. This treatment is similar to the approach in both the US and the EU and has the advantage of coherence in dealing with different types of cases. However, the Market Definition Guidelines fail to mention the caveats when applying market definition methods to different types of cases such as for example the elevated benchmark price that was caused by pre-merger market power. Gaps thus remain with regard to certain technical but important issues, such as the need for proper techniques to avoid the “Cellophane fallacy”.

The Interim Rules Two years after the Market Definition Guidelines were announced and one year after the US released the 2010 Merger Guidelines, MOFCOM published the Interim Rules on the Assessment of Competitive Impacts of Concentrations of Undertakings (“the Interim Rules”), which took effect on 5 September 2011.8 The Interim Rules provide the road map as to how MOFCOM will assess mergers and other concentrations under the AML. The Interim Rules bear several notable features. First, the Market Definition Guidelines and the Interim Rules have been treated with different legal approaches. As seen above, the former was issued by the Anti-Monopoly Commission as guidelines, while the latter was enacted by MOFCOM as a set of regulatory rules. Second, Article 4 of the Interim Rules states that two types of competition harms for horizontal mergers will be considered: unilateral effects and the coordinated effects. Such an approach is similar to those in the US and the EU. In addition, Article 4 states that competition harm for non-horizontal mergers will also be taken into account. From this perspective, one may conclude that China’s approach to assess the competition effects of non-horizontal mergers is closer to the EU than to the US, where the competition effects of non-horizontal mergers are clearly downplayed. However, there is one major difference between the Chinese and the European approaches: in the EU, a higher standard of proof for non-horizontal mergers is explicitly required, which is consistent with modern economic theory. In China, however, the issue of the standard of proof for objecting to non-horizontal mergers is not 6 The revised 2010 US Horizontal Merger Guidelines are available at http://www.justice.gov/atr/ public/guidelines/hmg-2010.html. 7 See Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, 2004 OJ C31/5. 8 See MOFCOM Notice No. 55 [2011] Interim Rules on Assessment of Competitive Impacts of Concentrations of Undertakings, 29 August 2011, http://fldj.mofcom.gov.cn/aarticle/ zcfb/201109/20110907753173.html.

306  Merger Control in European and Global Perspective mentioned explicitly. This leaves open the possibility that more non-horizontal mergers might be blocked or approved with remedies if a lower standard of proof applies.

III. Preliminary case analysis The merger decisions published by MOFCOM have concerned horizontal and non-horizontal mergers. Although horizontal and non-horizontal mergers seem to be treated similarly in the Interim Rules,9 economic theory and the practice in other jurisdictions suggest that the competition problems that can potentially arise in these two types of merger scenarios differ, and the standards of proof should be different. We therefore discuss these types of mergers separately.

a. Horizontal mergers Under the AML, the legal principle for merger review is to evaluate whether a merger may restrict or exclude competition, which is consistent with international practice.10 However, since many important implementing rules in the Market Definition Guidelines and the Interim Rules are simply too vague to be understood with precision, their enforcement is bound to be marked by certain “Chinese characteristics”.

(1) Market definition In most cases, MOFCOM had a clear definition of the relevant product markets. For example, in the cases of Pfizer/Wyeth,11 Panasonic/Sanyo12 and Novartis/Alcon,13 multiple relevant product markets were defined. As to the relevant geographical markets, however, MOFCOM was not always clear. In Pfizer/Wyeth,14 MOFCOM clearly defined China as the geographic market. But in Novartis/Alcon15 it was not clear whether the relevant geographic market was the world market or the Chinese national market. See the Interim Rules, cited above note 8. See the US Horizontal Merger Guidelines, and the EU Horizontal Guidelines, cited above notes 6 and 7. 11 See MOFCOM, Pfizer/Wyeth Decision, Notice No. 77 [2009], 29 September 2009, http://fldj. mofcom.gov.cn/aarticle/ztxx/200909/20090906541443.html 12 See MOFCOM, Panasonic/Sanyo Decision, Notice No. 82 [2009], 30 October 2009, http://fldj. mofcom.gov.cn/aarticle/ztxx/200910/20091006593175.html 13 See MOFCOM, Novartis/Alcon Decision, Notice No. 53 [2010], 13 August 2010, http://fldj. mofcom.gov.cn/aarticle/ztxx/201008/20100807080639.html. 14 See MOFCOM, Pfizer/Wyeth Decision, cited above note 11. 15 See MOFCOM, Novartis/Alcon Decision, cited above note 13. 9

10

Xinzhu Zhang and Vanessa Yanhua Zhang 

307  

An interesting question is what approach MOFCOM has used to define relevant markets. From the information released, it seems that MOFCOM has depended mostly on a qualitative economic approach, which is to analyze demand and supply substitutability as contemplated in the Market Definition Guidelines.16 For example, there is no indication that MOFCOM has even used critical loss analysis, a common methodology by which to implement the SSNIP test in many jurisdictions. More generally, there is no clear indication that a quantitative economics approach has been used to define relevant markets.

(2) Market power Market power is the pre-condition for any merger to create competitive harm. The AML and the Interim Rules have specified the legal principles applied in order to infer marker power. But they fail to provide detailed rules on how to implement those principles in practice. This would therefore appear to be an area in which there is significant legal uncertainty. Indeed, market share is one of the most important indicia that have been used by MOFCOM in its merger reviews. In particular, MOFCOM is more concerned with market shares that are close to or over 50 percent. In the Pfizer/Wyeth merger,17 for example, MOFCOM found that the combined post-merger market share of Pfizer and Wyeth would be 49.4 percent in the swine mycoplasmal pneumonia vaccine market, (Pfizer 38 percent, Wyeth 11.4 percent), which was considered significantly higher compared to their nearest rival Intervet (18.35 percent) and other individual competitors (less than 10 percent). With such a post-merger market share, MOFCOM was worried that the merging parties would have the ability or market power to exploit the market and unilaterally raise the price.18 Besides market share, MOFCOM has also taken into account market concentration as another important measure of market power. For example, in Pfizer/Wyeth,19 MOFCOM indicated that the post-merger HHI would be 2182 and that the change of HHI pre- and post-merger (i.e., the “delta”) would be 336. With this market concentration information, MOFCOM concluded that the swine mycoplasmal pneumonia vaccine market in China was highly concentrated and that the proposed transaction would have anticompetitive effects. The use of market shares and concentration seems to drive MOFCOM’s legal approach in ascertaining marker power. But since there are no safe harbors specified in the Interim Rules, one may wonder on what basis MOFCOM has treated such indicators in practice. One possibility is that they might have used the benchmarks common in other jurisdictions such as the EU and US.20 Another possibility is that See Article 4 of the Market Definition Guidelines, cited above note 5. See MOFCOM, Pfizer/Wyeth Decision, cited above note 11. 18 Ibid. 19 Ibid. 20 In the EU, the Commission would not identify horizontal competition concerns “in a market with a post-merger HHI below 1000”. It would not identify horizontal competition concerns “in a merger with a post-merger HHI between 1000 and 2000 and a delta below 250, or a merger with a 16 17

308  Merger Control in European and Global Perspective MOFCOM might have been guided by the market share thresholds mentioned in Article 19 of the AML, which are used to assess the abuse of dominance; but this raises the question of whether the same standard of proof for presuming market power should be adopted for these different types of potentially anticompetitive behaviours. One may also wonder how rebuttal evidence, if any is allowed, might be evaluated in practice.

(3) Competitive effects The Interim Rules stipulate in Article 4 that two types of competitive harm will be considered for antitrust review of horizontal mergers,21 namely, unilateral effects and coordinated effects. Such legal rules are consistent with economic theory and with international practice. However, the Interim Rules are silent as to whether there is any difference in the assessment of these two types of competitive harm. In particular, the Rules fail to specify how to assess unilateral effects in differentiated product markets. For example, they give no indication about whether diversion ratios should be emphasized in addition to market shares and concentration data. In the horizontal merger cases we have analyzed, MOFCOM was concerned about the presence of unilateral price and/or non-price effects. As mentioned above, MOFCOM has to presume this type of competition harm mainly based on market shares and/or market concentration. But one problem might be that it has placed undue emphasis on post-merger market shares or concentration ratios instead of the change of market structure. For example, in the Novartis/ Alcon22 case, although Novartis had a market share of over 60 percent in China’s ophthalmological anti-infective, anti-inflammatory/anti-infective combinations market, Alcon’s market share was only less than 1 percent. In other words, the merger would not cause a lessening of competition in this market due to Alcon’s negligible market share. In the cases of Novartis/Alcon23 and Uralkali/Silvinit,24 MOFCOM was worried about not only unilateral effects but also coordinated effects. In particular, MOFCOM considered only coordinated effects in the contact lenses care products market in the Novartis/Alcon25 case, but was concerned about both types of competition effects in the Uralkali/Silvinit26 case. In considering coordinated post-merger HHI above 2000 and a delta below 150”, except in certain special circumstances listed. See paragraphs 19-20 of the EU Horizontal Merger Guidelines, cited above note 7. In the US, the antitrust enforcement agencies would consider that mergers resulting in an unconcentrated market, i.e., a market with an HHI below 1500, “are unlikely to have adverse competitive effects and ordinarily require no further analysis.” See the 2010 US Horizontal Merger Guidelines, cited above note 6. See Article 4 of the Interim Rules, cited above note 8. See MOFCOM, Novartis/Alcon Decision, cited above note 13. Ibid. 24  See MOFCOM, Uralkali/Silvinit Decision, Notice No. 33 [2011], 2 June 2011, http://fldj. mofcom.gov.cn/aarticle/zcfb/201106/20110607583288.html. 25 See MOFCOM, Novartis/Alcon Decision, cited above note 13. 26 See MOFCOM, Uralkali/Silvinit Decision, cited above note 24. 21 22 23

Xinzhu Zhang and Vanessa Yanhua Zhang 

309  

effects, it seems that MOFCOM has depended largely on the combined share of the largest two or three firms. However, from the information released by MOFCOM, it was not clear whether the market would become not only more concentrated but also more symmetrical.

(4) Factors suggesting limited market power Even though market shares and concentration data have been taken as a basis on which to presume market power and competition harm, counter-indications were also considered, albeit perhaps to a lesser extent. In particular, we refer here to conditions of entry and countervailing buyer power. i. Entry In all cases, MOFCOM has had a brief entry analysis. In general, MOFCOM focused on industrial specifics such as contractual relationships (GM/Delphi27), economies of scale (GM/Delphi28), switching costs (Panasonic/Sanyo29), intellectual property rights (Pfizer/Wyeth,30 Panasonic/Sanyo,31 Novartis/Alcon32 and Alpha V/Savio33), and monopoly of natural reserves (Uralkali/Silvinit34 and GE/Shenhua35) etc., to assess whether there were significant entry barriers. The interesting question seems to be: although the Interim Rules have specified the principles by which to conduct entry analysis,36 which emphasize that entry must be possible, timely and sufficient, why haven’t there been specific rules on how to implement these principles? In particular, it is unclear what standard of proof applies for purposes of establishing the presence or absence of significant entry barriers. ii. Buyer power MOFCOM also considers buyer power in merger reviews to assess whether it can outweigh the increased market power created by the proposed mergers. For example, in Panasonic/Sanyo, MOFCOM found that buyer power was not strong enough to offset the anticompetitive effects of the merger.37 Although some large downstream users may have bargaining power vis-à-vis the merged entity, the countervailing effect cannot be extended to small and medium-sized users who do not have such power. 27 See MOFCOM, GM/Delphi Decision, Notice No. 76 [2009], 28 September 2009, http://fldj. mofcom.gov.cn/aarticle/ztxx/200909/20090906540211.html. 28 Ibid. 29 See MOFCOM, Panasonic/Sanyo Decision, cited above note 12. 30 See MOFCOM, Pfizer/Wyeth Decision, cited above note 11. 31 See MOFCOM, Panasonic/Sanyo Decision, cited above note 12. 32 See MOFCOM, Novartis/Alcon Decision, cited above note 13. 33 See MOFCOM, Alpha V/Savio Decision, Notice No. 73 [2011], 31 October 2011, http://fldj. mofcom.gov.cn/aarticle/zcfb/201111/20111107809156.html. 34 See MOFCOM, Uralkali/Silvinit Decision, cited above note 24. 35 See MOFCOM, GE/Shenhua Decision, Notice No. 74 [2011], 10 November 2011, http://fldj. mofcom.gov.cn/aarticle/zcfb/201111/20111107824342.html. 36 See Article 7 of the Interim Rules, cited above note 8. 37 See MOFCOM, Panasonic/Sanyo Decision, cited above note 12.

310  Merger Control in European and Global Perspective

b. Non-horizontal mergers For non-horizontal merger reviews, two issues are particularly important. One is the higher standard of proof required in comparison with horizontal merger reviews. The other is consistency between the review of non-horizontal mergers and the investigation and prosecution of abuse of dominance. Compared with other jurisdictions such as the US and EU, it seems that MOFCOM has been more concerned with the possibility of competitive harm from non-horizontal mergers. Indeed, the only case that has been blocked so far, i.e. the Coca-cola/ Huiyuan38 merger, involved portfolio effects in a conglomerate merger setting. And in several vertical merger cases with conditional approval, MOFCOM has focused on foreclosure effects. Even though there are not many details in the Interim Rules, MOFCOM has mainly followed the way the European Commission deals with non-horizontal mergers. In reviewing vertical mergers, MOFCOM often considers foreclosure effects to be the major competitive harm. This effect may emerge in an upstream or downstream industry depending on the market power of the relevant merging parties in the production chain. The key point is that the foreclosure effect must be causally linked to the vertical merger in question. In the GM/Delphi merger,39 for example, MOFCOM found that Delphi was the exclusive auto parts supplier to many domestic Chinese automakers that competed with GM. This exclusive supply relationship and the competition between GM and its rivals in the downstream market made MOFCOM concerned that the proposed transaction might have anticompetitive impacts on the stability of supply, prices, and quality of auto parts that used to be produced by Delphi and sold to other domestic automakers. The proposed transaction might thus eliminate competition in the downstream automobile industry by foreclosing other domestic automakers. In particular, MOFCOM was also worried about a hold-up problem between Delphi and other domestic automakers and concerned that Delphi might increase switching costs for downstream automakers, thus dissuading them from turning to other auto parts producers. In the GE/Shenhua40 case, MOFCOM was concerned by the fact that Shenhua was the major supplier of the specific raw coal necessary for the Coal-Water Slurry (CWS) gasification technology. Thus the proposed transaction, if not remedied, might eliminate competition in the market for the CWS gasification technology by foreclosing other competitors that owned the technology. Based on the information released, there might be two problems with these arguments. One is that MOFCOM failed to prove that the merging parties would have the ability to foreclose competitors downstream. In the GM/Delphi case, for example, at least based on economic theory and international practice, harm could only materialize if Delphi had significant market power in the auto parts market. 38 See MOFCOM, Coca-cola/Huiyuan Decision, Notice No. 22 [2009], 18 March 2009, http://fldj. mofcom.gov.cn/aarticle/ztxx/200903/20090306108494.html. 39 See MOFCOM, GM/Delphi Decision, cited above note 27. 40 See MOFCOM, GE/Shenhua Decision, cited above note 35.

Xinzhu Zhang and Vanessa Yanhua Zhang 

311  

But MOFCOM only mentioned that Delphi was a leader in global and Chinese auto parts markets and there was no further analysis, nor information regarding market shares or the concentration ratio for the auto parts market. Meanwhile, in the GE/Shenhua case,41 MOFCOM only mentioned that Shenhua was the major supplier of the specific raw coal. However, it did not provide any information on whether Shenhua was dominant in the upstream market, so that the JV would have the ability to foreclose its competitors in the downstream market. The other problem is that MOFCOM has failed to show that post-merger the merging parties would have incentives to foreclose competitors. In the GM/Delphi case42 or the GE/Shenhua case,43 for example, if the merging parties legitimately had market power in the upstream market and therefore would earn abnormal profits, why would it have an incentive to exclude competitors in the downstream market? This reminds us of the Chicago critique of the single monopoly profit theory. In other words, more evidence should be needed to prove that such a vertical merger could indeed increase profit in some other markets or could help protect rents in the auto parts or the CWS gasification technology market. Meanwhile, GM might also increase procurement of auto parts from Delphi, which would potentially make it more difficult for other domestic auto parts makers to enter GM’s procurement system. The merging parties might put other auto parts makers in an inferior position in the upstream auto parts market by foreclosing Delphi’s existing and potential competitors. Obviously, MOFCOM was also concerned with GM’s market power in the downstream auto market. To prove foreclosure theory in this case, a higher standard of proof is needed, which means that MOFCOM should prove not only that GM had market power in the downstream market and thus had the ability to exclude competition, but also that it had incentives to do so.

IV. Patterns and implications As more rules and regulations are issued and more merger cases reviewed in China, MOFCOM is building its capacity to deal with cases more efficiently and effectively. From the published decisions and from the merger filings in which we have participated (either as independent economists for MOFCOM or as economists engaged by filing firms to prepare competition analytical reports), we observe certain patterns emerging which provide important indications as to the direction of MOFCOM’s enforcement policy in the future.

41 42 43

Ibid. See MOFCOM, GM/Delphi Decision, cited above note 27. See MOFCOM, GE/Shenhua Decision, cited above note 35.

312  Merger Control in European and Global Perspective

a. Reliance on market share data to assess market power In this early period of enforcement, MOFCOM has assigned a lot of weight to market shares when assessing market power. Indeed, in most if not all cases, market shares and, to a lesser extent, concentration ratios are required information and must be submitted to MOFCOM. This information will then typically be used by MOFCOM to determine whether the merging parties can exercise market power in the relevant market. Since there is no regulation with regard to how and what MOFCOM should request from the filing party, MOFCOM can essentially demand any information it deems necessary at any time. For example, if the relevant market defined by MOFCOM is different from what was claimed by the filing parties, MOFCOM may also request market share information in relation to certain product or geographic markets that it may be interested in. So far, the issue of sources of evidence has not yet been raised. Indeed, information from third parties is generally accepted even though information from official sources is given priority.44

b. More concerns about non-horizontal mergers In assessing competitive effects of horizontal mergers, China’s experiences are more or less consistent with international practice in the sense that MOFCOM has focused on unilateral effects and coordinated effects. However, in non-horizontal merger reviews, it seems that MOFCOM has been more concerned with non-horizontal competition effects than other jurisdictions such as the United States and the European Union. In such cases, MOFCOM has not recognized a need for a higher standard of proof for claims of competition harm. This may have something to do with the unique organizational structure of enforcement agencies. Indeed, unlike the jurisdictions where merger and non-merger cases are typically investigated in an integrated way within a single agency, in China the decentralization of enforcement powers (which effectively separates merger enforcement from non-merger enforcement) might create some coordination problems due to (negative) externalities that merger and non-merger enforcement agencies impose on each other.

c. Use of a trustee to monitor the remedies In recently published decisions, MOFCOM has required both structural and behavioral remedies. Due to its capacity constraints and the increasing number 44 Interestingly, there appeared to be some incoherence here. In the Renren v. Baidu case, which was a case before the People’s Court, the plaintiff submitted two pieces of market share information, one appearing on Baidu’s website and the other on the Securities Daily, a national newspaper. But the court dismissed such information, which would likely be accepted by MOFCOM.

Xinzhu Zhang and Vanessa Yanhua Zhang 

313  

of merger filings, it is difficult for MOFCOM to supervise the implementation of remedies. Often, MOFCOM allows the use of trustees to monitor the implementation process. In the Novartis/Alcon case,45 for example, MOFCOM ordered Novartis to appoint a monitoring trustee to supervise the implementation of remedies according to the newly-issued divestiture guidelines.46 In Uralkali/Silvinit,47 MOFCOM allowed the merged entity to appoint a monitoring trustee to report to MOFCOM on the implementation of the behavioral remedies to ensure compliance.

d. Differences compared to the US and EU approaches In the review process and in published decisions, MOFCOM has chosen its own approach to implement the rules and provisions of merger control, even if the principles behind the rules and provisions appear to be very similar to those of the US and the EU. Two good examples are the GM/Delphi and Seagate/Samsung cases, which involved vertical and horizontal mergers respectively. Both mergers were cleared in the US and the EU without any conditions. MOFCOM, however, imposed remedies on both mergers. In the Seagate/Samsung case,48 in particular, MOFCOM imposed a very rare type of remedy to require Seagate to operate Samsung’s hard disk drive business as a separate business and independent competitor for at least one year, which would postpone the deal from closing and increase the transaction costs for the merging parties.49

V. Conclusion China’s merger control policy has combined the principles of U.S. and EU merger control while forging its own way forward. Compared to other jurisdictions, it has grown dramatically within a short time period of just three years. Although it has drawn some criticism from scholars and practitioners, and though there is much room for improvement, MOFCOM has been on the right track to build an independent and transparent merger review system. Even though China is not yet a member of International Competition Network (“ICN”), it is open to international antitrust enforcement cooperation and to becoming an active member of the global competition community. See MOFCOM, Novartis/Alcon Decision, cited above note 13. See Press Release, MOFCOM, Provisions on Divestiture of Assets or Businesses to Implement Concentrations of Undertakings (“Divestiture Guidelines”) (5 July 2010), http://fldj.mofcom.gov.cn/ aarticle/zcfb/201007/20100707012000.html. 47 See MOFCOM, Uralkali/Silvinit Decision, cited above note 24. 48 See MOFCOM, Seagate/Samsung Decision, Notice No. 90 [2011], 12 December 2011, http://fldj. mofcom.gov.cn/aarticle/zcfb/201112/20111207874274.html. 49 For more discussion of this type of ring-fencing remedy, see François Renard, “A Practitioner’s Look at Merger Control Remedies in China”, CPI Antitrust Chronicle, January 2012 (2). 45 46

Abel Mateus*

The New Brazilian Merger Control Regime

1. Introduction The new Brazilian regime for merger control was established by Law 12529 of 30 November 2011;1 it replaced Law 8884 of 11 June 11 1994 and Law 9781 of 19 January 1999, among other legal texts. The 2011 law brings the legal regime largely in line with OECD and ICN guidelines and recommendations. The Brazilian System of Defense of Competition (Sistema Brasileiro de Defesa da Concorrência) was composed of three institutions distributed between the Ministry of Justice and the Treasury. The new regime consolidates the merger review process and assigns it to a single competition agency, following a trend observed in the last decade among OECD countries, and responds to one of the major criticisms of the OECD.2 It establishes an obligatory pre-notification system and new thresholds for notification. It also defines more precisely the definition of concentration and the criteria for analysis of a concentration. It extends the term for analysis of concentrations by the Authorities. But some of the flaws and limitations of the previous laws have not been corrected, like the uncertainty with regard to the term for merger review by the Authorities and the failure to clearly establish two systems for merger review: a simplified and an in-depth investigation, similar to those in most OECD countries.3 In institutional terms, the two most important changes were the reduction of antitrust enforcement authorities from 3 to 2, and the increase of the terms of the pivot institution board members from 2 to 4 years. The next section describes the new regime and the major changes introduced. Section 3 discusses problems of implementation and some limitations still remaining, and Section 4 concludes.

2. The new regime and major changes Law No. 8.884/1994 introduced the current institutional framework of the Brazilian Competition Policy System (“BCPS”), comprised of two investigative * Universidade Nova Lisboa; University College London. 1 The new law took effect on 29 May 2012. 2 “Competition Law and Policy in Brazil”, OECD Policy Brief, September 2005, page 4. 3 The authorities had already established a “fast track” system for simple mergers.

316  Merger Control in European and Global Perspective and advisory agencies, the Secretary of Economic Monitoring at the Ministry of Finance (“SEAE/MF”) and the Antitrust Division of the Secretary of Economic Law at the Ministry of Justice (“SDE/MJ”),4 and a third component, the Council for Economic Defense (“CADE”), an administrative tribunal that issues final rulings in both merger and conduct cases. The inefficiencies of the current system became apparent fairly quickly, most of them related to the mandatory postmerger review system, the overlapping functions of the three agencies, and their lack of resources. As a result, policy makers began proposing amendments to the antitrust statute beginning in early 2000, but most were not enacted. Another important limitation was the excessive rotation of the management at CADE, where members were chosen by the government for terms of only 2 years. The new regime for merger control involved reforms in the following areas:

Introduction of pre-merger notification system and new notification thresholds The new law establishes a mandatory pre-merger notification system, whereby the parties must secure approval before closing the deal. This is a major change compared to the previous system, which permitted parties to close the deal before CADE´s approval. A transaction must be notified5 if one of the parties has reported revenues in Brazil of at least 400 million reais6 (US $215 million) in the previous year, and at least one of the other parties had revenues of at least 30 million reais (US $16 million). The new law eliminates the previous rule that established mandatory notification when one of the parties had a 20 percent market share. There is a clawback clause that allows CADE to review mergers that fall outside of those thresholds up to 1 year after the transaction.7

Consolidation of merger review responsibilities within a single competition agency The new law consolidates investigative, prosecutorial and adjudicative functions into one agency (CADE): CADE will remain the decision-making body in charge of rendering final and binding administrative decisions with respect to notified mergers. The “Superintendência Geral” will investigate anticompetitive practices and will be empowered to render administrative decisions to approve mergers.8 The SDE has concentrated on anti-cartel enforcement. The 45,000 reais (US $24,000) notification fee is retained and is allocated entirely to CADE. 6 This is the same threshold that was established by the 1999 law. 7 This is likely to happen only in cases where the parties involved have a high market share in the relevant market, but low overall turnover in Brazil (e.g., in very local geographic markets). 8 The “Superintendência Geral” will conduct merger investigations and decide whether a transaction will be (i) approved without further inquiry within 60 days, or (ii) deemed complex and subject to additional investigation. 4 5

Abel Mateus 

317  

The Department of Economic Studies, led by CADE´s chief economist, will be responsible for rendering non-binding economic opinions and preparing economic studies. In order to ensure a certain level of internal autonomy and control, CADE will be restructured to include: (i) an administrative tribunal composed of six Commissioners and a President; (ii) a Directorate General for Competition (“Superintendência Geral”, or SG); and (iii) an Economics Department. The new Directorate General will perform the former functions of SDE’s Antitrust Division and SEAE. SEAE will continue to exist but it will deal exclusively with “competition advocacy” before the Brazilian regulatory agencies and other governmental bodies.

Criteria for approval or rejection of concentrations Article 88, paragraph 5, establishes the criteria for the prohibition of a concentration: “Concentrations that eliminate competition in a substantial part of the relevant market, that create or reinforce a dominant position or that may result in a dominant position in a relevant market of goods or services, will be prohibited”. However, paragraph 6 provides for exceptions to that rule, authorizing concentrations which, “within the strictly necessary limits, pursue the following objectives: I – cumulatively or alternatively: a) increase productivity or competitive position, b) improve the quality of a good or service, or c) contributes to efficiency and technological or economic development; and II – will transfer to consumers a relevant part of the resulting benefits.” Although these are more precise definitions of the criteria established in Article 54, paragraphs 1 and 2 of Law 8884 of 11 June 1994, they parallel the previous criteria. Paragraph 5 closely follows the EU substantive standard, which mixes the “substantial lessening of competition” criterion with the “dominant position” criterion. However, the exceptions established in paragraph 6, which seem to echo Article 101 TFEU, are not usually considered in other legislative regimes. It is clear that quality, efficiency and technological development are all dimensions of competition or part of the dominance test.

Modification of penalties for anticompetitive behaviour for violation of competition law Corporations can now be fined from .1 to 20 percent of a company’s (or group’s) gross revenue earned in a relevant market from the year before the initiation of the investigation. This is a change from the current system from 1 to 30 percent of a company´s turnover, without specification of a particular market.

318  Merger Control in European and Global Perspective In case of lack of notification the merger transaction is annulled and there is a penalty between 60 thousand (US $32 thousand) and 60 million reais ($32 million). The same fine applies where false information is provided or if the conditions for approval of the merger have not been respected.

3. Formal steps for merger review Chapter II of the new law establishes the administrative procedure for reviewing mergers. Article 53 stipulates that the notification has to be made to CADE, according to information and forms pre-defined by the authorities. After the notification, the SG must analyze and decide on the completeness of the information presented, requiring the parties to amend the notification if any key item is missing. Once complete, the SG will evaluate the notification and must decide whether or not it will review the matter. If not, the act is officially approved. If the case is to be reviewed, the SG can require further investigations in case more information is needed, and it publishes information about the proposed concentration. The SG can declare that the concentration is complex and that complementary information is required, in which case the period of review is extended. At the end of its review, the SG issues a formal decision, either provisionally approving the act or challenging it before the Tribunal. In the first case (i.e., provisional approval), an appeal to the Tribunal can be lodged within 15 days by any interested party or by a relevant regulatory agency. Such an appeal will be assigned to a randomly chosen Reporting Commissioner, who can review or file it (therefore approving the transaction). If no appeals are presented within the 15-day period, the transaction is formally approved. If the SG decides to challenge the transaction, then it must demonstrate that the “concentration act” will or can produce adverse effects on competition, and it recommends that the Tribunal either approve the deal with restrictions or block it (Article 57). In such a case, the parties have 30 days to present a formal defence of the deal, providing all studies or opinions they consider to be relevant (Article 58). Afterwards, the Tribunal randomly assigns the concentration to one of the members of the board. The Reporting Commissioner can ask for any extra information it deems necessary, including other opinions by either the Department of Economic Studies or the Attorney’s Office (Articles 59 and 60). It can also impose conditions for the approval or reversibility of the transaction. The final decision will be issued by the new CADE Tribunal in a formal judgment session by unanimous or majority vote. The new law provides for the possibility that parties negotiating with CADE may propose commitments to deal with possible anticompetitive effects identified by the authority. Such commitments will be negotiated with either the SG or the Reporting Commissioner, and must be approved by the Tribunal.

Abel Mateus 

319  

4. Some limitations and problems of implementation There are still some insufficiencies in the law that would need to be addressed, to the extent possible, by case law and by regulations enacted by CADE.9 The first problem is that there is no automatic approval of transactions upon expiration of the statutory time period for review by the authorities. A final administrative decision must be issued within 240 days from the date of notification,10 which may be extended by 60 days at the request of merging parties or by 90 days if CADE determines that it requires more time for review, for a maximum of 330 days.11 The government vetoed automatic approval beyond that time limit because of possible “harm to society”, thus raising issues of uncertainty with regard to the statutory period. Second, the government removed the provision that allowed the “Superintendência Geral” to negotiate settlements with merging parties before CADE had commenced its review of the merger. Thus, settlements can be negotiated at any time. Third, following changes introduced by the government, the new law began to apply immediately to all mergers notified after the law became effective. There is no transition period. Another weak point in the law is the relatively coarse definition of “concentration”, compared with the laws of most OECD countries.12 According to Article 90, a concentration arises when: (i) two or more companies merge; (ii) one company acquires sole or joint control of the stock or assets of another, or even a minority shareholding; (iii) an incorporation of other companies takes place; or (iv) a joint venture, an association or a consortium is formed. The new provisions will not apply to consortia that are formed in connection with public bids. The new law also does not set out deadlines or a recommended period for filing.

5. Conclusions The new antitrust law represents a substantial improvement over the previous laws, and responds to most of the weaknesses previously identified by the OECD and ICN. There are still some areas where further clarification and legal certainty will be welcomed by firms. It will be important to see whether CADE will allow, and how it will proceed with respect to, pre-merger communication reviews. CADE should also provide guidance about all cooperative agreements and other 9 The BCPS has applied the 2001 Horizontal Merger Guidelines, approved by the Joint Ordinance SDE-SEAE # 50, which follows standards similar to those of other jurisdictions. 10 As of November 2011, the average merger review time in Brazil was 148 days. 11 The provision applicable to the review period of simple cases (up to 20 calendar days) was excluded from the final version approved by the Congress. 12 See, e.g., Article 3 of the EU Merger Regulation (Regulation 139/2004).

320  Merger Control in European and Global Perspective transactions which fall below the turnover threshold but which it intends to review under the clawback procedure, in view of the “open-handedness” of the provision. It will also be interesting to see how lengthy the review of simple cases will be. Finally, CADE had serious problems due to a lack of human resources. The new law increases its staff by 200 persons, which may not be enough given its enhanced responsibilities. Moreover, it may take some time to fully staff the vacant new positions. But with the modernized legal statutes and one of the best track records in Latin America, the Brazilian system for protection of competition can now look forward to still better law enforcement and policy enhancement.

References CADE, Revista de Direito Economico, December 1999 Clark, John, “Competition Law and Policy Developments in Brazil”, 2(3) OECD Journal of Competition law and Policy 181 (2000) OECD Policy Brief, “Competition Law and Policy in Brazil”, September 2005 Martinez, Ana Paula and Mariana Tavares de Araújo, “Brazil´s New Competition Law: Promising but Challenging”, CPI Antitrust Chronicle, November 2011 (2)

Maher Dabbah*

Merger Control in Middle Eastern Countries: A Perspective on Challenges and Opportunities

1. Introduction The area of merger control has in recent times become the focus of particular attention in different parts of the developing world. It is no longer mere academic indulgence to discuss merger control in developing and emerging economies. Looking around the world, several of these economies have established some form of merger regulation. Business firms, including some of the world’s leading businesses, who are involved in cross-border merger transactions – increasingly – have to notify their mergers to competition authorities in these economies. A number of such firms have had their mergers subjected to extremely close scrutiny in these economies: some have faced conditional clearance of their mergers;1 and there are firms who were even confronted with merger prohibition.2 Undoubtedly, this represents a dramatic shift in position in relation to developing and emerging economies. Only as recently as 5 or 6 years ago, the notification of cross-border mergers (in particular) in some of these economies took the form of pushing a small envelope containing a simple ‘notification letter’ under the door of a government official. This is no longer the case. Nowadays, merger notifications need to be handled properly in these economies: questions will be asked by the relevant competition officials; discussions between the competition authority and the merging firms will take place; further information will be sought; and ultimately the authority will give a decision. The picture described above is quite representative of the situation prevailing in some Middle Eastern countries. A number of these countries have made admirable progress in building effective domestic merger control regimes, which is important to recognize. Nonetheless, in the majority of Middle Eastern countries, effective merger control remains absent: some countries which have merger rules do not have a proper merger review mechanism in place; some countries have extremely brief merger rules in both wording and scope; some countries do not have merger * Chair, Competition Law and Policy; Director, Interdisciplinary Centre for Competition Law and Policy (ICC), Queen Mary, University of London. 1 A number of high profile cases may be mentioned, including: the Panasonic/Sanyo merger (2010), and the Coca-Cola/Cadbury Schweppes merger (1998). 2 This happened in the Coca-Cola/Huiyuan Juice Group merger (2008), when China’s MOFCOM decided to block this transaction.

322  Merger Control in European and Global Perspective rules at all; and, some countries – a very small minority – do not even have a competition law regime. It is understandable why the area of merger control has witnessed modest progress in the majority of Mediterranean countries. Nonetheless, this situation merits improvement. The need for improvement, in this case, is triggered by, among other things, the presumption that has in recent years developed in favour of having merger control in economies similar to those found in the Middle East. This presumption rests on a number of factors, the most notable of which are: the special nature of markets in these economies; the adverse effect some mergers may have on competition and consumers; and the fact that merger control can contribute to a better structuring of markets, stronger economic performance, and better outcomes for businesses. To obtain the desired improvement in the relevant Middle Eastern countries in the area of merger control is not an easy task. Specifically, it would be important to overcome many challenges facing these countries. Equally, however, it is important to recognize the opportunities which exist for these countries in this important area of competition law and policy. The discussion below will define and highlight the context of these challenges and opportunities in the case of Middle Eastern countries. The discussion will not, however, attempt to identity and assess all of these challenges individually, as this is beyond the scope of the discussion.3

2. A flavour of the challenges and opportunities in context It is sometimes the case that when people consider the challenges facing countries like Mediterranean nations in the field of competition law, they tend to group mergers with business phenomena, such as cartels, abuse of dominance, vertical restraints and even state aid; and, on that basis, they tend to develop a “common” approach to all of these different business phenomena in addressing these challenges. It is indeed the case that some of the challenges facing these countries are common to all of these business phenomena. However, other challenges are unique to mergers as business operations and merger control as a regulatory mechanism for assessing these operations. Specifically, one should highlight, as a starting point, the special nature of mergers – which sets them apart from other business phenomena in the field of competition law. As is widely known, mergers are – by nature – structural operations. This nature forms the “centre of gravity” of many challenges, which all countries – including Mediterranean ones – face in the area of merger control. 3 For a discussion of challenges facing developing economies in the area of merger control, and in particular those challenges concerning cross-border merger review, see Maher Dabbah, “Cross-border merger control: challenges for developing and emerging economies”, report prepared for the OECD (2011), http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=DAF/COMP/ GF(2011)1&docLanguage=En..

Maher Dabbah 

323  

Insofar as mergers are structural operations, the link between mergers and any possible adverse effect on competition is not easily established in practice and is therefore not always apparent. More importantly, perhaps, to be able to establish this link, a framework is required which should demand that certain mergers must be notified in advance – ex ante – for assessment.4 Nonetheless, the ex ante option means that there will always be mergers – the majority in fact – which will end up being notified but which do not raise any competition concern. This, in itself, is an articulation of one of the key challenges in the area of merger control: at what level should the competition authority set the thresholds for mandatory notification? The challenge here is: • On the one hand, not to set the thresholds at a level which is too high and so risk problematic mergers escaping scrutiny; and • On the other hand, not set the thresholds at a level which is too low, which would result in many small and insignificant mergers being caught. In determining the appropriate balance between “too high” and “too low” in this case, the relevant country would need to effectively balance between important interests. These include: first, and above all, the interests of competition and of consumers who deserve protection from possible anticompetitive outcomes of merger operations; secondly, the wider public interest; thirdly, the circumstances of the competition authority especially in terms of constraints on its resources; and fourthly, the interests of merging firms. Achieving the appropriate balance in this case is not a simple task, in particular given that these interests are not necessarily in conflict with one another. For example, where the notification thresholds are set too low, this has implications and consequences of a greater usurping of the resources of the competition authority and – at the same time – consequences of high costs for more business firms who decide to merge. Nonetheless, determining the appropriate thresholds for merger notification is not the only challenge to be noted in this regard. Related to the issue of thresholds, there is the question of whether these thresholds should apply to “market shares” or “turnover”. It may be worth mentioning the approach formulated by the International Competition Network (ICN) to this question: the ICN favours using turnover as opposed to market shares in order to ensure objectivity in merger notification.5 A number of Mediterranean countries, however, use market share thresholds. This includes Jordan and Morocco. In other countries, market share thresholds are used at least in part or as an alternative, as is the case in Tunisia and Turkey. The discussion above offered some flavour from the angle of challenges. However, merger control is not just about challenges: it is also about opportunities for Mediterranean countries. 4 The ex post option is being discounted here on the ground that the vast majority of regimes around the world include an ex ante mechanism and because, on balance, the ex ante option enjoys greater merits than the ex post option. 5  See ICN Recommended Practices for merger notification procedures, http://www. internationalcompetitionnetwork.org/uploads/library/doc588.pdf.

324  Merger Control in European and Global Perspective One should recognize that there is a big opportunity for competition authorities in the area of merger control, namely the ability to determine and condition the level of competition in markets directly and effectively. This opportunity can also extend, in some cases, to facilitating competition in local markets in the relevant jurisdictions. Competition authorities can achieve this because they are able to apply their actions directly to the structure of markets, whether through preventing mergers (which are structural operations) or through imposing structural obligations as part of conditional merger clearance. This is particularly important in relation to markets that are being opened up or scheduled to go through privatization. Many local markets in Middle East countries fit this description. An example illustrating the importance of merger control in the context of liberalization comes from Turkey. One of the anxieties of the Turkish Competition Authority in the late 1990s was over the creation of monopolies as part of liberalization in local markets. Thus, in 2000, the Authority rejected the privatization of IGSAS because of possible adverse effects on competition in the market for nitrogen-based and composite fertilisers. This rejection was motivated by the strong position of the buyer in the relevant market. The Authority eventually allowed the sale of IGSAS in 2003 to another buyer which had no activity in the relevant market. Competition was thus facilitated in this market. This certainly is an inspiring example, although one must be realistic: such enforcement action in a particular context cannot easily be envisaged in the case of all Mediterranean competition authorities. Three particular reasons may be offered here. First, not every country in the region is in favour of using merger control for intervention in markets when they are still at a liberalization stage. Tunisia, for example, advocated against such intervention in the past on the ground that this could send the wrong message to businesses.6 Secondly, in some countries in the region there are exclusionary provisions in the competition law preventing its application in relation to certain firms, as in the case of Egypt,7 Qatar,8 Yemen9 and Kuwait,10 or in relation to certain sectors like banking, as in the case of Turkey.11 Thirdly, in several Mediterranean countries the relevant minister12 or even the prime minister13 may play a decisive role in merger regulation. The involvement 6 See the contribution by Tunisia to the OECD roundtable on ‘Cross-border merger control: challenges for developing and emerging economies’, as part of the OECD Global Competition Forum (2011), http://www.oecd.org/officialdocuments/displaydocumentpdf/?cote=DAF/COMP/GF/ WD(2010)77&doclanguage=en. 7 See Article 9 of the Law on the Protection of Competition and the Prohibition of Monopolistic Practices 2005. 8 See Article 2 of the Law on the Protection of Competition 2006. 9 See Article 4 of the Competition Law 1999. 10 See Article 6 of the Law on Protection of Competition 2007. 11 See Maher Dabbah, Competition Law and Policy in the Middle East, Cambridge University Press, 2007, chapter 4. 12 See the position in Tunisia and Jordan. 13 See, for example, the position in Morocco, although a proposal has been put forward to change this position and give the power to make decisions to the competition authority instead.

Maher Dabbah 

325  

of politicians in the decision-making process in merger cases opens up not only the possibility for pushing aside competition considerations but also the possibility of guaranteeing the dominance of non-competition considerations. The latter possibility is of particular significance in the case of several Middle Eastern countries given that non-competition considerations feature quite prominently in a number of competition law regimes in these countries. For example, Article 11(A) (1) of the Jordanian Competition Act 2004 states that the minister for trade will approve a merger if the merger: does not negatively impact competition or if it has positive economic benefits that outweigh any negative impact on competition, such as enhanced employment opportunities, encouraging exports and enhancing international competitiveness of local firms, and attracting foreign investment. The latter consideration of attracting foreign investment is one that can easily dominate in practice. One high profile example can be found in relation to Morocco. The dominance of foreign investment, as a consideration in the merger review process in the country, derives mainly from the fact that foreign firms involved in a merger may enjoy a particularly strong bargaining position.14

3. Reflections One particular conclusion that may be drawn in light of the discussion in the previous part is that the full picture that emerges when looking at merger control in Mediterranean countries is quite complex and has an element of ambiguity. The importance of this conclusion is enhanced by one crucial fact: merger control – as a mechanism – is inherently uncertain in both nature and operation. For this reason, there should be as much clarity as possible when seeking to regulate merger operations. There is not, however, a magic button that one may press and cause the complexity and uncertainty to disappear. Moreover, at any rate, merger control will never – anywhere and anytime in this world – become free from complexity or uncertainty. One thing that can be done, nonetheless, is to eliminate unnecessary complexity and to minimize the uncertainty. For this to happen in the case of Mediterranean countries, a number of concrete steps must be taken. These steps include the maintenance of a suitable economic environment for businesses to flourish and, where commercially sensible, to merge; and the establishment of a suitable legal framework for merger control, overseen and implemented by independent competition authorities with the necessary expertise and capacity. These steps require a comprehensive approach based on simultaneous actions on a number of fronts. The need for such an approach is explained by the special nature of merger control, which is a multi14 See the contribution by Morocco to the OECD roundtable on “Cross-border merger control: challenges for developing and emerging economies”, as part of the OECD Global Competition Forum (2011), http://www.oecd.org/officialdocuments/displaydocumentpdf/?cote=DAF/COMP/GF/ WD(2010)80&doclanguage=en.

326  Merger Control in European and Global Perspective faceted topic with interfaces between: competition policy and other public policy considerations (most notably social and industrial policy); jurisdictional, procedural and substantive issues; legal regimes and business interests; and (in the case of cross-border mergers) global, regional and domestic interests and considerations. It is never easy to manage such complex interaction between such diverse considerations and forces, which is itself an appropriate articulation of the challenges faced by Mediterranean countries in the area of merger control.

V. Merger control and best practices

5 PANEL DISCUSSION

Chair: Philip Lowe

Speakers: John Boyce Rachel Brandenburger Jochen Burrichter Thomas Deisenhofer Kirsten Edwards

Cal Goldman William Kovacic Howard Shelanski Mario Siragusa Sven Völcker



328  Merger Control in European and Global Perspective

PANEL 5: Merger control and best practices Philip Lowe: This final session is about best practices. What is best, and whose practices they are, is going to be revealed to us by our three speakers. First of all, John. John Boyce: Well, I must confess that when I was asked to focus on best practices in merger control I was a little bit uncertain about what I was meant to be addressing. Was I supposed to be commenting on those guidelines that some competition agencies issue, which are labelled “best practices”? Or was I being given a bit more flexibility, to talk about what I think is the best thing competition authorities should be doing and how we, as practitioners, should do our jobs? The first thing I did, actually, was to go to the dictionary to find out what “best practices” means. I was hoping to find some sort of Shakespearian quote or something showing that it was first used by Henry II (laughter). But to my surprise, the oldest reference in the Oxford English dictionary to the use of the word “best practices” was only in 1984. That was perhaps in the days when accountants were more innovative than they now are, because it pre-dated the concentration in that industry (more laughter). According to the OED, in understanding accounting in 1984, that was the first time that the concept was used. It goes on to talk about another use of it in 1993 by John Kay, who some of you will know as an economist who wrote about corporate success.1 He explained that, if you want to run a successful corporation, one thing you might do, obviously, is to look at what other firms are doing and copy it. This strategy, he said, is more felicitously expressed as “adopting best practice”. So that’s not a bad idea. Just see what other people are doing, copy it, and that’s best practice. Well, things have moved on a little bit since then. One dictionary definition which I did quite like is from businessdictionary.com, which describes “best practices” as “methods and techniques that have consistently shown results superior than those achieved with other means, and which are used as benchmarks to strive for”. It did, however, go on to say that there is “no best practice that is best for everyone (laughter), or in every situation, and no best practice remains best for very long, as people keep on finding better ways of doing things.” Then, if you look at one of my favourite dictionary sources, Wikipedia, it’s very modern, very democratic. It comes up with similar concepts and it actually picks up some of the terminology that Bill was using earlier on. Wikipedia suggests that best practice is to be measured by its effectiveness: getting the best results. Best practice should also be about getting things done in the most efficient way, with the least amount of effort. Now, that all sounds very sensible to me, but Wikipedia actually goes on to say that there is a danger of best practices becoming a bit of a business buzz word. It says that, “as the term has become more popular, some more 1 See John Kay, Foundations of Corporate Success: How Business Strategies Add Value, OUP, 1993/1995. For additional information on Kay, see http://www.johnkay.com/about.

Panel 5 

329  

organisations have begun using the term ‘best practices’ to refer to what are in fact merely rules, causing a linguistic drift in which a new term such as ‘good ideas’ is needed to refer to what would previously have been called ‘best practices’”. I think that really sets the scene for what I think best practices should really be about. It’s all about learning from experience. That’s from your own experience, but also by looking at what other people are doing and by learning from their experiences. They might have done something wrong, so you say that’s what we want to avoid doing. It’s all about learning by doing and learning by watching. If you look at the whole area of international merger control that we’ve been talking about for the last couple of days, it’s a prime area where best practice makes sense. Increasingly, regulators are applying similar rules, and increasingly they clear similar deals, or even the same deal, so we can try to learn from what others are doing. In my paper, I try to focus on three areas where best practices make sense. The first area is best practices by agencies and by lawyers, facilitating the process of advising clients about how to get deals through. We’ve been talking about alliteration, well some time ago as a junior lawyer I learned the 6 P’s: proper preparation prevents piss poor performance (laughter). Now, increasingly that is a problem when you’re advising on international deals; where should you be notifying? I think firms are doing this regularly; you’ve got your databases, you’ve got the guides that people go to. Even on the most mundane deals, you may have to notify in a number of jurisdictions. I was just talking to Nick during the break about how, when I was a junior lawyer, the bane of our lives was being asked: “Can you review this agreement to see whether it is caught by the Restrictive Trade Practices Act in the UK?” Now, I’d have to do this and I’d think that this isn’t what I wanted to do as a competition lawyer. I wanted to deal with effects and problems. So, I think that as a legal community we were all quite pleased to see that piece of arcane legislation repealed, but what’s come along to replace that for our junior lawyers now and for the junior case handlers of the European Commission is dealing with cases that raise no competition issues whatsoever. Now, this is where I think there are some best practices out there. The ICN has been fantastic at trying to point people in the right direction by coming up with common techniques for cases, that is, both cases that shouldn’t raise problems and those that do. One of the key buzzwords for me in this area is transparency. There can be clarity as to what rules should be about, and if agencies are transparent about how their rules operate in practice that makes life much easier for us in practice. If I’ve got a couple of suggestions, my first one picks up on the point that the ICN is not a group of legislators, it’s just agencies so they don’t make any binding rules. However, I think there is still some scope for advocacy by the agencies, just making it clearer to their governments that the rules don’t make sense. It sounds as if John is having the problem the other way around at the moment, in that he’s being pushed in the direction of rules that could be more bureaucratic. I’ve talked a bit about the issue of rules, which are triggered purely by jurisdictional thresholds. I just wonder, raising the same point again, whether

330  Merger Control in European and Global Perspective there is scope for regulators – I used the word “regulator” there actually! – to exercise some self-restraint on whether to enforce the rules. Thomas, you said that you don’t have the power to do that, but to some extent you do. Let’s say a Korean company and a Japanese company set up a coffee shop in Tanzania. Say they happened to meet the thresholds and didn’t think of notifying the European Union, but someone spotted it when they were reading the newspaper in Tanzania while on holiday from the European Commission. I don’t think you would be using the Commission’s resources effectively by requiring them to file a notification. Fold up the paper and move on! (laughter) Another idea that I’ve got for the ICN is that if you hunt around the ICN’s website, you’ll find some summaries that some jurisdictions have put in, showing what their rules are, what their thresholds are. They’re pretty long, and quite difficult to understand, so I wonder whether there might be some scope for cleaning them up a bit and encouraging members to do that. You can do that behind closed doors without NGAs having to hear about it. Possibly agencies could even self-assess and spell out where the rules are ICN best practice-compliant and where they’re not. There’s nothing wrong with having rules that don’t comply with the ICN’s best practices, but it is useful for us as practitioners to know when rules don’t comply. So those are the areas where I think more can be done. A lot has been done to make life easier for us advising clients at the outset on deals. The other area of best practice is for the conduct of investigations, and we talked a lot about that in the previous session. Best practice is all about using the 3 Cs: convergence, cooperation and comity. Thomas talked about challenges, which is another C word. I like C words when it comes to competition! You were talking about the cultural challenges, the confidentiality issues – another C word that you didn’t even mention was a C word – and the challenges of communication that you have dealing with the other regulators. We heard Rachel’s more complicated MRTD acronym, but I took those letters and just threw them into the Scrabble bag and they came back as 4 Cs: mindfulness is consciousness of what the other regulators are doing; respect, well Nick said it – consideration, or courtesy or care. There you are, that’s three words (laughter). Trust is confidence – have confidence in what your other ICN members are doing; and dialogue is, of course, communication. So, there are plenty of C words out there. The last area I wanted to touch on briefly is remedies, which is an important area… Audience: Commitments! (laughter) Boyce: Yes, commitments, sorry! This is an area where, as a practitioner, it is very important to get the regulators working together because the risk of having divergent commitments can actually increase the antitrust “price” and make deals unattractive. Philip was talking right at the end of the last panel about the airlines sector, and I spent some of the past two years dealing with a transatlantic alliance case.2 I found that quite bizarre in some ways, because you have the 2

M.5747, Iberia/British Airways (14 July 2010).

Panel 5 

331  

sort of odd US practices dealing with cooperation in the airline sector, where the DOJ has got no jurisdiction at all and the Department of Transportation deals with it in an incredibly so-called “transparent” way where everything is on the docket. Also, how can you have cooperation between the Commission and the DOT under those circumstances? But you get around that somehow. The normal US approach, where orders to show cause come up quite quickly, as happened with Star Alliance and Sky Team, didn’t seem to happen in our case for some reason. That was because the Commission was cooperating so well, and not just with the Department of Transportation. I’m complimenting the Commission here, actually! (laughter) There was very close cooperation between the Commission and the DOT and there was also very close cooperation between the Commission and the DOJ. Where there didn’t seem to be close cooperation was between the DOT and the DOJ, which I found quite intriguing. I’ve stayed away from talking about best practices in the sense of guidelines because I think Jochen’s going to say a bit about that. In my paper3 I’ve tried to summarize in an annex the available best practice documents. I also commend to you, if you’re not familiar with them, the ICN’s Recommended Practices;4 I’ve summarized them in a separate annex in my paper. They’re very good and they’ve withstood the test of time, being written some five years ago. I don’t think they need to be updated much, even though one of the recommendations is that practices should be updated from time to time. The European Commission is a bit delinquent in that respect, because the Commission’s best practices for mergers pre-date the 2004 merger regulations,5 so all the references are wrong. I’ve also in an annex put down some recommended practices for lawyers, which I think we generally all try to comply with. We have guidelines like that, and I’m sure you do too, that are there to help our junior lawyers work out how you should be dealing with relatively simple cases, which may need to be notified all over the world and where you need to coordinate with local lawyers as required. To conclude, I think best practice is a great thing to strive for. If we keep on developing best practices and coming up with new, good ideas, then ultimately we will get to Utopia. But you’ve got to keep on trying to get there, because even when you think you are there, you’re not. It’s going to change, so keep on moving forward. Lowe: As school children in the North of England, we were actually taught to recite this little rhyme: “good, better, best. Never let it rest, until the good is better and the better best.” (laughter). Next we have Jochen. Jochen Burrichter: Well, it’s very difficult for me to add to what John has already said. I would like to categorize, or to analyze, the existing best practice rules which are in place. The best practices which you can observe provide guidelines 3 Boyce, “Best Practice in Merger Control: It Ain’t What You Do It’s the Way That You Do It … And That’s What Gets Results”, this volume, pp. 343 et seq. 4 http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 5 http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf.

332  Merger Control in European and Global Perspective on how merger reviews can be conducted, or should be conducted, if issued by international institutions in terms of transparency, case handling, communication, confidentiality and inter-agency coordination. If you try to categorize the existing best-practice rules, you can distinguish between the external best practices, which are issued for the benefit of merger control agencies by international institutions like the ICN, or between agencies. The second category is internal best practices, as I call them. That is, the best practices issued by the agencies themselves for the merging parties and for their own practice. The third category, which John covered at the end, is best practices issued by companies or by law firms. Let me concentrate on the two first categories. The external best practices for agencies are an important means to influence convergence. We have two categories, as I already said: firstly, the international organizations issue best practices. By far the most significant driver with regard to these best practices is the ICN, which seeks to facilitate cooperation and the exchange of know-how between competition authorities globally. The Merger Working Group published a comprehensive set of materials on best practices, which John already recommended that we should study. The most important contributions are the Guiding Principles for Merger Notification and Review6 and the Recommended Practices for Merger Notification and Review Procedures.7 Further important guidance for merger control agencies comes from the OECD. In 2005, the Council of the OECD adopted a Recommendation that touched fairness and cooperation.8 And then another document was issued by the Business and Industry Advisory Committee of the OECD; that committee published their Recommended Framework for Best Practices in International Merger Control Procedures in October 2001.9 The other important category of best practices consists of bilateral best practices, and by far the most important one is the Best Practices on Cooperation in Merger Investigations of the US-EU Merger Working Group.10 If you try to analyze the existing best practices, what is the scope? If you look at the external best practices, you see that, consistent with the rationale of external best practices, i.e., providing guidance for merger control authorities on how to shape their procedures, external best practices only establish general and broad principles for an efficient and transparent set up. These principles are usually verbalized as recommendations to agencies, therefore external best practices provide a model rather than a formal code. So you can conform to them but you are not obliged to do that. Procedural issues that are usually dealt with include recommendations on the structural alignment of notification thresholds with, for the sake of legal certainty and reduction of unnecessary transaction costs, a clear preference for revenue-based thresholds rather than market share http://www.internationalcompetitionnetwork.org/uploads/library/doc591.pdf. http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 8 OECD, Recommendation of the Council on Merger Review (2005), http://www.oecd.org/ dataoecd/3/41/40537528.pdf. In April 2012, the OECD’s Competition Committee published a report (based on the proceedings of Working Party No. 3) entitled Procedural Fairness and Transparency: Key Points. See http://www.oecd.org/dataoecd/16/34/50235955.pdf. 9 http://www.biac.org/statements/comp/BIAC-ICCMergerPaper.pdf. 10 http://www.ftc.gov/os/2011/10/111014eumerger.pdf. 6 7

Panel 5 

333  

thresholds. Moreover, the recommendations are pronounced regarding the agencies’ requirements on the timing of a notification, a clear determination of the triggering event and preferably no deadline, and the content of the notification. If you analyze the internal best practices which exist, frequently they deal with the details regarding the case team set up, the decision-making process, pre-filing guidance, communication issues – for example the timing of the Statement of Objections – guidance on the conduct of meetings, and access of third parties to such meetings. There is also relevant guidance on the right of access to the file, timing, accessible documents, entitled persons, procedures on the requirement for the initial notification, and safeguarding confidentiality with regard to the business secrets provided by the merging parties. The aims and benefits of merger best practices can be summarized very quickly. First of all, with regard to agencies, it is clear that merger control agencies benefit from the external best practices because they assist antitrust agencies with respect to a structuring of the relevant merger review procedures, therefore acting as a catalyst for better convergence. They also help to streamline the intra-agency procedures and therefore boost efficiency with respect to the merger review. The promotion of external best practices is desirable, but I think, as John said, there is still something to do. The increasing proliferation of merger control regimes and the variety of jurisdictions which affect globally active undertakings heighten the need for national agencies to publish more internal best practices on the conduct of their procedure. With regard to external best practices we have the ICN’s Recommended Practices,11 which is an excellent document that still can be improved, as John already explained. If one would try to define the minimum requirements of the internal best practices, I think it is clear that each internal best practice should deal at least with a minimum standardized set of issues which cover the most important aspects of merger control procedures. I think these are: pre-filing consultations, the scope of requests for information, the conduct of meetings, involvement of third parties, access to the file and confidentiality issues. To what extent are these best practices binding? An authority that issues such best practices normally should be bound to what they have pronounced; if there is a good reason, they may deviate from it. Another aspect that I want to mention is that in favour of the parties, or in favour of a better procedure, the binding effect should be limited. That especially applies, for example to access to file, if in appropriate cases access to file can be given earlier than you normally find in the best practices. So I would like to conclude with that and let me add that most of the proposals that John made, I would also have made. Thank you. Mario Siragusa: I will concentrate on what Jochen characterized as internal best practices, and in particular the Commission’s best practices. The Commission has made a substantial effort trying to codify and standardize its internal procedures, and there are in fact two groups of best practices: one specific to mergers,12 and 11 12

http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. Link above note 5.

334  Merger Control in European and Global Perspective a second, broader best practice for antitrust proceedings.13 On mergers, the 2004 best practices include, of course, a standard model for commitments and the trustee mandate. Then there are the more recent remedy notices,14 so that’s the block of best practices which pertain to mergers. We recently saw more best practices for antitrust proceedings published.15 There, we have of course the best practices for the submission of economic evidence16 and the guidance on the role of the Hearing Officer.17 So these are the two groups that I was talking about. It seems to me that all this documentation plays a very important and excellent role, and is inspired basically by enhancing the cooperation spirit of the relationship between the Commission and the companies involved in those various processes. That is clear, for instance, from the efforts that have been made to clarify the pre-notification process. What I find particularly important is the agreement that has been reached for the Commission to pursue fact-finding also in that process and even to talk to third parties, with the agreement of the merging entities. I think this is a very interesting development, which has clarified a very important portion of what happens during the pre-notification process. Also, I think these best practices have been inspired generally by an effort to be more transparent. The state of play meetings and the access to some of the key documents before the SO is sent all seems to me to be a step in the direction of more transparency. There have also been important decisions in the more general antitrust best practices, published more recently. In particular, I found this idea that there should be an earlier formal opening of the proceedings helpful, so that everybody knows that the proceeding is starting. Also, publications on the development of the proceeding, the issuing of the SO and the final decision, so that the opening of the procedure will be publicly announced so that everybody is informed about the progress of the proceedings. Now I would simply like to make a couple of remarks on ways in which to improve this, because I think everything can always be improved. First of all, a reflection on what the nature of all these documents is. I suppose that, particularly to the extent that they standardize proceedings that are applied by the Commission in individual cases, the Court will interpret those in the same way it has done with the Commission’s other guidelines. So we may see a development of case law 13 See now Commission Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU, 2011 OJ C308/6. 14 See Commission Notice on remedies acceptable under the Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, 2008 OJ C267/1. 15 See above note 13. 16 For the submission of economic evidence (in antitrust as well as merger cases), see http:// ec.europa.eu/competition/antitrust/legislation/best_practices_submission_en.pdf. 17  See http://ec.europa.eu/competition/consultations/2010_best_practices/hearing_officers.pdf. At the time of the 2010 Workshop proceedings, the terms of reference of the Hearing Officer were stipulated in Commission Decision on the terms of reference of hearing officers in certain competition proceedings, 2001 OJ L162/21. Those terms of reference were later expanded to some degree (e.g., during the investigative phase, or in the context of commitment proceedings under Article 9, etc.) by Decision 2011/695/EU of the President of the European Commission of 13 October 2011 on the function and terms of reference of the hearing officer in certain competition proceedings, 2011 OJ L275/29. For details, see Wouter Wils, “The Role of the Hearing Officer in Competition Proceedings before the European Commission”, 35 World Competition 431 (2012).

Panel 5 

335  

which says that a state of play meeting can take place and, if requested, must take place. So we can imagine this at least for certain aspects of these guidelines. I can imagine that there are other areas of the best practices which probably are more general and which may not be subject to judicial intervention. But this is probably something that we may see developing in the next few years. As to the possible improvements, I found that there are four areas in which I think some improvements can be made. More generally, on access to the file, I think that it should be possible to allow access from the beginning of the proceedings, both in merger cases and in antitrust cases. Once the procedure is open, and of course in the case of mergers with notification and in the case of antitrust proceedings with the formal opening of the proceedings, the file is being built. In some of our national procedures, parties have the right to access the file periodically to be informed as to the exact status of the file and the documents contained therein. This allows the party to be on a level playing field with the Commission, because when the Commission comes to know, for instance, what third parties are saying on the proposed merger, I do not see why the merging parties should not be put in exactly the same position as the Commission. This is even more important in antitrust proceedings, but it certainly could apply also to mergers. So I think that giving access to key documents before the SO is a good step in the right direction, but I propose as a next step to see whether it is possible to give full access to the file periodically during the course of the proceedings. A second suggestion relates to contacts with third parties. The Commission will of course meet and talk to third parties in the process of merger review, as in the process of antitrust proceedings. But here there is this distinction between formal and informal contacts, which I find troublesome. More generally, I think that all meetings that Commission officials have with parties – both parties to the proceedings and third parties – should somehow be recorded. There should be minutes of the meetings, summary minutes of what has been said, and this should be put in the file, with access for all interested parties. Again, this is done by some national authorities and I don’t see why this should not also be done at the Commission level. This is particularly if the results of those meetings somehow affect the Commission’s thinking and can be used by the Commission when it makes its findings. There should be some record of it in the file; otherwise there is no way it could be subject to review by the Court. The third area where improvements could be made concerns the role of the Hearing Officer. Now, I’m not sure if this can be done by way of best practices. Maybe this needs an ad hoc communication of the Commission. But certainly there were some expectations when the Commission announced the recent publication of the best practices that this could have been the occasion to somehow modify and enhance the role of the Hearing Officer. I don’t think that this has been done. The Hearing Officer’s role has been codified and explained, but I did not find anything new in the document. I’d be interested to hear whether you agree with this analysis or whether you think that there has been a change in that respect. I

336  Merger Control in European and Global Perspective certainly think that the Hearing Officer could play a more comprehensive role, not limited to procedure and access to file issues.18 My last comment is on an improvement which has been made and which probably relates more to general antitrust proceedings and not so much to merger review. It has been announced that there will be an early, official opening of the antitrust proceedings. Generally, what we have seen is that this opening is done with a very short communication, sometimes only one page or half a page as an indication of what the area is that the Commission is investigating. Not much information is given as to the scope and the real possible allegation or preliminary theory on which the opening of the case is based. To the extent that the Guidelines say that the opening of the procedure will come after an initial assessment of the case, maybe it could be useful to have a more clear indication of the nature and scope of the matter when the Commission announces the opening of the proceedings. So these are my comments on the specifics. Otherwise, I think that it is a very good initiative and I am sure that the Commission will continue to update its best practices because of the need for continual improvement. This is particularly because some of those issues covered by the best practices the Commission really have to do with due process and with the way in which the specific proceeding is going to be followed by the Commission. Thank you. Lowe: Thanks very much. John said it right at the beginning: you can’t have best practice unless you have promised to improve it regularly. For those of you who have not been to the Competition Workshop before, it has been a major driver of improvement on substantive issues and on procedural issues. The 2009 Workshop featured a barrage of criticism from some of the people around this table, but others as well, saying that the Commission didn’t do enough in this area.19 I think you can actually attribute the appearance of a document on best practice in antitrust proceedings and the submission of economic evidence and the organization of hearings precisely to the pressure you put on us on that occasion. This is just as, three or four years ago, I would say that thanks to you and to the Workshop, I was able to argue – with great difficulty – for a Guidance Paper on the application of Article 102 to unilateral conduct.20 So we have a clear indication of a cycle of debate and reaction. It’s not surprising that you don’t see a change in the precise role of the Hearing Officer, because it would have been rather odd for Neelie Kroes or myself to have left to Commissioner Almunia the test of having to do something radical. That’s up to him and his team to consider. Now, would anyone else like to comment on these best practices? 18 As seen the previous footnote, the Hearing Officer’s terms of reference were enhanced to some extent pursuant to Decision 2011/695/EU of the President of the Commission. 19 See Ehlermann and Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011. 20 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.

Panel 5 

337  

Sven Völcker: Just briefly on a couple of points. First, on Mario’s suggestion that best practices might become justiciable: I personally have my doubts, speaking under Judge Forwood’s control, of course, that the Court would ever get to that point. Looking at the past litigated merger cases, perhaps the procedural deficiencies have not been well pleaded, but they certainly have not succeeded – with the notable exception of Schneider,21 which may have been a special case. The Courts’ substantive review has been much more searching than perhaps its willingness to accept that there were certain flaws in the procedure here and there. I’m also not sure we should wish for that kind of review, because the moment it is clear that the Commission will be held to the best practices in every case by the Court, we will see a freezing of those practices and they will not evolve to the stage of being the very best. The other point is generally transparency, which I think is something the DOJ has been very outspoken on. In terms of access to file, there have been a lot of practical improvements that are very noticeable over the last year or two, down to things like an electronic index with hyperlinks to the document which saves you, in the very short time frame that you have to respond to a Statement of Objections, an enormous amount of time. I also think things like supplementary access to the file, really up until the point that the Commission needs to get close to taking its decision, are very helpful. I’m not sure we need to go to the process that the Bundeskartellamt and perhaps other national authorities have of really giving rolling access to the file. I think we all know, after having looked at these files, that 95% of it is pretty irrelevant and we would prefer in the first phase that the case team be fully focused on resolving the key issues of the case. Nevertheless, I think there is one thing that I’ve never really understood, which is that, in theory, if there are key documents in a case, you get access to them under the best practices the day after the opening of Phase II. But why do you get access to them only after the Commission has taken that decision to go into Phase II, which we all know has very significant consequences for the parties and the timing of the deal? If there is really a key document, whether it is a separate complaint or a request for information which has a 30-page response to the last question – “Do you have any concerns about this transaction?” – why not give access to that to the parties immediately? Certainly, in one case that I’ve had recently, I think that getting that information outside the framework of the best practices was extremely helpful for resolving a lot of technical issues in that case. In that vein, Rachel, I guess that the DOJ doesn’t have any best practices in terms of access to the file, because there is no file in that sense. However, I wonder whether giving access to key documents at a stage before the DOJ actually goes to court must also help the cause of transparency. Nicholas Forwood: So far as the legal status of best practice statements is concerned, there are obviously two ways that that could come before us. First of all, in the context of a plea of legitimate expectations, perhaps its limits are not quite 21

Case T-310/01, Schneider Electric SA v Commission [2002] ECR II-4071.

338  Merger Control in European and Global Perspective as broad as you might think they were, at least in relation to the sort of things that can be covered by it. What is becoming more important of course is the principle of good administration, which has now got support in the Charter of Fundamental Rights.22 Now, a person is entitled to have the administrative authorities deal with his case properly and effectively and so on. Clearly, statements as to best practice, which we have to remember is not necessarily exactly the same as good practice – best may be ideal, but good may be good enough – are clearly going to be relevant in that connection. One particular area where it may be relevant is in relation to the control of complex economic assessments in areas where there is, in some form or another, a marginal control. There, of course, as the Court reminds us in Tetra Laval,23 you verify the basic facts and so on. You also look to see, in the words of the General Confession of Sins, whether you have left undone those things that you ought to have done, or whether you have done those things you ought not to have done. In that connection, I can well see that if there is a statement in a best practice document that one ought to do X, Y, or Z, but that hasn’t been done in a particular case, then that may be the sort of thing that might be thrown in to undermine a complex assessment, to the extent that we would not otherwise be looking at it closely and forming our own view if indeed that be the case. Kirsten Edwards: In terms of submission of economic evidence, I really welcomed the best practice guidelines, but I think a lot of it is actually down to communication between the individual case team and the parties and their advisors. I’ve had very different experiences even since the best practice guidelines have been published in this respect. My best experience, which I think really demonstrated best practice on the side of the Commission and, not to be too modest, on our side too, was in Kraft/Cadbury,24 where we did a vast amount of work in Phase I. This was one of these cases where there was a merger simulation in Phase I, but the only reason we could do that much in Phase I was partly because we spent so much effort trying to pre-empt all of the questions that the case team and the Chief Economist’s Team would come up with, because we didn’t have time to go through a lengthy process of questioning. The case team was also very up front with us in that case, at every stage, on exactly what they had seen so far, what their views were at that time, and what more they needed to see to enable them to decide one way or another about it. Because of that, we did manage to get to a Phase I clearance with divestments in a way that everybody was happy with. 22 The right to good administration is provided for in Article 41 of the Charter, 2010, OJ C83/02. Article 41 provides, in its first three paragraphs, that “1. Every person has the right to have his or her affairs handled impartially, fairly and within a reasonable time by the institutions, bodies, offices and agencies of the Union. 2. This right includes: (a) the right of every person to be heard, before any individual measure which would affect him or her adversely is taken; (b) the right of every person to have access to his or her file, while respecting the legitimate interests of confidentiality and of professional and business secrecy; (c) the obligation of the administration to give reasons for its decisions. 3. Every person has the right to have the Union make good any damage caused by its institutions or by its servants in the performance of their duties, in accordance with the general principles common to the laws of the Member States.” 23 Case 13/03 P, Commission v Tetra Laval [2005] ECR I-1113. 24 M.5644, Kraft Foods/Cadbury (6 January 2010).

Panel 5 

339  

I think everyone was really proactive in that case, but I’ve been on cases where on our side we’ve just had no idea what the Commission was thinking at a particular point and I’m sure the Commission was thinking: what are these crazy economists going to put in front of them next? (laughter) So I think a large part of it is about communication, but I also think one of the challenges is actually talking to the parties themselves about best practice, because I think most people around this table and in the profession are in a lot of agreement about what best practice should be, particularly as regards economic evidence. However, when it comes to talking to your client and saying that they need to be a bit more here or there, until it comes out of the Commission’s mouth they are not willing to do anything. They are always willing to put their money where their mouth is and do as you advise, but often they think: “Why should we do this, why isn’t the Commission doing this, why is this on us?” So I think there is still a challenge in trying to get the companies who are being investigated to understand where we are coming from with the best practice. Lowe: A pre-condition for that is that the Commission has to know where it is coming from when it actually receives the original pre-notification, and starts to develop a theory or potential theories of harm and has an investigative technique which isn’t a fishing expedition. The opposite to that is when it has absolutely no theory of harm, thinks there is no problem, then it gets a key document which looks brilliant. Then it needs a non-confidential version to be able to share it with everyone else, and it’s two days before the deadline, and we can’t meet what Mario said is the right approach, which is to share key documents which are fundamental to opening Phase II. But Thomas is going to tell us about his own practical experiences. Thomas Deisenhofer: First of all, one general comment on due process, which is taken extremely seriously by Commissioner Almunia and Alexander Italianer.25 One counter-intuitive thing is that you think you can’t get enough of due process. The reality is that, depending on where you stand, there are trade-offs. There is first of all due process and speed, and effectiveness. Then there are tradeoffs between the due process of the one and the due process of the other, which is the most important thing. Confidentiality involves tradeoffs on all sides too. All these points that you have made are extremely delicate points where the difficulty is to find the right balance and to calibrate it correctly, so that you provide as much due process as you can while maintaining speed and while maintaining the rights of third parties. Depending on where you stand as a lawyer, obviously you will defend more this side than the other side, but due process is not an unlimited resource. The second thing is how seriously we take it in practice. It’s not just what is written in the rule that’s important, but what we do. I think Sven mentioned 25 See, e.g., Joaquín Almunia, “Fair process in EU competition enforcement”, SPEECH/11/396 of 30 May 2011; Alexander Italianer, “Safeguarding due process in antitrust proceedings”, Fordham speech, 23 September 2010.

340  Merger Control in European and Global Perspective key documents; it’s true that there is currently no rule that key documents before Phase I can be given, but as Philip said, on the one hand the case team has in reality 15 days from notification during which a lot of time is spent waiting for the replies to questions we try to send out the day the notification arrives. We send them out so that the companies have a bit of time to reply, then we try to coordinate inside. So there might be a document coming up, and then the internal coordination, and then on day 15 we have to speak to the parties. I think it’s right to tell the parties where we’re going, but if on top of that you then have to produce non-confidential versions of key documents, with the difficulties of getting them from third parties, and then manage possible remedies or whatever, that would be a challenge if it were to become mandatory. What is happening in practice, as you’ve rightly described, is that if it’s possible, and if the complainant has no problem with it, then directly or indirectly you can promptly identify relevant technical issues. Routinely, at least at my level, we try to give non-confidential versions of documents even in Phase I in order to sort out things because the ultimate objective is to get to the issues as soon as possible. My last point is on attitude. It’s all about attitude: attitude of the enforcer, and of the parties and third parties. If there is a give and take, and if people are reasonable, then solutions are normally found beyond the written best practices. Bill Kovacic: I have a response to Sven’s question posed to Rachel, and she can certainly address it herself as well. I think there’s a basic tradeoff inherent in the US system, that is that parties like the fact that we can’t do anything that hurts until we’re in front of a judge and a judge says so. That inhibits access to the file, and I think if we had the capacity to take the decision in the first instance it would be much more forthcoming. The tradeoff in the system design is that as long as we have an adversarial contest to resolve the matter, there is always going to be an inhibition to putting all of the evidentiary cards face up. If we had a process that resembled that of the Commission, if the original Hart-Scott vision had been in place, or if the FTC for example could take the steps that the European Commission could take, I imagine that we would have a similar bundle of disclosure safeguards ex ante, but the possibility of litigation makes it hard to be as revealing as we might be. Thank you. Rachel Brandenburger: All I would add is that, from my experience in private practice counselling transatlantically, although the procedures were different and one spent quite a lot of time explaining to clients why it worked this way and that way, I’m not aware of cases in front of either US agency where I felt that I did not understand what the theory of harm was, or what the evidence was in front of the US agencies, even if I did not have access to the file. If you have other experiences, then we need to hear about them. Howard Shelanski: You may remember the proposals that Charles James made in 2001–2002, where he said that the DOJ was willing to put things on a different

Panel 5 

341  

path if it had an earlier common revelation of information, especially in instances where a consensus might be achievable. I don’t know exactly what the experience with that turned out to be, but it was a commitment to put more information forward earlier. I hear continuing complaints as well, Sven, about whether, even without access to the file, we are forthcoming enough. I don’t know whether that’s because we don’t say enough on our side at the FTC, or whether we say it quite clearly but the parties don’t want to believe it. Cal Goldman: Just one or two quick additions. When I was running the Canadian Competition Bureau I consulted with my US colleagues about issuing news releases even when we took no-challenge decisions. I was cautioned that if we did that, then we would end up being sued. Well, we’ve come a long way. The Canadian and the US agencies now issue all kinds of information. That’s a best practice. Secondly, the US authorities, especially the FTC, should be commended for stakeholder consultations on an international scale. I suggest that what the FTC has done on global hearings, consulting the global community and not just stakeholders in the US, is a very good example of trying to achieve as much input through a form of best practice as is possible. It’s a model for other jurisdictions. Thirdly, I’d like to make a small point concerning the OECD. Peer review at the OECD does not generally include BIAC. I think it should. It would be constructive to consider including BIAC at least in the country reviews up until the point where there is any discussion of confidential enforcement actions. I don’t see any harm, I can see only good coming of it. Thank you. Lowe: Thank you very much. I think we’ll draw a line under that discussion of best practices. I just want to say one or two things in conclusion of this day and a half. In June of last year, I took an executive decision that we should discuss mergers. Not because I thought that there would be a huge amount of disagreement on many issues, but because it was an opportunity to discuss the new guidelines envisaged in the US and the UK. I think we did have a very stimulating discussion on the use of methodologies, including market definition issues and other investigative techniques. I didn’t notice a great deal of alarm, but certainly some caution about using those techniques in the right context, and recognizing too that virtually all our techniques need to complement each other to arrive at a robust result. On the standards of assessment and evidential requirements before the EU Courts, I think we’re still going to debate that quite a lot. At the moment I don’t see at least the Commission in Europe, or the EU Courts, actually moving on any particular issue of structure or procedure. Maybe that will become necessary if indeed we have another spate of Court activity with respect to certain cases, and there are some in the pipeline. On the procedural issues of convergence particularly, I thought that we would not expect to have a great deal of advance compared with the work already done in the ICN context and elsewhere, but I did see, in both yesterday’s session and this morning, continuing emphasis on evaluating performance and

342  Merger Control in European and Global Perspective indeed confronting ourselves with either the criticism of over-enforcement or under-enforcement. What that would mean in terms of ongoing work inside our agencies almost certainly needs clarification. We emphasized how important it was to build up a relationship of cooperation and confidence between those people who are actually going to tackle the same issues in the same markets. I did raise a small alarm bell, which I think will become larger, that the more our markets become global, the more pure competition will be mixed up, at least in the public perception and in the government’s perception, with other issues such resource scarcity, security of supply, and general concerns about concentration. These may take any agency, if put under pressure, beyond what it would regard as the normal standards of proof necessary to prohibit a merger. I’m just saying that this is an alarm bell, and it may be less loud if we are able to look more closely at the longer-term effects of further concentration in global markets. It comes back to the issue of what the strength of potential competition is outside any particular jurisdiction, and what the serious assumptions being made by one jurisdiction are in taking a decision in relation to the effects in its own jurisdiction. There, thank goodness that there is the Bundeskartellamt to look at BHP Billiton. With the Commission, obviously. But, without some constructive cooperation between the agencies in relation to transactions like that, I think that there would be a serious issue of government failure of addressing major transactions, which have maximum impact in practically all jurisdictions. On the last issues of best practice, we can only get better. Looking ahead, Mel and I will be thinking about next year’s subject, and we’ll undoubtedly choose a more controversial one where we can scratch each other’s eyes out on procedural issues or inter-institutional issues. I thank you very much, particularly those who have come a long way to be with us to share in this discussion. As I say, the Workshop hasn’t been as bloody as some of the previous ones. Last year, when we discussed due process there were far greater, far more powerful epithets used than anything in the last 36 hours, but those lively discussions did lead us to interesting conclusions.26 We can say that there’s a lot of progress being made, but there are some alarm bells ringing and we shouldn’t be complacent. I thank you again for your participation, and as always we will try to set an agenda next year to match the standards set by Claus Ehlermann and all of you in times gone by. Thank you (applause).

26

See Ehlermann and Marquis, Evaluation of Evidence and its Judicial Review, cited above note 19.

John Boyce1

Best Practice in Merger Control: It Ain’t What You Do It’s the Way That You Do It … And That’s What Gets Results

What is meant by “best practice”? The notion that merger control agencies should perform their functions in a way that conforms to “best practice” is hardly controversial. It is a laudable aspiration, just as the scouting movement worldwide expects its members to promise to do their best. It does though beg the question of what is meant by “best practice”, a concept used in society more generally – whether in the medical profession, the world of computing technology, or the wider business community. Some inspiration can be drawn from the following dictionary definitions, which also illustrate that the term’s usage is relatively new (indeed, it was only during the last decade that competition agencies began to use it) and that there is a risk of the term being overused or misused. Oxford English Dictionary Best practice: n. chiefly Business (as a mass noun) the practice which is accepted by consensus or prescribed by regulation as correct; the preferred or most appropriate style. 1984 C. HITCHING & D. STONE Understand Accounting! xvi. 315 Some of these have long been regarded as ‘best practice’, and most have already been defined within the accounting standards. 1993 J. KAY Found. Corporate Success VII. xxi. 350 To look at what other firms do, and copy it. This strategy is more felicitously expressed as adopting best practice. 2005 J. DIAMOND Collapse (2006) xv. 474 The essence of FSC certification is that consumers can believe it, because it is... the result of an examination, against internationally accepted standards of best practice.

BusinessDictionary.com Best practice: Methods and techniques that have consistently shown results superior than those achieved with other means, and which are used as benchmarks to strive for. There is, however, no practice that is best for everyone or in every situation, and no best practice remains best for very long as people keep on finding better ways of doing things.

1

Partner, Slaughter and May, Brussels.

344  Merger Control in European and Global Perspective Wikipedia A best practice is a technique, method, process, activity, incentive, or reward which conventional wisdom regards as more effective at delivering a particular outcome than any other technique, method, process, etc. when applied to a particular condition or circumstance. The idea is that with proper processes, checks, and testing, a desired outcome can be delivered with fewer problems and unforeseen complications. Best practices can also be defined as the most efficient (least amount of effort) and effective (best results) way of accomplishing a task, based on repeatable procedures that have proven themselves over time for large numbers of people. A given best practice is only applicable to particular condition or circumstance and may have to be modified or adapted for similar circumstances. In addition, a “best” practice can evolve to become better as improvements are discovered. Despite the need to improve on processes as the environment changes, best-practice is considered by some as a business buzzword used to describe the process of developing and following a standard way of doing things that multiple organizations can use for management, policy, and especially software systems. As the term has become more popular, some organizations have begun using the term “best practices” to refer to what are in fact merely ‘rules’, causing a linguistic drift in which a new term such as “good ideas” is needed to refer to what would previously have been called “best practices.”

Why is best practice a good idea for merger control rules? Each time a competition agency applies its merger control rules to a specific transaction, it needs to interact with the businesses subject to review (and their advisers) and also with any other interested third parties which may be affected by the transaction (customers, competitors, etc.). When reviewing multijurisdictional mergers, agencies may also interact with other competition agencies – which are in turn performing similar functions, liaising with the same merging parties (albeit possibly with different individuals within the organisations or with different advisers) and with the same or often different interested third parties. Successful management of these various relationships is a key to ensuring the most efficient review process (which includes minimizing the hurdles placed on the business community and avoiding unnecessary duplication) and the most effective results (in terms of approving transactions within a timeframe that is compatible with the parties’ reasonable commercial expectations and objectives, with focused remedies where required). These challenges can multiply in multijurisdictional cases. Where major conflicts or problems arise, this is often due to a lack of understanding between the various participants of the roles played by others in the process and the constraints under which they are operating. For example, businesses sometimes fail to understand the objectives and purpose of the different

John Boyce 

345  

competition agencies’ reviews, whereas agencies sometimes fail to appreciate the deadlines and pressures faced by the businesses concerned. In the context of merger control investigations (particularly of cross-border mergers subject to multiple reviews), promoting “best practice” is in the interest of merging parties, the competition agencies and society as a whole (including consumers and taxpayers). It can help expedite the implementation of transactions which are compatible with the competition rules, as well as the identification of any issues giving rise to competition concerns which can be addressed by appropriate remedies or changes. The accomplishment of these tasks can be facilitated by effective and pragmatic cooperation amongst the various participants – within frameworks which are broadly transparent and encourage efficient coordinated reviews, without duplication of effort on the part of the businesses and agencies involved. While formalized best practice guidelines do play a useful role in guiding participants in how to conduct themselves to achieve these objectives, written guidelines must be applied purposefully and flexibly, in light of the fundamental objectives best practice aims to achieve, rather than being applied like a rulebook. Formal best practice guidelines should be seen as means to achieve an end – with the fundamental objective of instilling and maintaining a culture of best practice within and among competition authorities (and with businesses participating in and affected by the review of merger transactions). If, on the facts of a particular transaction (in a particular sector, at a particular time), there are more efficient and effective ways for the parties and agencies concerned to accomplish their tasks, best practice mechanisms should allow the parties and agencies to adapt to those circumstances – provided that the rights of all the various participants are respected. The rest of this Paper identifies specific aspects of the merger review process which can benefit from closer coordination and continued harmonization of best practice standards (in particular for multijurisdictional mergers), doing so by reference to different stages of the process. The observations below draw from 2 experience of existing best practices guidelines.

Best practice to facilitate preparatory work by parties (prior to contact with competition agencies) Merging parties and their advisers have an important role to play in preparing the groundwork for a smooth review process. The competition agencies will generally not be involved in those commercial deliberations – when parties may be evaluating all sorts of strategic and day-to-day options for developing their business activities, including possible mergers and acquisitions. Nevertheless, it is important for competition agencies to understand the realities of commercial life 2 Annex 1 identifies a number of existing best practice guidelines applicable to merger control reviews.

346  Merger Control in European and Global Perspective in which companies are operating and which form the backdrop for those merger deals which do ultimately see the light of day. Indeed, at one extreme there are deal proposals which may be considered in detail but which never come to fruition because the parties reach the conclusion that the “antitrust price” (in terms of the level of remedies that would likely be required) destroys the economic rationale for the deal. There will also be possible deals where the “regulatory cost” – in terms of the uncertainties, delays, legal and other fees and filing costs that would be imposed by multiple reviews – may be sufficient to rule out proceeding with the transaction, even if ultimately it might have been cleared unconditionally by all competition agencies. Best practice considerations mean that parties should give due consideration to these risks before announcing or agreeing to a specific transaction. That said, commercial imperatives and confidentiality concerns are such that the extent to which this can be done varies considerably from case to case. Competition agencies around the world can facilitate this risk assessment process for businesses (large and small) by increased transparency of their merger control rules and convergence towards internationally accepted standards of best practice. Such transparency assists the business community. It improves awareness and understanding of different countries’ jurisdictional thresholds (which affect the number and cost of merger filings that may be required for the deal) and of the procedural steps that need to be followed (which can have an impact on the deal timetable). It also helps international businesses to understand how the different agencies might apply their substantive tests to deals they may be considering (which may have an impact on whether the deal can be implemented as planned). While the ICN has made valiant efforts to encourage greater transparency 3 and convergence by its members in these areas, much remains to be done. For example: • Although the ICN adopted Recommended Practices for the design and operation of merger control regimes in 2005, few authorities’ regimes are fully compliant. In this regard, some ICN members could play more of an advocacy role in encouraging their domestic legislators to bring national merger control regimes more into line with these internationally accepted standards of best practice. Indeed, there are some areas where agencies could make efforts to conform to these recommended ICN standards without needing legislative changes. In this regard, in particular for “foreign mergers” with no appreciable effects on competition in the jurisdiction concerned, the agency could exercise self-restraint by refraining from requiring formal prenotification (or from imposing fines for failure to notify) on the basis that the deal would most likely not have been notifiable if its domestic legislation were revised to comply with the ICN’s Recommended Practices. 3 The ICN Recommended Practices (summarized in Annex II) encourage legal certainty and increased transparency in these areas, in particular in RP1, RP2, RP3, RP4 and RP8. RP13 also highlights the importance of regular review to ensure that regimes develop in line with internationally accepted standards of best practice.

John Boyce 

347  

• Many ICN jurisdictions still do not include up-to-date and easy-to-follow summaries of their merger control rules on the ICN website. It ought to be relatively straightforward for each ICN member to produce such a simple summary using a standard ICN template. This could usefully be accompanied by a brief selfassessment by each agency of whether its current rules comply with each of the ICN’s 13 Recommended Practices. A process of “peer review” could then help ensure consistency in self-assessment, and could lead to greater convergence between agencies towards having merger control systems which are enforced fairly, consistently and in a manner that complies with ICN best practice.

Best practice for the conduct of investigations by the competition agencies In the interest of efficient enforcement, best practice generally seeks to reduce unnecessary or unreasonable burdens and costs on business (in particular in cases which do not raise significant substantive issues in the jurisdiction concerned). Accordingly, the ICN Recommended Practices (and/or some of the other existing 4 guidelines) recognize that: • Initial notification requirements should be limited to information needed to verify that a transaction exceeds jurisdictional thresholds, to determine whether it raises competitive issues meriting further investigation, and to take steps necessary to terminate the review of transactions that do not merit further investigation. • Agencies should avoid imposing unnecessary burdens on parties to transactions that do not present material competitive concerns (or on third party addressees of questionnaires). Agencies should be sensitive to the pressures faced by businesses and be prepared to be flexible regarding some of their formal requirements for notification. • Agencies should limit translation requirements and formal authentication burdens; cooperation between agencies should help facilitate this process. • Merger filing fees should be discouraged or kept at levels which reflect the amount of work undertaken by the agency; they should not be used as an indirect tax on merger activity (particularly as regards deals between “foreign” companies). It is widely accepted that best practice involves agencies’ affording procedural fairness to merging parties and to third parties with a legitimate interest in the merger under review. Competition agencies should manage the review process to 5 ensure that it is implemented fairly, efficiently, and consistently. 4 5

ICN RP5. ICN RP7.

348  Merger Control in European and Global Perspective Enforcement priorities mean that each agency should focus its greatest efforts on cases raising substantive issues in the jurisdiction concerned. Guidelines 6 generally encourage the following: • Investigation procedures should include opportunities for meetings or discussions between the competition agency and the merging parties at key points in the investigation, and parties should be kept informed of the progress of their case on a regular basis. • Where feasible and appropriate, parties should coordinate at an early stage with agencies regarding the scope of any detailed economic analysis that may be required. • Prior to a final adverse enforcement decision on the merits, merging parties should be provided with sufficient and timely information on the facts and competitive concerns forming the basis for the proposed adverse decision and should have a meaningful opportunity to respond to such concerns. • Third parties should be allowed to express their views during the merger review process, and notifying parties should be able to review and comment on any substantiated submissions from third parties running counter to the notifying parties’ own contentions. There has been some controversy regarding the way in which reviews should be conducted when two or more agencies are reviewing similar issues raised by the same transaction. The ICN Recorded Practices emphasise the importance of inter7 agency coordination. In this respect: • Merging parties have a role to play in facilitating inter-agency coordination, in particular through the timing of notifications and the provision of voluntary waivers allowing agencies to share confidential information with each other. Agencies can encourage the provision of such waivers by allaying fears of unauthorized disclosure, e.g. by providing credibly secure physical storage for physical information and introducing measures to ensure the accountability of individuals with access to such information. • Agencies should seek to coordinate – both formally and informally – their reviews of mergers which may raise competitive issues of common concern. For instance, representatives of one authority might attend certain key events 8 in another’s investigation process. Coordination can be beneficial with respect to information requests, analysis, timing, remedies and peer review. • However, it is important for agencies not to get over-obsessed with being treated “equally”, particularly in cases which may be subject to review by many different agencies. Commercial pressures (to ensure that deals can be approved within an acceptable timeframe) and the different procedural timetables that apply in different jurisdictions mean that some agencies should be able to start their investigations before others (and conclude those ICN RP6. In particular at ICN RP10. 8 As sometimes occurs within EU-US best practice. 6 7

John Boyce 

349  

investigations before others have reached their key decision points). Dealing in an orderly fashion with different jurisdictions (which may have different timelines, as well as different perspectives of whether the deal raises problems in their territories) is a legitimate business objective for the merging parties. There have been a few exceptional cases where agencies have had concerns that the parties were “gaming the system” and playing off one agency against another; however, this would be a high risk strategy for the parties, and one which should not be viewed as “best practice”. Such a strategy risks antagonizing agencies that are generally well-placed to coordinate their investigations (formal and informal); they also have the ability to “punish” such behaviour by launching extended investigations or – if there are genuine competition concerns – by concluding their investigations with demands or findings which are different from, or even incompatible with, those already reached elsewhere.

Best practice for the development of remedies It is also widely accepted that agencies and parties alike can benefit from clear 9 procedures for developing remedies tailored to particular cases. In this regard: • Meetings between the competition agencies and parties at key stages of the investigation should serve as a forum to discuss possible remedy proposals. The competition agency should provide guidance on the general appropriateness of draft remedies proposals prior to their formal submission. • Procedures and practices should be established to ensure that remedies are effective and easily administrable. Appropriate means should be provided to ensure implementation, monitoring of compliance, and enforcement of the remedy. • Reviewing agencies should seek remedies tailored to cure domestic competitive concerns and endeavour to avoid inconsistency with remedies in other reviewing jurisdictions.

9

ICN RP11.

350  Merger Control in European and Global Perspective

Annex 1: Existing Best Practice Guidelines for Merger Review Cases NB The various best practice guidelines listed at Part A below focus on existing guidance on how agencies should go about conducting merger investigations efficiently and effectively. They do not cover guidelines issued by some agencies on how their national jurisdictional rules are to be interpreted or on procedural requirements that need to be followed. Although not covered by this Paper, Part B lists some of the various other guidelines issued by agencies on how to apply the relevant substantive tests under the merger control rules (e.g. applicable SLC, SIEC or dominance tests). Part A: Best Practice Guidelines for Merger Reviews Authority/Organisation Title (reference)

Date

Merger Streamlining Group

Best Practices for the review of international mergers (Discussion Draft) (http://www.globalcompetitionforum.org/ regions/n_america/canada/MergerSep21. pdf?id=323)

2001

Business and Advisory Committee to the OECD and the International Chamber of Commerce

Recommended Framework for best 2001 practices in international merger control procedures (http://www.biac.org/statements/ comp/BIAC-ICCMergerPaper.pdf)

ICN

Recommended Practices for Merger Notification Procedures (http://www. internationalcompetitionnetwork.org/ uploads/library/doc588.pdf)

Original comments 2002, last updated 2005

US Federal Trade Commission

Statement of the Federal Trade Commission’s Bureau of Competition on Guidelines for Merger Investigations (http://www.ftc.gov/os/2002/12/ bcguidelines021211.htm)

2002

Statement of the Federal Trade Commission’s Bureau of Competition on Negotiating Merger Remedies (http://www.ftc.gov/bc/bestpractices/ bestpractices030401.shtm)

2003

Best Practices for Data, and Economics and Financial Analyses in Antitrust Investigations (http://www.ftc.gov/be/ bestpractices.shtm)

John Boyce  Authority/Organisation Title (reference)

Date

US-EU Merger Working Group

Best Practices on cooperation in merger investigations (http://ec.europa.eu/ competition/mergers/legislation/eu_us.pdf)

2002

European Commission – DG Competition

Best Practices on the conduct of EC merger proceedings (http://ec.europa. eu/competition/mergers/legislation/ proceedings.pdf)

2004

Best Practice Guidelines on Divestiture Commitments (http://ec.europa.eu/ competition/mergers/legislation/note. pdf; http://ec.europa.eu/competition/ mergers/legislation/commitments.pdf; http://ec.europa.eu/competition/mergers/ legislation/trustee_mandate.pdf)

2003

Best Practices for the submission of economic evidence and data collection in cases concerning the application of Articles 101 and 102 TFEU and in merger cases (http://ec.europa.eu/competition/ consultations/2010_best_practices/best_ practice_submissions.pdf)

2010

OECD

Recommendation of the Council on Merger Review (http://www.oecd.org/ dataoecd/3/41/40537528.pdf)

2005

UK Office of Fair Trading

Good practice in the design and presentation 2010 of consumer survey evidence in merger inquiries - consultation (http://www.oft.gov. uk/shared_oft/consultations/OFT1230con. pdfhttp://www.oft.gov.uk/shared_oft/ mergers/642749/OFT1254.pdf)

UK Competition Commission

Suggested best practice for submissions of technical economic analysis from parties to the Competition Commission (http://www. competition-commission.org.uk/rep_pub/ corporate_documents/corporate_policies/ best_practice.pdf)

2009

German Bundeskartellamt

Best practices for expert economic opinions (http://www.bundeskartellamt. de/wEnglisch/download/pdf/Merkblaetter/ Bekanntmachung_Standards_Englisch_ final.pdf)

2010

351  

352  Merger Control in European and Global Perspective Part B: Guidelines for Substantive Merger Analysis Authority/ Organisation

Title (reference)

Date

ICN

Recommended Practices for Merger Analysis (http://www. internationalcompetitionnetwork.org/ uploads/library/doc316.pdf)

Original comments 2008, last updated 2010

European Commission – DG Competition

Guidelines on the assessment of horizontal 2004 mergers under the Council Regulation on the control of concentrations between undertakings (http://eur-lex.europa.eu/ LexUriServ/LexUriServ.do?uri=OJ:C:20 04:031:0005:0018:EN:PDFhttp://eur-lex. europa.eu/LexUriServ/LexUriServ.do?uri= OJ:C:2008:265:0006:0025:EN:PDF) Guidelines on the assessment of 2010 non-horizontal mergers under the Council Regulation on the control of concentrations between undertakings (http://eur-lex.europa.eu/LexUriServ/ LexUriServ.do?uri=OJ:C:2008:265:0006:0 025:EN:PDF)

UK Competition Commission and Office of Fair Trading

Merger Assessment Guidelines (http://www.oft.gov.uk/shared_oft/ mergers/642749/OFT1254.pdf)

2010

Australian Competition & Consumer Commission

Merger guidelines (http://www.accc.gov. au/content/item.phtml?itemId=809866&n odeId=7cfe08f3df2fe6090df7b6239c47d 063&fn=Merger%20guidelines%202008. pdf)

2008

Canadian Competition Bureau

Merger Enforcement Guidelines (http:// www.competitionbureau.gc.ca/eic/site/ cb-bc.nsf/vwapj/2004%20MEGs.Final. pdf/$file/2004%20MEGs.Final.pdf)

2004

Annex 2: Icn Recommended Practices (last updated June 2005) The ICN has adopted 13 Recommended Practices for the design and operation of merger review regimes. These recommendations are not binding; indeed, many members of the ICN are from jurisdictions with rules which do not conform (in varying degrees) to the recommendations. Nevertheless these recommendations do provide an indication of what the international community generally believes should be key elements of a sensible merger control regime.

John Boyce 

353  

1. Nexus to Reviewing Jurisdiction A. Jurisdiction should only be asserted if transaction has nexus with jurisdiction concerned. B. Merger notification thresholds should incorporate appropriate standards of materiality as to the level of local nexus required for notification. C. Determination of a transaction’s nexus to the jurisdiction should be based on activity within that jurisdiction, as measured by reference to: – activities of at least two parties to the transaction in the local territory; and/or – activities of the acquired business in the local territory. 2. Clear and Objective Notification Thresholds Thresholds should be: A. Clear and understandable; B. Based on objectively quantifiable criteria; C. Based on information that is readily accessible to the merging parties. 3. Reasonable Rules for Timing of Notification A. Parties should be permitted to notify proposed mergers upon certification of a good faith intent to consummate the proposed transaction. B. Suspensive jurisdictions (which prohibit closing pending review by agency or for a specified time period following notification): Should not impose deadlines for merger notification. C. Non-suspensive jurisdictions (which do not prohibit closing pending review by agency): Should allow parties reasonable time to notify following clearly defined triggering event. 4. Reasonable Rules for Review Periods A. Merger reviews should be completed within a reasonable time period. B. Merger review systems should include simplified procedure, providing for expedited review and clearance of notified transactions that do not raise material competitive concerns. C. Suspensive jurisdictions: Initial waiting periods should expire within specified period of notification, and any extended waiting periods should expire within determinable timeframe. D. Non-suspensive jurisdictions: Should complete initial review within specified period of notification, and any extended reviews within determinable timeframe. E. Jurisdictions should adopt appropriately tailored procedures to accommodate particular circumstances associated with non-consensual transactions and sales in bankruptcy. 5. Reasonable Requirements for Initial Notification A. Initial notification requirements should be limited to information needed to verify that a transaction exceeds jurisdictional thresholds, to determine whether it raises competitive issues meriting further investigation, and to take steps necessary to terminate review of transactions that do not merit further investigation.

354  Merger Control in European and Global Perspective B. Initial notification requirements and/or practices should be implemented so as to avoid imposing unnecessary burdens on parties to transactions that do not present material competitive concerns. C. Competition agencies should provide for the possibility of pre-notification guidance to parties on the notifiability of the transaction and the content of the intended notification. D. Jurisdictions should limit translation requirements and formal authentication burdens. 6. Fair Procedures for Conduct of Merger Investigations A. Investigations should be conducted in a manner that promotes an effective, efficient, transparent and predicable merger review process. B. Investigation procedures should include opportunities for meetings or discussions between the competition agency and the merging parties at key points in the investigation. C. Parties should be advised not later than the beginning of a second-stage inquiry why the competition agency did not clear the transaction within the initial review period. D. Where investigation periods are not subject to definitive deadlines, procedures should be adopted to ensure that the investigation is completed without undue delay. E. Competition agencies should seek to avoid imposing unnecessary or unreasonable costs and burdens on merging parties and third parties in connection with merger investigations. F. Merger investigations should be conducted with due regard for applicable legal privileges and related confidentiality doctrines. 7. Procedural Fairness A. Procedural fairness should be afforded to merging parties and third parties with a legitimate interest in the merger under review. B. Prior to a final adverse enforcement decision on the merits, merging parties should be provided with sufficient and timely information on the facts and competitive concerns forming the basis for the proposed adverse decision and should have a meaningful opportunity to respond to such concerns. C. Third parties should be allowed to express their views during the merger review process. D. The competition agency should manage the review process to ensure that it is implemented fairly, efficiently, and consistently. E. Merger review systems should provide an opportunity for timely review by a separate adjudicative body of a competition agency’s final adverse decision on the merits. 8. Transparency A. Merger control laws should be applied with a high level of transparency, subject to the appropriate protection of confidential information.

John Boyce 

355  

B. Merger control regimes should be transparent with respect to, at least: – the jurisdictional scope of the merger control law; – the competition agency’s decision-making procedures; and – principles and criteria the agency uses to apply substantive review standard. C. Competition agencies should promote transparency, by making information about the current state of merger control law, policy, and practice readily available to the public. 9. Confidentiality A. Business secrets and other confidential information received from the parties and third parties should be subject to appropriate confidentiality protections. B. Competition agencies should promote transparency of the confidentiality laws, policies, and practices applicable to their merger control procedures. C. Competition agencies should defer contacts with third parties until the proposed transaction becomes public where such deferral would not adversely affect the agency’s ability to conduct its investigation effectively or complete its review within applicable deadlines. D. Confidentiality rules should strike an appropriate balance between protecting the confidentiality of third-party submissions and procedural fairness considerations. E. Competition agencies should avoid unnecessary public disclosure of confidential information in public announcements, court or administrative proceedings, decisions, and other communications regarding a pending transaction. 10. Inter-agency Coordination A. Competition agencies should seek to coordinate their reviews of mergers that may raise competitive issues of common concern. B. Inter-agency coordination should be conducted in accordance with applicable laws and other legal instruments and doctrines. C. Inter-agency coordination should be tailored to the particular transaction under review and the needs of the competition agencies conducting the merger investigations. D. Competition agencies should encourage and facilitate the merging parties’ cooperation in the coordination process. E. Reviewing agencies should seek remedies tailored to cure domestic competitive concerns and endeavour to avoid inconsistency with remedies in other reviewing jurisdictions. 11. Remedies A. Remedies should address identified competitive harm arising from proposed transaction. B. Review system should provide a transparent framework for proposal, discussion and adoption of remedies. C. Procedures and practices should be established to ensure that remedies are effective and easily administrable. D. Appropriate means should be provided to ensure implementation, monitoring of compliance, and enforcement of remedies.

356  Merger Control in European and Global Perspective 12. Competition Agency Powers Competition agencies should have: A. Authority and tools needed for effective enforcement of their merger review laws; B. Sufficient staffing and expertise to discharge enforcement responsibilities effectively; C. Sufficient independence for objective application and enforcement of merger review laws. 13. Periodic Review of Merger Control Provisions A. Jurisdictions should periodically review their merger control provisions to seek continual improvement in the merger review process. B. Jurisdictions should consider reforms to their merger control laws and procedures that promote convergence towards recognised best practices.

Annex 3: Suggested Best Practice for Companies and Advisers on Transactions Subject to Multijurisdictional Merger Reviews Parties to international/cross-border deals may have to obtain regulatory consents or approvals in a number of countries. When contemplating or advising on such multijurisdictional deals, companies and their advisers may benefit from following a number of common-sense general principles: • Familiarize yourself with rules in key jurisdictions and those with outof-the-ordinary rules: While no one can be expected to know the details of all the different merger control rules, companies and advisers who regularly deal with multijurisdictional deals should familiarize themselves 10 with the broad legislative framework in key jurisdictions. This should put the principal advisers and their clients in a better position to develop a coordinated strategy to obtain all necessary clearances in a timely fashion and to liaise with local advisers in the jurisdictions concerned if appropriate. For these purposes, advisers should be aware of certain recurring problem jurisdictions which have merger control rules which are not in line with 11 international best practice. • Clarify responsibilities: Establish at the earliest opportunity which individuals (within the acquiring and selling companies or their principal legal or other advisers) will be responsible for gathering information and co-ordinating regulatory filings for the transaction in contemplation – 10 Law firms which regularly deal with multijurisdictional mergers should generally maintain up-todate internal guidance on the basic merger control rules in jurisdictions affected by deals on which they advise. There are also various published guides which seek to summarise this information. 11 As seen above, the ICN’s Merger Review Working Group has adopted 13 “Recommended Practices” for the design and operation of merger review regimes. Last revised in June 2005, these Recommended Practices are summarized in Annex 2.

John Boyce 

357  

competition, foreign investment and otherwise. This should assist in keeping down costs and avoiding unnecessary delays. The potential consequences of not notifying a deal in a timely fashion to a relevant agency (risk of unenforceability, potential fines, subsequent divestment, etc.) will differ from case to case and depending on whether the party concerned is the purchaser, vendor or target. Pressures to notify in different jurisdictions will also depend on the deal structure (e.g. public bid, private auction or sale, joint venture), risk of complaints, timetable for signing/closing, governing law and the companies’ general reputations in the jurisdictions concerned. • Be sensible and pragmatic: As a starting point for multijurisdictional merger control analysis, the companies involved and their principal advisers should generally prepare a brief overview of what the parties to the merger do (i.e. what goods or services their activities relate to) and where they do it (i.e. in which countries they manufacture and/or market their goods or supply their 12 services). There is little point in immediately involving lawyers in every jurisdiction where the parties make sales, as this risks involving people who are unfamiliar with the deal/markets and/or who are not experienced in competition law/practice. (Such a “scatter gun” approach can backfire, resulting in unfocused advice on local laws, unnecessary delays and expense.) Where appropriate, transaction-specific local advice should be sought. For key jurisdictions, focused local law advice should help to clarify whether a notification is necessary/advisable for the particular transaction (and whether it can be made after, rather than before, closing). • Distinguish competition rules from foreign investment rules and special sector regulation: Some countries have rules restricting or controlling foreign investment. Many countries also have special rules which apply to companies active in certain sectors, e.g. utilities (water, gas, electricity, post, telecommunications), transport (aviation, rail), media (TV, radio, newspapers), culture (art, music), financial services (banks, insurance), health (pharmaceuticals, hospitals), or defence. Where a deal involves the acquisition of local assets or shares in countries with foreign investment or special sectoral regulations, it will be important to observe those rules. As regards competition-based merger control rules, however, the competition agencies in most countries worldwide (notably those participating in the ICN) increasingly apply their domestic merger control rules with comparable standards and in a broadly consistent way, i.e. the emphasis tends to be on ensuring mergers do not substantially lessen competition (“SLC”), significantly impede effective competition (“SIEC”) or result in or increase a situation of market “dominance” (although some jurisdictions apply vaguer tests such as whether there may be harm to the “public interest”). Companies and their advisers should always bear 12 Even if the transaction might technically qualify for a large number of filings, advisers’ initial requests for data should not burden the companies’ business operations with onerous requests for local market share information. Initially, advisers should only need basic turnover and manufacturing data to help identify which jurisdictions may be significant or material. Precisely what figure may be appropriate for such materiality purposes will depend on the facts of the deal.

358  Merger Control in European and Global Perspective









in mind whether or not the particular proposed deal may have such negative substantive effects in any of the different jurisdictions concerned. Bear in mind substantive effects: Most competition agencies are less interested in deals which do not have appreciable competitive effects in their jurisdictions. If a deal has no appreciable effects in (or nexus with) a particular country, it probably ought not to be notifiable there. Some countries’ rules expressly recognize this effects doctrine. Even where they do not, principles of international comity as well as enforcement priorities may mean that they will not rigorously enforce their rules if a deal is not formally notified to them (in circumstances where, from their perspective, it is a “foreign” merger with no appreciable effects on competition in their 13 jurisdiction). Identify nexus to possible reviewing jurisdictions: In identifying which merger control jurisdictions are relevant for the transaction, companies and their advisers should generally focus their main efforts on countries where the involved parties have material activities, in particular production facilities or substantial sales. If interests to be acquired include direct control of legal entities/manufacturing facilities in the jurisdiction concerned, failure to comply with mandatory notification requirements could affect the transfer of legal ownership. This is less likely to be a significant risk, however, if the parties’ only activities involve export sales to the country (e.g. via a third party distributor). Needless to say, the purchaser should usually be more concerned about valid transfer of legal ownership than the vendor or target company. Identify key jurisdictions: Often an international deal will involve at least one or two obvious key jurisdictions – e.g. the parties’ “home” jurisdictions and/or countries with a developed system of merger control where prima facie the deal raises some competition issues. Generally, parties should focus on developing materials for these key jurisdictions (including in prenotification discussions with the agencies if appropriate); those materials and notifications can then provide a useful template for notifications in other jurisdictions. Parties should also bear in mind that: agencies in other jurisdictions may look to the agencies in the key jurisdictions for guidance on how to treat a deal, e.g. through the formal structures of the ECN (European Competition Network) in the EU, or informal contacts 14 between members of the ICN (International Competition Network). In some cases, the agencies may request “waivers” to facilitate these contracts (and it will generally be in the interest of the parties to offer such waivers to help facilitate an orderly multijurisdictional review process);

13 Thus, the ICN Recommended Practices (see Annex 2) provide that authorities should only assert jurisdiction over transactions which have an appropriate nexus with the jurisdiction concerned (e.g. material sales or assets within the territory on the part of the acquired party, such that the deal potentially has a material effect on the local economy). 14 In the context of cases handled by national competition authorities (NCAs) in the EU, there is the possibility of such cases being referred up to the European Commission (under Article 22 EUMR) if the parties have not themselves opted to refer the case to the Commission (under Article 4(5) EUMR).

John Boyce 

359  

• the coordination and timing of formal notifications can be very important. For example, it may be sensible to delay formal filings in some jurisdictions until after filings have been made in the key jurisdictions (although filing deadlines in some jurisdictions may limit the scope for prioritizing filings in this way); a clearance in a key jurisdiction can then provide other agencies with a useful precedent or “template” for their own clearance decisions. Alternatively, sometimes there will be scope for pressing ahead with some non-key jurisdictions before filing elsewhere, so as to ensure that those clearances come in early on, rather than holding up closing of the deal later in the process. Establishing an appropriate timetable for the different filing needs to be decided on a case-by-case basis.

Jochen Burrichter* and Manuel Zandt**

Merger Control and Best Practice

Undertakings face an increasing number of merger control regimes in merger cases. Following a significant trend visible in particular since the 1990s, more than 80 jurisdictions1 subject merging parties to their merger control proceedings today. As there are no binding international rules on how to form and conduct merger control procedures, international merger projects face a large variety of procedural rules (e.g., relating to access to file or state of play meetings). This constitutes a considerable challenge for the merging parties as more and more undertakings conduct business in various jurisdictions resulting in the need to notify transactions to an increasing number of agencies with genuine procedural rules. Only a few agencies have published guidelines on their procedures in the form of Best Practices. This article analyzes the status quo with respect to such Best Practices, assesses the need for further proliferation of Best Practices and addresses some possible improvements.

I. Defining “best practices” The term “Best Practices” – as it is used in this paper – refers to non-legislative and non-binding sets of procedural rules issued by a national merger control authority or by a competition law-related institution with respect to merger control proceedings. Best Practices provide guidelines on how merger reviews will be conducted (if issued by an agency) or should be conducted (if issued by an international institution), e.g. in terms of transparency, case handling, communication, confidentiality and inter-agency coordination. Therefore, Best Practices rather deal with what is frequently referred to as the “due process” than with guidance on the substantive analysis of a merger. Different types of Best Practices exist and can be categorized as follows: • External Best Practices issued for the benefit of merger control agencies. These Best Practices are usually published by international organizations as guidance or recommendations for agencies on how to carry on their procedures; * Jochen Burrichter is a Partner at Hengeler Mueller, Düsseldorf and Brussels. ** Manuel Zandt was, at the time of writing, a Senior Associate at Hengeler Mueller, Düsseldorf. 1 Jonathan Galloway, “Convergence in International Merger Control”, 5(2) The Competition Law Review 179, 180 (2009) (citing Richard Whish, Competition Law, 6th edition, OUP, 2009, p. 801).

362  Merger Control in European and Global Perspective • Internal Best Practices issued by merger control agencies: These Best Practices are addressed by the authorities to merging parties to provide them with guidance on specific procedural aspects of the relevant merger control regime; • Best Practices for the merging parties. These guidelines recommend certain actions and behaviour for undertakings concerned by merger control proceedings. It is a frequently heard contention that Best Practices should be more actively promoted in order to foster the predictability of the outcome of merger control proceedings and a better understanding of the conduct of such proceedings by the concerned undertakings. This request needs to be seen in the context of another term which is frequently referred to (albeit more commonly in the context of substantive issues) when international developments with respect to merger control law are discussed, namely “convergence”. Generally speaking, convergence describes the process of moving differing positions towards union or even uniformity. Therefore, with respect to merger control procedures, the term “convergence”, involves not so much a loose harmonization of national procedures and sporadic cooperation between merger control agencies, but rather a persistent unification of procedural standards and approaches to merger review.2 Accordingly, convergence goes far beyond what “promoting Best Practices” implies: the promotion of Best Practices (in particular as far as external Best Practices are concerned) is just a measure of harmonization and, consequently, the first step if not a prerequisite for finally achieving convergence with respect to merger review standards (e.g. relating to the length of review periods or to timing issues with respect to the triggering of the obligation to notify a transaction). Holbrook put it this way: “Convergence is a process that begins with harmonization and cooperation and expands systematically over time.”3

II. Existing best practices Various Best Practices have been published. However, attention is not equally spread between the different categories of Best Practices. While the focus is on external and internal Best Practices, procedural guidelines for merging parties are only rarely directly addressed.

2 Cf. the alignment of the Canadian merger review rules to the US regime, in particular with respect to the new US-style two-stage waiting period process. For further details, see Steve Szentesi, “Convergence and Canada’s New Merger Control Law”, http://lawprofessors.typepad.com/files/ canada-convergence.docx; and the contributions of Cal Goldman and Adam Fanaki to this volume. 3 Cf. Dane Holbrook, “International Merger Control Convergence: Resolving Multijurisdictional Review Problems”, 7 UCLA Journal of International Law & Foreign Affairs 345, 348 (2002).

Jochen Burrichter and Manuel Zandt 

363  

1. External best practices for agencies Initial efforts to foster harmonization of procedural rules were made in 2000 by the International Competition Policy Advisory Committee (ICPAC), which prepared a comprehensive report4 on merger control proceedings for the US Department of Justice. One of the core proposals of the ICPAC Report was that national merger proceedings should be aligned on the basis of guidelines on merger control review. Today, the originators of external Best Practices are tripartite: besides competition law-related international organizations, private initiatives have also published procedural guidelines for antitrust authorities. Moreover, bilateral efforts have been undertaken between various merger control agencies in order to define specific rules for their cooperation in merger control cases.

a. International organizations By far the most significant driver with respect to external Best Practices is the International Competition Network (ICN), which seeks to facilitate cooperation and the exchange of know-how between competition law authorities globally. In 2002, its Merger Working Group5 published a comprehensive set of materials on Best Practices. The most important contributions are the Guiding Principles for Merger Notification and Review6 and the Recommended Practices for Merger Notification and Review Procedures.7 The former comprises only one page but it sets out the general formula which should govern merger control proceedings. For that purpose, it sets forth eight general principles (sovereignty, transparency, nondiscrimination, procedural fairness, efficiency and effectiveness, coordination, convergence and confidentiality). The Recommended Practices elaborate on these principles with respect to all aspects of merger control proceedings and specify them through more than fifty recommendations on how to establish due process of merger control proceedings. Further guidance for merger control agencies comes from the Organisation for Economic Co-operation and Development (OECD). In 2005, the Council of the OECD adopted its Recommendation on merger review8 which – while addressing similar issues such as efficiency, fairness and cooperation – are much less specific than the ICN’s Recommended Practices. Nevertheless, they provide general and abstract guidance on how national merger control agencies should conduct their proceedings. An additional set of external Best Practices comes from the Business and Industry Advisory Committee to the OECD (BIAC) in cooperation with the International http://www.justice.gov/atr/icpac/finalreport.htm. One of the Merger Working Group’s subgroups is specifically dedicated to merger notifications and review procedures. 6 http://www.internationalcompetitionnetwork.org/uploads/library/doc591.pdf. 7 http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf (last updated 2005). 8 http://www.oecd.org/dataoecd/3/41/40537528.pdf. 4 5

364  Merger Control in European and Global Perspective Chamber of Commerce (ICC) which published their Recommended framework for best practices in international merger control procedures9 in October 2001. It focuses on guidelines relating to the avoidance of unnecessary transaction costs, e.g. through recommendations on the initial notification requirements, and relates to procedural safeguards such as confidentiality.

b. Private initiatives Private initiatives are much less prominent with respect to publishing external Best Practices than international organizations. Apparently, only one private set of recommendations exists. In 2001, the Merger Streamlining Group,10 a group funded by globally active companies and consisting of antitrust practitioners, published its Best Practices for the Review of International Mergers (Discussion Draft).11 These guidelines contain a set of 45 recommendations relating to the general design of merger review procedures, agency practices and – making it distinct – the conduct of merging parties. They feature an apparent view on the issue of Best Practices through a lawyer’s lens: discouraging agencies from charging filing fees and the request for stopping the clock options can be considered as representing the demands of advocates.

c. Bilateral best practices Bilateral Best Practices should also be categorized as external Best Practices, as they are not addressed to the merging parties like internal Best Practices but govern the conduct of a merger control authority. Due to their specific nature, bilateral Best Practices are limited in scope. They usually aim at promoting consistent decisions in multijurisdictional merger cases and focus on coordination, cooperation (in particular with respect to the exchange of information) and communication issues. One of the most prominent examples is the “Best Practices on cooperation in 12 merger investigations” of the US-EU Merger Working Group which forms the basis of an intense exchange between the European Commission and the US merger authorities.13 There are similar arrangements between the EU and Canada and between the EU and Japan.

http://www.biac.org/statements/comp/BIAC-ICCMergerPaper.pdf. http://38.99.129.197/PracticeArea.aspx?ParID=bdbdc2a3-d34f-4535-b884-17a54f1391e9. 11  http://www.mcmillan.ca/Upload/Publication/NCampbell_BestPractices_ReviewIntlMergers _0901.pdf. 12   http://www.justice.gov/atr/public/international/docs/200405.pdf. More recently, see http:// ec.europa.eu/competition/mergers/legislation/best_practices_2011_en.pdf. 13 For examples in the past, see Frank Montag, “Konvergenz bei internationalen Fusionen”, FIW Heft 188, p. 50. 9

10

Jochen Burrichter and Manuel Zandt 

365  

2. Internal best practices published by agencies Compared with the large number of merger control regimes, apparently only a surprisingly low number of antitrust authorities have published internal Best Practices in order to provide merging parties with guidance on their procedural approach.14 In contrast, substantially more agencies provide guidelines on the principles of their substantive merger assessment and on the definition of certain terms relevant for the national merger control regime. Not even all EU Member States have issued internal Best Practices. Besides the European Commission15 itself, only the Netherlands,16 Ireland (twofold)17 and France (as part of catch-all guidelines on substantive assessment and procedural issues)18 have internal Best Practices in place which deal with (more or less) all relevant procedural issues in the context of merger control proceedings. These countries are accompanied by Romania19 and the UK,20 both of which, however, lack a full set of procedural rules: while Romania only provides guidance on the issue of access to file, the internal Best Practices in the UK deal only with the procedures before the Competition Commission but do not contain guidance with respect to the OFT. Apart from these and Norway21 as the only EFTA State, no other European jurisdiction has to date published internal Best Practices. On the other side of the Atlantic, only the US (split into several documents)22 and Canada23 lay down some principles of their merger review process in internal 14 On the basis of the information provided on www.iclg.co.uk, www.globalcompetitionforum.org and on the basis of a Google-search; it appears that no comprehensive list on this issue has been published yet. 15 “Best practices on the conduct of EC merger control proceedings”, http://ec.europa.eu/ competition/mergers/legislation/proceedings.pdf. 16 “NMa Best Practices on the Conduct of Concentration Control Proceedings”, http://www. nmanet.nl/Images/90106%20Spelregels%20voor%20concentratiezaken%20-%20vertaling_tcm16135884.pdf. 17 “Mergers and Acquisition Procedures”, http://www.tca.ie/images/uploaded/documents/Merger_ Procedures.pdf; “Access to the File in Merger Cases”, http://www.tca.ie/images/uploaded/documents/ Merger_File_Access_Procedures.pdf. 18 “Lignes directrices relatives au contrôle des concentrations – Procédure et analyse”, http://www. economie.gouv.fr/directions_services/dgccrf/concurrence/concentrations/lignesdirectrices_2007_ anglais.pdf. 19 “Guidelines on the rules for access to the Competition Council file in cases pursuant to Art. 5 and 6 of Competition Law No. 21/1996 and in the economic concentration cases”, http://www. competition.ro/documente/Instructiuni%20acces%20dosar_14852en.pdf. 20 “Competition Commission Rules of Procedure”, http://www.competition-commission.org.uk/ rep_pub/rules_and_guide/pdf/cc1.pdf. 21 “Best Practices on the conduct of merger control proceedings”, http://www.konkurransetilsynet. no/Global/English/Merger_Control_Proceedings_Best_Practices.pdf. 22 “What is the Premerger Notification program? An Overview”, “To File or Not to File – When you must file a premerger notification report form” and “A Model Request for Additional Information and Documentary Material”, http://www.ftc.gov/bc/hsr/introguides/introguides.shtm. See also “Statement of the Federal Trade Commission’s Bureau of Competition on Guidelines for Merger Investigations”, http://www.ftc.gov/os/2002/12/bcguidelines021211.htm; “Best Practices for Data, and Economics and Financial Analyses in Antitrust Investigations”, http://www.ftc.gov/be/bestpractices.shtm. 23 “Fees and Service Standards Handbook for Mergers and Merger-Related Matters”, http:// www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/vwapj/Final-Fees-Service-Handbook-Oct-2010-e. pdf/$FILE/Final-Fees-Service-Handbook-Oct-2010-e.pdf.

366  Merger Control in European and Global Perspective Best Practices for the benefit of merging parties. However, both sets of guidelines are – compared to the standards set by the European Commission or France – limited in scope. While the US guidance focuses on substantive issues and addresses procedural matters only partially, the Canadian Best Practices relate only to specific issues such as review periods, market contacts and fees. As far as Asia and Oceania are concerned, the situation is substantially the same as with Europe and the Americas: Only two jurisdictions, Singapore24 and Australia,25 have rather comprehensive internal Best Practices in place. Additionally, Japan26 provides some guidance for merging parties, but does so largely in the context of explanations on substantive issues.

3. Best practices for the merging parties Apparently, only a single set of Best Practices is available which offers procedural guidance on the proper conduct of the merging parties themselves. This set forms part of the Merger Streamlining Group’s “Best Practices for the Review of International Mergers” (see above). Apart from this, Best Practices relating to the merging parties are indirectly addressed in certain external and internal Best Practice guidelines, e.g. in the context of making available confidential information or with respect to recommendations on pre-filing contacts or regarding the synchronization of the initiation of merger control in different jurisdictions.

III. The scope of best practices While the different types of Best Practices contain distinct sets of guidelines, they reflect the commitment of their originators to provide guidance on the timely, consistent and transparent assessment of mergers.

1. External best practices Considering the rationale of external Best Practices, i.e. in particular to provide guidance for merger control authorities on how to shape their procedures, external Best Practices only establish general and broad principles for an efficient and 24 “CCS Guidelines on Merger Procedures”, http://app.ccs.gov.sg/cms/user_documents/main/ pdf/mergerprocedures_Jul07FINAL.pdf. The Merger Procedure Guidelines were revised in 2012. See http://www.ccs.gov.sg/content/dam/ccs/PDFs/CCSGuidelines/Revised%20Merger%20 Guidelines%20Procedures%202012.pdf. 25 “Merger review process guidelines”, http://www.accc.gov.au/content/index.phtml/itemId/740765. 26 “Guidelines to application of the Antimonopoly Act concerning Review of Business Combination” (last revised in 2011), http://www.jftc.go.jp/en/legislation_guidelines/ama/pdf/110713.2.pdf.

Jochen Burrichter and Manuel Zandt 

367  

transparent set up. These principles are usually verbalized as recommendations to the agencies. Therefore, external Best Practices rather provide a model than a formal code.27 Procedural issues usually dealt with include recommendations on the structural alignment of notification thresholds with – for the sake of legal certainty and reduction of unnecessary transaction costs – a clear preference for revenue-based thresholds rather than market share thresholds. Moreover, recommendations concern the agencies’ requirements on the timing of a notification (clear determination of the triggering event, preferably no deadline) and the content of the initial notification. Guidance on transparent procedures plays a crucial role, and this includes advocating the publication of all relevant rules, policies and decisions, which should be based only on competition-related criteria. Another major issue relates to inter-agency cooperation and communication (in particular with respect to the collection and evaluation of evidence).

2. Internal best practices Internal Best Practices are more specific than external Best Practices. They tie the principles laid down in external Best Practices down to explicit rules. Surprisingly, while internal Best Practices can be regarded as “implementing” external Best Practices, only a few explicitly make reference to external Best Practice guidelines, e.g. the preamble of Australia’s Merger review process guidelines (“the process guidelines continue to draw on the ICN’s eight guiding principles for merger review”). In contrast, Norway’s Best Practices on the conduct of merger control proceedings even state that “they are built on the experience of the Competition Authority in the conduct of merger control proceedings” as well as on a consultation process. The internal Best Practices listed in Section II above are heterogeneous in nature. It has already been noted that not all available internal Best Practices are so comprehensive that they can be considered as addressing all necessary potential issues which could arise in the course of merger control proceedings (e.g. Romania, the UK or Japan). Some have a clear focus on specific issues (e.g. Australia on communication issues). Some read more or less like an introduction to the relevant merger legislation (e.g. Singapore, and to some extent the US), while others appear to constitute a codification of the relevant agency’s procedural approach (e.g. Norway) and therefore effectively provide additional value. Moreover, the context of the internal Best Practices differs: while most of the guidelines are stand-alone documents, others form part of comprehensive guidance on substantive issues as well (in particular France). Frequent issues dealt with in internal Best Practices include details regarding the case team set-up and the decision-making process, pre-filing guidance, 27 Cf. the Best Practices for the Review of International Mergers of the Merger Streamlining Group, cited above note 10.

368  Merger Control in European and Global Perspective communication issues (e.g. the timing of the statement of objections) and guidance on the conduct of meetings (e.g. accessibility of third parties). Also relevant is guidance on the right to access the file (timing, accessible documents, entitled persons, procedures), on the requirements for the initial notification and on safeguarding confidentiality with respect to the business secrets provided by the merging parties.

3. Best practices for merging parties Since only one set of Best Practices specifically deals with the adequate conduct of the merging parties (as part of the Merger Streamlining Group’s Best Practices for the Review of International Mergers (see Section II)), no “standard” for such Best Practices has yet evolved. Not surprisingly, the Merger Streamlining Group recommends a cooperative approach with the merger control agency. Merging parties should be open for a candid communication with the authority (e.g. with respect to other jurisdictions concerned by the transaction including persistent status reports on foreign proceedings). Besides such obvious Best Practices, which are in the interest of the merging parties anyways, the authors of the Merger Streamlining Group also leave their practitioner’s lens on the subject matter aside, and recommend to volunteer all relevant competition issues arising in connection with the transaction and to refrain from seeking to use one clearance decision by one agency as leverage vis-à-vis another one.

IV. Aims and benefits – why Best Practices are useful Best Practices have benefits for all addressees, for merger control agencies as well as for merging parties. In the end, however, Best Practices mainly deliver benefits for the parties to a notifiable transaction.

1. Aims and benefits with respect to agencies Merger control agencies benefit from external Best Practices as well as from Best Practices for merging parties (although less pronounced in this respect). External Best Practices assist antitrust agencies with respect to the structuring of the relevant merger review procedures (thereby acting as a catalyst on the way to convergence). The agencies can align their procedures to precedents which are well-established and well-proven. If national proceedings conform to external Best Practices, this will foster procedural transparency and, consequently, will lead to a greater participation of the merging parties as well as to greater accountability

Jochen Burrichter and Manuel Zandt 

369  

and effectiveness of the decision at the end of the merger review. External Best Practices also help to streamline the intra-agency procedures and thereby boost efficiency with respect to merger review. In contrast, the benefits an agency can derive from Best Practices for the merging parties are presumably very limited. Their main focus is on cooperation issues; however, a cooperative level playing field with the agency is in the interest of the merging parties anyway. The added value of such Best Practices for merger agencies seems to be modest.

2. Aims and benefits with respect to undertakings The benefits of Best Practices for the merging parties are mainly driven by internal Best Practices. External Best Practices contribute only indirectly to these benefits to the extent the relevant internal Best Practice guidelines conform to the recommendations of external Best Practices. Provided the relevant guidelines do not focus on specific issues (such as access to file like in Romania), internal Best Practices provide comprehensive guidance for the merging parties and their advisors on how merger control procedures are generally conducted in a specific jurisdiction. Without such guidelines it would prove difficult to ascertain the relevant procedural standards. However, internal Best Practices summarize the “due process” of the relevant merger review so that the content and the current status of a specific merger case become transparent for the merging parties. Internal Best Practices are also beneficial for a better understanding of the procedures and for fair treatment of all undertakings (as the merger control authority – principally – binds itself to the published guidelines). Moreover, they result in a higher degree of predictability of the outcome of the proceedings for the business side. Undertakings also benefit from increased efficiencies (due to the improved legal certainty and the reduction of time and costs spent on the analysis of the relevant merger regime). In a globalized world, this is of particular importance, as international transaction may be subject to various merger regimes.

V. The effectiveness of Best Practices While the ICN and the Merger Streamlining Group have undertaken comparative efforts by way of balancing various national merger regimes against the ICN’s Recommended Practices, no comprehensive survey has been undertaken so far specifically to analyze the effectiveness of Best Practices for merger control procedures. The ICN’s Implementation Handbook28 only lists examples of legislative texts, rules and practices that conform to selected ICN standards. The 28

http://www.internationalcompetitionnetwork.org/uploads/library/doc325.pdf.

370  Merger Control in European and Global Perspective surveys of the Merger Streamlining Group29 had the same approach. Therefore, these surveys constitute static analyses but do not measure the success of Best Practices in the daily practice of merger control proceedings. However, indications in that respect can be derived from DG Competition’s Stakeholder Survey 2010 (see below).

1. General remarks on the effectiveness of Best Practices According to the authors of the Merger Streamlining Group’s Best Practices for the Review of International Mergers (see above), Best Practices have to achieve “balance between the fairness, due process and cost interests of private parties and the public interest in the ability of competition agencies to conduct effective and efficient merger reviews”. This balance provides an adequate benchmark for the assessment of the effectiveness of Best Practices.

a. External best practices It is impossible for external Best Practices to achieve this balance directly. As a fundamental principle of merger control (listed as no. 1 in the ICN’s Guiding Principles) each jurisdiction is sovereign to apply its own laws and rules to mergers with an appreciable effect in its markets. Therefore, national merger control agencies have a legitimate interest to review mergers pursuant to their own national substantive rules. As a consequence, external Best Practices on procedural issues necessarily need to be (i) non-binding and (ii) vague, as national merger regimes may substantially differ (e.g. the public enforcement by administrative experts in an administrative system versus judicial enforcement in a judicial system). Therefore, external Best Practices can only indirectly contribute to the balance to be achieved due to the need for adoption by national agencies as part of their respective national procedures. However, this indirect impact is effective – this can be derived from the recent adoption of new merger control regimes in China and India which – according to Galloway – “appear to be largely consistent with the ICN best practices”.30

b. Internal best practices The balance mentioned above can only be directly achieved by internal Best Practices, as they lay down the relevant procedural rules applicable to the specific 29 “Implementation of the ICN’s Recommended Merger Practices: A Work-in-(Early)-Progress”, http://www.mcmillan.ca/Upload/Publication/ANCampbell_JWRowley_ImplementationICN MergerPractices_0705.pdf; “Best Practices for Merger Review: Analysis and Recommendations for Review Processes in the United States and the European Union”, http://www.mcmillan.ca/ Upload/Publication/Best%20Practices%20for%20Merger%20Review_Analysis%20and%20 Recommendations_Rowley_1102.pdf. 30 Galloway, cited above note 1, at p. 191.

Jochen Burrichter and Manuel Zandt 

371  

transaction. However, internal Best Practices face structural deficits which diminish their effectiveness in achieving the balance. One of the main problems is that internal Best Practices are not binding on the relevant authority. None of the available internal Best Practices appears to have been enacted in the form of a legislative act. This leaves internal Best Practices guidelines open for adaptation to the specific circumstances of the transaction. Some agencies explicitly reserve a certain degree of flexibility, such as – besides the European Commission31 – the authorities in Norway,32 France,33 Ireland,34 Singapore35 and the Netherlands.36 While such flexibility is understandable and sometimes necessary to address particular requirements of the case, it counteracts other benefits linked with internal Best Practices which should not be neglected: legal certainty and transparency. Moreover, even if particular internal Best Practices guidelines are comprehensive and relate to the most important issues in the context of merger proceedings, they can never be exhaustive, as statutory provisions other than merger control law (e.g. administrative laws) apply as well, and influence the conduct of the proceedings. Internal Best Practices are therefore necessarily always only supplementary to statutory provisions; this diminishes the benefit of transparency and predictability and, in the end, the effectiveness of the relevant guidelines. Further, internal Best Practices frequently use open and generic language (“may”, “should”, “sufficiently ahead of meetings”, “in less straightforward cases”) in order to allow the authority a certain leeway with respect to the burden it puts on the merging parties, mainly when specific rights of the agency are concerned. This is of particular importance with respect to the fact-finding and provision of evidence. The European Commission, for example, “may obtain all necessary information from relevant persons”.37 While this is identical to the wording of Article 11(1) of the EU Merger Regulation, the Commission missed the chance to elaborate on its approach on the question of what is “necessary”. These considerations would have been well-situated in its Best Practices. The current situation, however, favours overly cumbersome requests on businesses and, therefore, puts effective proceedings at risk.

31 Recital 2: “The specificity of an individual case may require an adaptation of, or deviation from these Best Practices depending on the case at hand.” 32 Article 1(3): “do not have any binding effect on the way the Competition Authority will handle a case.” 33 Preamble: “not themselves in any way binding.” 34 Article 1(1): “The document is not intended to be a binding statement of how discretion will be exercised in a particular situation and should not be taken as such.” 35 Article 1(4): “They [The Guidelines] are not exhaustive. And do not set a limit on the investigation and enforcement activities of the CCS. In applying these guidelines, the facts and circumstances of each case will be considered.” 36 Recital 3: “not intended to set its procedures and practices in stone.” 37 Recital 26 of the Best Practices on the conduct of EC merger control proceedings.

372  Merger Control in European and Global Perspective

2. DG Competition Stakeholder Survey 2010 The impact of these broad powers of the European Commission in the context of fact-finding – amongst various other aspects – was dealt with in the DG Competition Stakeholder Survey38 initiated by the Commission in January 2010. The purpose of the survey was to determine the views of the relevant stakeholders on DG Competition’s actions, an indicator for the effectiveness of the proceedings of the Commission. While the survey, according to the final report (published in midOctober), revealed substantial praise for the effectiveness of DG Competition’s work and the integrity of its staff, it also highlighted several areas of criticism, in particular on the part of antitrust practitioners. DG Competition has announced that it will use the results of the survey as input for its best practices discussions.39 The survey was not conducted specifically with respect to merger control proceedings. Nevertheless, a selection of aspects may be fruitful for the purposes of this chapter. • Respondents complained that it was difficult to predict final decisions of the Commission during the early stages of a case; however, predictability was generally perceived to be better with respect to merger decisions.40 • Some respondents would like the Commission (which has been known to “tweak” timings) to be more transparent about its internal timetable of cases.41 • The observance of procedural rules (including Best Practices) was considered to be an area of good performance by the Commission. However, some complainants regretted the absence of rules on applying pressure on notifying parties (e.g. limiting the number of questions being asked during pre-merger guidance).42 • After formal notification, the burden on businesses imposed by the Commission was also criticized (in particular due to a disproportionate volume of questions; lack of understanding of the Commission that data systems of undertakings cannot always produce the requested data).43 • The timing of merger decisions was generally well regarded, while the timing of most other procedures was considered unacceptably long.44 However, even with regard to mergers, some respondents complained that the review process can take a lot longer than originally envisaged. 38 The aggregate report of the survey is available at http://ec.europa.eu/competition/publications/ reports/aggregate_report_en.pdf. The Commission also published separate reports with respect to the stakeholders which contributed to the survey. The report based on the lawyer’s responses is available at http://ec.europa.eu/competition/publications/reports/lawyers_en.pdf. 39 See MEMO/10/498 of 18 October 2010. 40 Stakeholder Report – Lawyers, p. 9. 41 Ibid., p. 19. 42 Ibid., p. 23. 43 Ibid., p. 25. 44 Ibid., p. 31.

Jochen Burrichter and Manuel Zandt 

373  

VI. Outlook External Best Practices have gained a widespread appreciation. However, the increasing proliferation of merger control regimes and the variety of jurisdictions which affect globally active undertakings heighten the need for national agencies to publish more internal Best Practices on the conduct of their procedures. In contrast, further promotion of Best Practice guidelines for merging parties does not appear to be a priority matter.

1. External best practices A comprehensive set of guidance for national legislation and agencies is available and issued by influential institutions, in particular the ICN. Apparently, the propagation of external Best Practices has already reached a satisfying level. Additional initiatives, e.g. supplementing the external Best Practices with typical procedural constellations as guidance for the agencies and the merging parties, should be originated by the ICN as the most influential and competent organization. Bilateral initiatives between national agencies or even private initiatives such as the Merger Streamlining Group may be of assistance with respect to achieving the aims of external Best Practices. However, in the end they are no adequate alternative to a nucleus such as the ICN. Private initiatives do not have – compared to the ICN – sufficient influence on national merger authorities. Moreover, bilateral Best Practices naturally will be restricted to a limited set of issues (mainly inter-agency cooperation) and, given the variety of distinct merger control regimes, they are likely to result only in an isolated application of Best Practices; this endangers the overall aim of convergence. However, it must be clear that a detailed and binding global set of external Best Practices aimed at aligning merger control procedures all over the world will not be achievable. Every jurisdiction is sovereign to assess merger cases having effects within a jurisdiction by the rules applicable in that jurisdiction. Such rules often originate from tradition rather than from factual reasons, which is why they are difficult to overcome. Moreover, any alignment by way of Best Practices comes to an end where national enforcement regimes are fundamentally different (administrative versus adversarial approach).

2. Internal best practices Compared with the large number of merger control regimes, a surprisingly low number of antitrust authorities have published Best Practice guidelines relating to their procedural rules. While substantive tests are becoming increasingly similar,

374  Merger Control in European and Global Perspective procedural rules largely remain distinctive around the world. National antitrust agencies should be encouraged to publish more internal Best Practices. Such explicit encouragement would be addressed best in the external Best Practices of the ICN and should supplement the current, fairly vague recommendation to conduct transparent proceedings. With respect to the European Union, a dynamic link to some core principles of the Best Practices of the European Commission could be possible (e.g. with respect to access to file) even for national cases, maybe supplemented by short guidelines on the relevant national specialties. Each internal Best Practices, which for the sake of certainty and predictability should refrain from generic and vague clauses, should deal at least with a minimum, standardized set of issues which cover the most important aspects of merger control proceedings. This standard set could be specified by the ICN and could include guidance on: • • • • •

pre-filing consultations; the scope of requests for information; the conduct of meetings (e.g. involvement of third parties); access to the file; and confidentiality issues.

Last but not least, agencies should explicitly bind themselves to their Best Practices. Opening clauses, such as those used in the Best Practices of Ireland, the Netherlands and Norway, may undermine the predictability and transparency that best practices are ideally designed to achieve. Deviations from the published guidelines, if any, should be limited to hardship cases, and they should require the consent of the merging parties or should only be possible to the benefit but not to the detriment of the parties.

3. Best practices for merging parties No further efforts seem necessary to foster the development of Best Practices on procedural matters for merging parties. Their main focus would be on cooperation issues anyway. However, a good relationship with the agencies should be one of the main interests of merging entities in order to ensure a swift and smooth clearance procedure. This does not have to be laid down in a formal set of rules. Many of the relevant issues are indirectly addressed in external or internal Best Practices as well, so there is no effective need to push Best Practices for merging parties.

Götz Drauz*

Promoting Best Practices within and among Competition Authorities and with Business

“Best Practices” are methods and techniques that have consistently shown results superior to those achieved with other means, and which are used as benchmarks to strive for.1 In the past twenty years, authorities around the world and other organizations have invested in the cultivation of good practice within and among merger control agencies, as well as with the business community. These efforts have enhanced efficiency and effectiveness in merger review; they have increased predictability and should also have reduced transaction costs for companies; and they have promoted competition by ensuring that multijurisdictional transactions will receive similar treatment and have similar outcomes in the relevant jurisdictions. However, much more should and can be done. The present paper examines first examples of promoting best practices in merger control within competition authorities, focusing in particular on the European Commission. It also reviews the efforts for cooperation and convergence among the world’s merger control agencies, with particular emphasis on the work of the International Competition Network (“ICN”), the Organisation for Economic Cooperation and Development (“OECD”), and as an example of best practice in a bilateral setting, the cooperation between the EU and the US. Finally, it reviews the paths available to the business community for instilling their perspective in the promulgation of best practices by merger control agencies.

Promoting best practices within competition authorities The European Commission’s example The European Commission is perhaps a good example for the adoption of best practices in merger control. Already in its early days of existence, DG Competition’s Merger Task Force (“MTF”) started developing new working methods. These were intended to foster a spirit of cooperation, particularly by moving away from a more hierarchical concept characterizing relations between authorities and companies in most parts of Europe and promoting open and constructive dialogue * Partner, Shearman and Sterling, Brussels. 1 See online Business Dictionary: http://www.businessdictionary.com/definition/best-practice.html.

376  Merger Control in European and Global Perspective that would help identify and address early on any competition issues arising from a proposed concentration. A first set of Best Practices in merger control was published in 1999. In the wake of the triple Court defeat of the Commission in 2002 before the Court of First Instance of the European Communities (now General Court of the European Union),2 these best practices were further developed and eventually formed the Best Practices on the conduct of merger control proceedings of 20 January 2004 as an integral part of the Commission’s merger review package3 to reform merger control in Europe. On 2 May 2003, the Commission had already published a Standard Model for Divestiture Commitments and a Standard Model for Trustee Mandates. These two standard texts are supplemented by an explanatory note and serve as best practice guidelines for the negotiation and implementation of remedies.4 The best practices developed in the framework of merger control paved the way for the preparation of a set of best practices in the context of antitrust proceedings. It is interesting to note that it took the European Commission some ten years before it eventually “exported” this concept from merger control to the wider antitrust world and published draft Best Practices in antitrust proceedings for public consultation.5 Each of these documents is briefly described below.

Best Practices on merger control proceedings The Best Practices of 2004 codified the system of pre-notification interaction between the parties to a concentration and DG Competition’s case teams which had developed over time. It allows parties to discuss with the Commission, on an informal and confidential basis, an envisaged concentration prior to notification. In this respect, the Best Practices encourage open and co-operative dialogue with business people and their legal advisers. These informal contacts at the pre-notification stage are useful, as they help ensure that any jurisdictional and other legal issues are dealt with early on. They also help avoid rejections of notifications by guiding parties to ensure that the notification form contains all necessary information. To achieve this, the Best Practices of 2004 on the conduct of merger control proceedings advises notifying parties to “[f]ully and frankly disclose information relating to all potentially affected markets and possible competition concerns”.6 2 See Case T-342/99, Airtours plc v. Commission [2002] ECR II-2585; Case T-310/01, Schneider Electric S.A. v. Commission [2002] ECR II-4071; Case T-5/02, Tetra Laval v. Commission [2002] ECR II-4381. 3 See http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf. 4 The standard texts and the explanatory note are published on DG Competition’s website. See http://ec.europa.eu/competition/mergers/legislation/divestiture_commitments. 5 See http://ec.europa.eu/competition/consultations/2010_best_practices/index.html. 6 Best Practices on merger control procedure, para. 16.

Götz Drauz 

377  

The cooperative spirit is also reflected in the opportunity given to parties who have submitted an incomplete Form CO to correct these omissions within a few days before a declaration of incompleteness, and the possibility to request a waiver of the obligation to provide information that is unnecessary for the examination of the case.7 The Best Practices also aim at enhancing the transparency of the proceedings. It foresees the possibility to hold “State of Play” meetings with notifying parties, and to hear the views of any interested third parties, such as customers, competitors, consumer organizations, and others through written submissions, or even through informal “triangular meetings”.8 In the same spirit, the Best Practices make it clear that, in cases where there are competition concerns, State of Play meetings may also be used for discussing potential remedies with the notifying party(ies).9 It also deals with issues of access to so-called “key documents” and important third party submissions contained in the Commission’s file following initiation of the proceedings and before a Statement of Objections (“SO”) is issued by the Commission services, and to the Commission’s file after the adoption of an SO until consultation of the Advisory Committee.10 Practice shows that these working methods are adhered to by DG Competition and companies alike, and have thus proven to be highly useful. By reducing – to the extent possible – hierarchical boundaries and formalities, and encouraging frank and open dialogue between the authority and the parties as early as the pre-notification phase, both sides help ensure more efficient review of complex transactions. However, from a practitioner’s standpoint, some merger investigations conducted by the Commission lately require more time and result in higher expenses. This phenomenon is not attributable to the current merger control framework as such; rather, it stems from an often dramatic lack of experience of case teams, which translates in ever-increasing demands for information of little probative value.

Best Practices on divestiture proceedings These best practice guidelines aim at streamlining the negotiation of the terms and conditions for commitments and trustee mandates. Merging parties need not negotiate terms and conditions for such commitments from scratch. The Standard Models are neither exhaustive, nor legally binding upon parties, but contain all standard provisions that should be included in divestiture commitments and trustee mandates, while allowing for some flexibility. Ibid., para. 19. Ibid., Section 5. 9 Ibid., paras. 40 et seq. 10 Ibid., Section 7.1. 7 8

378  Merger Control in European and Global Perspective The same spirit governs the adoption by the Commission of its 2008 Remedies Notice,11 which replaced the 2001 Remedies Notice. The revision is in line with the conclusions of the Commission’s Merger Remedies Study which was published in October 2005.12 In that study, the Commission undertook an ex post review of ninety-six remedies included in merger decisions adopted by the Commission between 1996 and 2000, analyzing the implementation and effectiveness of those remedies. This revision reflects the Commission’s intention to enhance the effectiveness of the divestiture procedures, and is overall positive. It is worth noting, however, that the new “Form RM” which was introduced for the submission of remedies proposals to the Commission is overly burdensome for companies and should be simplified in the future.

The 2010 Best Practices in antitrust proceedings The draft Best Practices in antitrust proceedings are contained in three documents: (i) Best Practices for antitrust proceedings, (ii) Best Practices for the submission of economic evidence (both in antitrust and merger proceedings) and (iii) Guidance on the role of the Hearing Officers in the context of antitrust proceedings. Each of these is briefly described below.

Best Practices for antitrust proceedings The draft Best Practices in antitrust proceedings set out the entire process of the Commission’s antitrust proceedings, from the Commission’s decision to look into a certain case until the potential adoption of a decision. Again, the aim is to further improve procedures by enhancing transparency and increasing the efficiency of the Commission’s investigations. According to the Commission, the draft Best Practices amends the Commission’s procedures by: • Earlier opening of formal proceedings, as soon as the initial assessment phase has been concluded; • Offering state of play meetings to the parties at key points of the proceedings; • Disclosing key submissions, including giving early access to the complaint, so that parties can already express their views in the investigative phase; • Publicly announcing the opening and closure of procedures, as well as when Statement of Objections has been sent; and 11 Commission Notice on remedies acceptable under the Council Regulation (EC) No 139/2004 and under Commission Regulation (EC) No 802/2004, 2008 OJ C267/1. 12 See http://ec.europa.eu/competition/mergers/legislation/remedies_study.pdf.

Götz Drauz 

379  

• Providing guidance on how the new instrument of commitment procedures is used in practice.13

Hearing Officers’ Guidance Paper The Hearing Officers’ role is to ensure that the rights of defence and other procedural rights of companies involved in competition proceedings are safeguarded. The aim of the Guidance Paper is to make that role more transparent. It therefore explains the Hearing Officers’ tasks as established in their mandate and outlines how they usually carry them out, while encouraging companies to make best use of an Oral Hearing.

Best Practices on submission of economic evidence The Best Practices on submission of economic evidence reflect the increasing importance of economics in merger and competition law cases. In order to streamline the submission of such economic evidence, the Best Practices on economic evidence outline the criteria which these submissions should fulfill. They also explain the practice of the competition department’s case teams and the Chief Economist when interacting with parties which submit economic evidence.

Some points of critique The Commission’s resolve to instill more transparency and predictability into its proceedings through the publication of Best Practices in antitrust proceedings, Best Practices for the Submission of Economic Evidence and Data Collection, and the Guidance on the Procedures of the Hearing Officers is a positive step. However, a number of prominent lacunae remain. For example, an issue where more clarity is required relates to the rules applicable to informal contacts. It would be more helpful if the Commission first explained on the basis of which criteria it draws the line between informal and formal meetings. In case the two types of meetings do not entail the same implications, it should clarify the probative value of any note drafted during or after an informal meeting, and its possible inclusion in the Commission file as an internal or accessible document. Also, the added value of triangular meetings should be clarified, since there is no record of such meetings to date. 13 See Press Release IP/10/2 of 6 January 2010 (“Improved transparency and predictability of proceedings”), http://europa.eu/rapid/pressReleasesAction.do?reference=IP/10/2&format=HTML&a ged=0&language=EN&guiLanguage=en.

380  Merger Control in European and Global Perspective The issue of legal professional privilege is crucial; addressing it in the Best Practices will help to enhance legal certainty, and is therefore welcome. However, the Best Practices fail to make reference to the treatment of legal professional privilege for lawyers who are not members of a bar of an EU Member State. The Commission should therefore consider providing guidance on whether it intends to acknowledge legal professional privilege in relation to these lawyers on the basis of either: (i) the same rules as those applicable to lawyers that are members of a bar in an EU Member State; or (ii) the rules of the bar of which the lawyer in question is a member. In our view, the Commission should take this step because investigations are often not limited to a single jurisdiction and competition authorities tend to welcome cooperation at an international level. Concerning the Hearing Officers’ Guidance Paper, it sadly misses the opportunity to propose more fundamental changes to the role of the Hearing Officer, changes that would make it attractive again for parties to request an Oral Hearing in merger cases. With regard to the Best Practices for the submission of economic evidence and data, an important clarification is suggested. The Commission indicates that responses to data requests “must be (i) complete; (ii) correct, and (iii) timely”.14 This statement appears to impose a high standard, in particular in comparison to the later explanations provided under section 3.4 (entitled “other recommendations”). For example, paragraph 65 indicates that “individual values within a variable must be consistent with the economic reality” and that “coherence between different variables is necessary”. While the data submitted should indeed aim to comply with such standards, in practice, data are not always “perfect” and it may not be possible to live up to such standards within the short deadlines in merger control. The Commission does recognize this in paragraphs 77 and 78, where it indicates for example that parties should draw the Commission’s attention as early as possible to the limitations in the data, and explain how steps have been taken to ensure reliability. This second standard is more realistic, and the Commission should therefore consider modifying paragraph 60 to indicate that responses to a data request “should be” complete, correct and timely.

Best Practices of other merger control agencies Apart from the European Commission, other merger control agencies have published best practices concerning merger control. For example: • In the US, the Federal Trade Commission’s Bureau of Competition published in 2002 a Statement on Guidelines for Merger Investigations,15 and in 2003, a Statement on Negotiating Merger Remedies,16 as well as a set of 14 15 16

Best Practices for the Submission of Economic Evidence and Data Collection, para. 60. See http://www.ftc.gov/os/2002/12/bcguidelines021211.htm. See http://www.ftc.gov/bc/bestpractices/bestpractices030401.shtm.

Götz Drauz 

381  

Best Practices for Data, and Economics and Financial Analyses in Antitrust Investigations.17 • In 2010, the UK Office of Fair Trading issued Good practice in the design and presentation of consumer survey evidence in merger inquiries and launched a public consultation,18 while the UK Competition Commission adopted in 2009 Suggested best practice for submissions of technical economic analysis from parties to the Competition Commission.19 • The same year in Germany, the Bundeskartellamt published its Best practices for expert economic opinions.20 Overall, the spirit of these documents is quite similar; their main aim is to increase transparency and cooperation between the reviewing merger control agency and the parties to a concentration, and to ensure the accuracy of the data provided, the effectiveness of the investigation proceedings and the avoidance of unnecessary costs for the parties.

Promoting Best Practices among competition authorities Since 1989, when the Merger Regulation21 came into force, and throughout the following years of accelerated globalization, merger control agencies around the world have been entrusted with the assessment of large-scale mergers and acquisitions between multinational companies. The increasing number of multijurisdictional mergers increased the occasions for different competition authorities around the world to work closely together with the same goal: to ensure that they do not adopt conflicting decisions and ultimately, to safeguard competition. Although the different jurisdictions worldwide are still far from harmonization of methodology, procedure and substantive analysis of mergers, many – and among them, the most influential merger control agencies around the world – cooperate closely to ensure a compatible outcome in terms of merger control enforcement. This need for more convergence is also obvious in the development of best practices, which aim to increase transparency and legal certainty for business while rendering the merger regulators’ work more efficient. See http://www.ftc.gov/be/bestpractices.shtm. See http://www.oft.gov.uk/OFTwork/consultations/closed/2011/merger-inquiries. The follow-up document (“Good practices in the design and implementation of consumer survey evidence in merger inquiries) was published by the OFT and the Competition Commission in March 2011. See http:// www.oft.gov.uk/shared_oft/consultations/merger-inquiries/Good-practice-guide.pdf. 19 See http://www.competition-commission.org.uk/rep_pub/corporate_documents/corporate_policies/ best_practice.pdf. 20 See http://www.bundeskartellamt.de/wEnglisch/download/pdf/Merkblaetter/Bekanntmachung_ Standards_Englisch_final.pdf. 21 Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings, 1989 OJ L395/1. 17 18

382  Merger Control in European and Global Perspective There are fora where such cooperation among merger control authorities is fostered with the view to exchanging experiences and developing best practices in merger control in a harmonious way. The present paper focuses, in particular, on the role of the ICN, the OECD, the bilateral efforts between the EU and the US, and other methods for the dissemination of best practices in merger cases.

Multilateral cooperation The ICN An important venue for the development of best practices in merger control among competition authorities is the ICN. Since its establishment in 2001, the ICN’s contribution to the effort of competition authorities around the world to promote convergence in their substantive merger analysis, and adopt compatible working methods and best practices has been pivotal.22 Today, the ICN members include the majority of the world’s competition agencies. The ICN has a number of working groups, one of which is the Merger Working Group.23 Its mission is to promote the adoption of best practices in the design and operation of merger review regimes. More specifically, its main aims are to: (i) enhance the effectiveness of each jurisdiction’s merger review mechanisms; (ii) facilitate procedural and substantive convergence; and (iii) reduce the public and private time and cost of multi-jurisdictional merger reviews.24 The ICN has adopted a number of documents produced by the Merger Working Group, including: • A set of eight guiding principles for merger notification and review, which concern issues such as transparency, non-discrimination, and procedural fairness;25 • A set of Recommended Practices for Merger Review Procedures, which in particular focus on the required link between a merger and the reviewing merger control agency, objective notification thresholds, and the efficient conduct of merger investigations;26 • A set of Recommended Practices for Merger Analysis;27 See ICN website, http://www.internationalcompetitionnetwork.org/about.aspx. See http://www.internationalcompetitionnetwork.org/working-groups/current/merger.aspx. This working group is co-chaired by the U.S. Department of Justice’s Antitrust Division and the Irish Competition Authority. For an analytical description of the ICN’s role and functions, see also Elizabeth Kraus and Maria Coppola, “The International Competition Network: A Virtual Reality” (2004), http:// www.ftc.gov/bc/international/docs/krauscoppolaicn04.pdf. 24 See the ICN’s website, http://www.internationalcompetitionnetwork.org/working-groups/current /merger.aspx. 25 See http://www.internationalcompetitionnetwork.org/uploads/library/doc591.pdf. 26 See http://www.internationalcompetitionnetwork.org/uploads/library/doc588.pdf. 27 See http://www.internationalcompetitionnetwork.org/uploads/library/doc316.pdf. 22 23

Götz Drauz 

383  

• Merger investigation techniques28 and the analytical framework for merger review, including substantive tests applied to mergers and remedies;29 and • A collection of national merger control rules and other related materials as well as templates for each jurisdiction.30 According to the ICN’s Statement of Achievements for the period June 2006 – May 2007,31 “[n]early half of the ICN member jurisdictions with merger laws have made changes that bring their merger notification and review systems into greater conformity with the Recommended Practices”. It follows from the above that, to a certain extent, the ICN’s effort to disseminate best practices among the world’s merger control regulators has been fruitful. The documents mentioned above are useful tools, likely to increase predictability concerning the investigation techniques and methodologies used in merger control, and to enhance cooperation among merger control agencies. The ICN declares it will continue to pursue the fulfillment of these goals during the years 2010–2014.32 However, the delay in achieving conformity of the merger review systems of all ICN members with the standards noted above is noteworthy. For the ICN’s mission to be achieved, it is important for Member States, in cooperation with national competition authorities, to make a conscious effort to accelerate the procedures necessary for the alignment of their national procedures with the ICN best practices.

The OECD The OECD comprises thirty developed and some developing countries from which competition agency representatives meet three times per year and exchange their experiences on competition issues and prepare recommendations concerning competition policy and enforcement.33 Among other areas, the OECD also contributes to the development of best practices in merger control through the OECD Competition Committee, which has been focusing on a broad range of issues related to the review of mergers under national competition laws. The OECD has produced a number of documents that foster the creation of common best practices among national merger control agencies: • In 1994, the OECD published a study prepared by Professors Richard Whish and Diane Wood with the title “Merger cases in the real world – A study of merger control procedures”.34 The study reviewed a number of transnational See http://www.internationalcompetitionnetwork.org/uploads/library/doc322.pdf. See http://www.international competitionnetwork.org/uploads/library/doc573.pdf. 30 See http://www.internationalcompetitionnetwork.org/working-groups/current/merger/templates. aspx. 31 See http://www.internationalcompetitionnetwork.org/uploads/library/doc392.pdf. 32 See http://www.internationalcompetitionnetwork.org/uploads/library/doc623.pdf. 33 See OECD’s Annual Report 2009 8, 32 (2009), http://www.oecd.org/dataoecd/38/39/43125523.pdf. 34 See http://www.oecd.org/dataoecd/4/40/31587583.pdf. 28 29

384  Merger Control in European and Global Perspective mergers, identified factors calling for greater cooperation among merger control agencies, and examined possibilities for the further development of cooperation and convergence in merger control worldwide. • In 1999, the OECD published a report on the Notification of Transnational Mergers.35 That report contains a review of the OECD members’ notification practices, and highlights the common elements of notification forms, aiming at reducing the challenges presented in multi-jurisdictional mergers. • These studies paved the way for some sector-specific or specialized reports, covering matters such as mergers in financial services (2000),36 media mergers (2003),37 merger remedies (2003),38 and vertical mergers (2008).39 • On 23 March 2005, the OECD Council adopted a Recommendation on merger review,40 which aims at greater convergence of merger review procedures, including cooperation among competition authorities, towards internationally recognized best practices. This recommendation is addressed to the OECD’s members. The recommendation invites members to ensure increased transparency and procedural fairness, and to make merger review procedures “effective, efficient, and timely”. It also promotes best practices which help merger control agencies and merging parties to save time and avoid unnecessary costs in multinational transactions. Member countries are also encouraged to facilitate effective cooperation and coordination of merger reviews. The OECD’s efforts, and its contribution in the merger field from an early stage clearly underpin the importance of promoting best practices among merger control agencies.

Bilateral cooperation: the EU-US example Since the 1980s, the European Commission and the US authorities have increasingly consulted each other on cases of transatlantic interest and effects. The need for a more systematic cooperation between the two sides of the Atlantic became increasingly clear. On 23 September 1991, the European Communities and the US entered into an Agreement “on the Application of Their Competition Laws”.41 The purpose of the Agreement is to “promote cooperation and coordination and lessen the possibility See http://www.oecd.org/dataoecd/40/2/2752153.pdf. See http://www.oecd.org/dataoecd/34/22/1920060.pdf. 37 See http://www.oecd.org/dataoecd/15/3/17372985.pdf. 38 See http://www.oecd.org/dataoecd/61/45/34305995.pdf. 39 See http://www.oecd.org/dataoecd/25/49/39891031.pdf. 40 See http://acts.oecd.org/Instruments/ShowInstrumentView.aspx?InstrumentID=195&Instrument PID=191&Lang=en&Book=False. 41 See Agreement between the Government of the United States of America and the Commission of the European Communities regarding the application of their competition laws – Exchange of interpretative letters with the Government of the United States of America, 1995 OJ L95/47, also available at http://eur-lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:21995A0427%2801%2 9:EN:HTML. 35 36

Götz Drauz 

385  

or impact of differences between the Parties in the application of their competition laws”.42 The Agreement provides for: • notification of cases being handled by the competition authorities of one Party, to the extent that these cases concern the important interests of the other Party (Article II), and exchange of information on general matters relating to the implementation of the competition rules (Article III); • cooperation and coordination of the actions of both Parties’ competition authorities (Article IV); • a “traditional comity” procedure by virtue of which each Party undertakes to take into account the important interests of the other Party when it takes measures to enforce its competition rules (Article VI); • a “positive comity” procedure by virtue of which either Party can invite the other Party to take, on the basis of the latter’s legislation, appropriate measures regarding anticompetitive behaviour implemented on its territory and which affects the important interests of the requesting Party (Article V).43 • regular bilateral meetings to share information on current enforcement activities and priorities, as well as on economic sectors of common interest, to discuss policy changes, and to discuss other matters of mutual interest relating to the application of competition laws. However, cooperation between the EU and the US was not always obvious. Probably the most classic example where the approach adopted by the EU and the US was diametrically opposed was the GE/Honeywell case.44 In 2000, General Electric announced its plans to acquire Honeywell. While the merger was cleared by the US authorities, it was blocked by the European Commission on 3 July 2001. The rationale for the prohibition was that, by combining GE’s dominance of the large commercial jet engine market with Honeywell’s extensive portfolio of avionics and non-avionics, the new company would be able to “bundle” products and stifle competition through the creation of a monopoly on a horizontal level. The US merger regulators did not share that view, finding, on the contrary, that the merger would enhance competition and have positive effects for consumers, as it would reduce prices. Underlining the divergent approaches to this case, United States Assistant Attorney General Charles James described the Commission’s analysis as “antithetical to the goals of antitrust law enforcement”.45 The need for more convergence in the approaches of the different merger agencies involved in the same transaction was more than obvious. In the same speech, Charles James emphasized that “virtually any large transaction involving international businesses is likely to be subject to review in both the U.S. and under the European Merger Regulation. And in cases in Ibid., Article I, para. 1. See ibid., as well as the summary at http://ec.europa.eu/competition/international/bilateral/usa. html. 44 M.2220, General Electric/Honeywell (3 July 2001). 45 See Charles James “International Antitrust in the Bush administration”, speech, Ottawa, 21 September 2001, http://www.justice.gov/atr/public/speeches/9100.htm. 42 43

386  Merger Control in European and Global Perspective which the U.S. and E.U. are reviewing the same transaction, both jurisdictions consider themselves to have a large stake in reaching non-conflicting outcomes, whenever possible. By notifying, consulting, and coordinating under our cooperation agreement, we have managed to avoid inconsistent results in the vast majority of cases. Indeed, the close collaboration on facts, analysis and remedies between the Antitrust Division and the European Commission in the Sprint/MCI, Alcoa/Reynolds, MCI/WorldCom, and Dresser/Halliburton investigations stand out as cases in point. And, although we reached different conclusions in GE/Honeywell, […], it is essential to put GE/Honeywell in the context of the many other cases that were resolved without any conflict whatsoever.”

Mr. James added that, “despite certain differences in our respective antitrust laws, we have tended to reach the same set of conclusions on a matter when we become fully engaged with one another on the analysis and are working from a common set of facts. The importance of reaching consistent outcomes is obvious. Besides imposing substantial costs on the merging firms involved, divergent outcomes undermine the public’s confidence in the work we do and risk politicizing antitrust – both of which can have adverse effects on sound and predictable antitrust enforcement.”

Similarly, European Commission Competition Commissioner at the time, Mario Monti, declared that “GE/Honeywell is a rare case where the transatlantic competition authorities have disagreed. I am determined to strengthen our bilateral cooperation in the future to try and reduce this risk further.”46 In retrospect it appears that the GE/Honeywell case acted as a catalyst to further improve bilateral cooperation.47 It was not until some eight years later, in the Oracle/Sun Microsystems case, that some tension arose again and led to some stark statements on the US side. While the US Department of Justice had approved the acquisition by Oracle of Sun Microsystems on 20 August 2009, on 3 September 2009 the European Commission opened an in-depth investigation into the deal,48 which led to the adoption of an SO on 9 November 2009. Ultimately, on 21 January 2010, the Commission also approved the transaction.49 Following GE/Honeywell, in 2002 the “US-EU Merger Working Group Best Practices on cooperation in merger investigations” was adopted.50 The EU-US Best Practices is not legally binding but provides an organizational framework for transatlantic cooperation in reviews of individual merger cases. Its main objectives are: “to promote fully-informed decision-making on the part of both sides’ authorities, to minimize the risk of divergent outcomes on both sides of the Atlantic, to facilitate 46 See Press Release IP/01/939 of 3 July 2001 (“The European Commission prohibits GE’s acquisition of Honeywell”). 47 An indicative list of cases where the EU and the US merger control agencies cooperated successfully is contained in John Parisi, “Cooperation Among Competition Authorities in Merger Regulation”, 43 Cornell International Law Journal 55 (2010). See also Charles Stark, “Improving Bilateral Antitrust Cooperation”, speech, 23 June 2000, http://www.justice.gov/atr/public/speeches/5075.pdf. 48 See Press Release IP/09/1271 of 3 September 2009 (“Mergers: Commission opens in-depth investigation into proposed takeover of Sun Microsystems by Oracle”). 49 M.5529, Oracle/Sun Microsystems (21 January 2010). 50 See http://ec.europa.eu/competition/international/bilateral/eu_us.pdf.

Götz Drauz 

387  

coherence and compatibility in remedies, to enhance the efficiency of their respective investigations, to reduce burdens on merging parties and third parties, and to increase the overall transparency of the merger review processes.”51

Under the EU-US Best Practices, cooperation is most effective when the investigation timetables of the reviewing agencies run more or less in parallel.52 Merging companies will therefore be offered the possibility to meet at an early stage with the agencies to discuss timing issues. Parties are also encouraged to permit the EU and US agencies to exchange information which they have submitted during the course of an investigation through confidentiality waivers and, where appropriate, to allow joint EU/US interviews of the companies concerned.53 The practices designate key points in the respective EU and US merger investigations when it may be appropriate for direct contacts to occur between senior officials on both sides.54 Finally, concerning remedies, the EU-US Best Practices provide that, to the extent consistent with their respective law enforcement responsibilities, the agencies should advise the parties to consider coordinating the timing and substance of remedy proposals being made to the EU and US agencies, so as to minimize the risk of inconsistent results or of subsequent difficulties in implementation.55

Other ways to increase inter-agency cooperation Besides the aforementioned written best practices, other useful ways for promoting cooperation and developing best practices in the way different authorities operate in merger investigations are extended lecture visits and meetings between the EU, US and national competition authorities, or temporary secondments of officials to other countries.56 The regular participation of officials dealing with merger cases from different countries in international conferences, seminars and workshops concerning best practices in merger policy and enforcement can also contribute to this effort.

Promoting Best Practices with Business Companies have every interest in familiarizing themselves with the working methods and extent of cooperation by merger control agencies. They also have an Ibid., para. 2. Ibid., para. 4. 53 Ibid., para. 7 54 Ibid., para. 12. 55 Ibid., para. 14. 56 See, e.g., Commission Decision C(2008) 6866 final of 12 November 2009 laying down rules on the secondment to the Commission of national experts and national experts in professional training, http://ec.europa.eu/civil_service/docs/regime_end_en.pdf. 51 52

388  Merger Control in European and Global Perspective interest in influencing the agencies’ practice towards increased efficiency, clarity, transparency and impartiality in the application of merger control. It is thus important for companies not only to follow closely all developments in the legislation and substantive analysis of merger control agencies, but also to be actively involved in the development and promulgation of best practices. There are several ways to foster the participation of business in the formulation of best practices. Encouraging contributions to public consultations is one of them. The contributions submitted by business and by independent legal advisors in response to such consultations, which the Commission routinely publishes on its webpage,57 are numerous and they often contain very interesting suggestions that can benefit, among other policy matters, the merger review process. Another way to achieve this is by inviting business representatives to participate in intergovernmental conferences with representatives from both sides, as well as in working groups and panels organized by international networks and organizations. For example, both the ICN and the OECD endorse the contributions of companies, through their legal counsel, as well as external independent attorneys or non-governmental agencies. Such events offer valuable opportunities for business representatives and agency officials to exchange experiences from multi-jurisdictional merger reviews and foster cooperation and coordination among the agencies.

Convergence? Undoubtedly, the efforts made to formulate best practices in merger control are highly beneficial for all participants in complex merger investigations. However, there is still a long way to go before true convergence is reached in the way the world’s competition authorities conduct their review of mergers. Apart from remaining differences in the substantive analysis of mergers from one jurisdiction to the other, one cannot help but observe the significant procedural differences in the implementation of merger control by different countries, particularly outside the EU. In continental Europe, and recently also in the UK through the planned merger between the Office of Fair Trading and the Competition Commission, there is a tendency to simplify merger review by having one administrative body in charge of the review process. However, other countries still insist on having much more complicated structures. For example, while in Brazil the only decision-making agency is CADE, the Secretaria de Direito Econômico (“SDE”) of the Ministry of Justice issues nonbinding opinions to and below CADE, the Secretaria de Acompanhamento Econômico (“SEAE”) of the Ministry of Finance and, dealing only with telecommunications (in this case replacing both the SDE and the SEAE), 57

See http://ec.europa.eu/competition/consultations/2010_best_practices/index.html.

Götz Drauz 

389  

Agência Nacional de Telecomunicações (“ANATEL”) of the Ministry of Telecommunications. Filing also depends on opinions issued by the Office of the Attorney General of CADE (“ProCADE”) and the Office of the Attorney General of the Republic. Accordingly, a typical filing in Brazil will go through in the following order: the SEAE, the SDE, the Office of the Attorney General of CADE, and the Office of the Attorney General of the Republic, before finally being decided by CADE. Until recently, all filings had to go through full reviews, regardless of the size of the companies and of the relevant transactions.58 In China, the Antimonopoly Bureau of the Ministry of Commerce (“MOFCOM”) has complete responsibility for Chinese merger review, subject to coordination by the Anti-Monopoly Commission (“AMC”). However, although neither the Foreign M&A Regulations nor the Chinese Antimonopoly Law (“AML”) expressly provides for antitrust-based merger review by specific industry regulators, in practice, MOFCOM regularly requests input from other sector regulators and government authorities during merger reviews, such as the National Development and Reform Commission (“NDRC”) and the Ministry of Industry and Information Technology (“MIIT”).59 Merger review also differs significantly from one country to the other in terms of timing. For example, typically, clearance of mergers in Brazil, Argentina60 and Colombia61 is granted between twelve and fourteen months from the date of notification. Moreover, in some cases notified in China and Russia, although the respective authorities have informed the parties that they have cleared the transaction, they have omitted sending them the official clearance decision until much later.62 Finally, the authorities in the latter jurisdictions tend to send very lengthy questionnaires, which are often unrelated to the transaction and could thus be seen as a “fishing expedition” aimed at obtaining general economic intelligence data.

Conclusion It becomes clear from the foregoing examples that the required level of convergence concerning best practices in merger review worldwide is yet to be reached. For instance, the ICN Recommended Practices are not fully applied in most ICN member countries. A permanent monitoring of compliance is needed, and the private bar could become instrumental in such an exercise. The regulatory cost produced by non-transparent, slow or heavy handed merger control systems is very considerable and risks eating away the benefits of trans-border and international 58 See “Brazil”, in GCR, The Antitrust Review of the Americas 2010. For an update, see Abel Mateus, “The New Brazilian Merger Control Regime”, this volume, pp. 315 et seq. 59 See “China”, in GCR, The Asia-Pacific Antitrust Review 2010. 60 See, e.g., M.4691, Schering-Plough/Organon Biosciences (11 October 2007); M.1879, Boeing/ Hughes (27 September 2000). 61 See, e.g., M.5860, Cooper/Danaher/JV (2 June 2010). 62 See, e.g., ibid.

390  Merger Control in European and Global Perspective restructuring and consolidation. It is of little consolation that, in the end, very few if any large international mergers are blocked if, on the way to clearance, too many unnecessary obstacles are placed in their path. The development, mutual exchange and constant improvement of best practices is certainly one way to foster transparency, predictability and efficiency in international merger control, and reduce processing time and cost.

References

I. Publication, speeches and presentations Aitken, Melanie, “Remarks to ABA/US Chamber Event”, 1 February 2010, http:// www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03216.html Aitken, Melanie, “Keynote Dinner Address”, Cambridge, Ontario, 3 February 2010, http://www.competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03205.html Aitken, Melanie, “Remarks to CBA Fall Competition Law Conference”, 30 September 2010, Gatineau, Quebec, http://www.competitionbureau.gc.ca/eic/ site/cb-bc.nsf/eng/03306.html Aktas, Nihat, Eric De Bodt and Richard Roll, “Market Response to European Regulation of Business Combinations”, 39 Journal of Financial and Quantitative Analysis 731 (2004) Aktas, Nihat, Eric De Bodt and Richard Roll, “Is European M&A Regulation Protectionist?”, 117 The Economic Journal 1096 (2007) Almunia, Joaquín, “Cooperation and Convergence: Competition Policy in the 21st Century”, SPEECH/10/183 of 27 April 2010 Almunia, Joaquín, “New Transatlantic Trends in Competition Policy”, SPEECH/10/305 of 10 June 2010 Almunia, Joaquín, “State of Play and Future Outlook, Competition Day”, SPEECH/10/576 of 21 October 2010 Almunia, Joaquín, “Compliance and Competition Policy”, SPEECH/10/586, of 25 October 2010 Almunia, Joaquín, “Fair Process in EU Competition Enforcement”, SPEECH/11/396 of 30 May 2011 Anthony, Sheila, “Antitrust and Intellectual Property: The Crossroads” Program: Riddles and Lessons from the Prescription Drug Wars: Antitrust Implications of Certain Types of Agreements Involving Intellectual Property”, speech, San Francisco, 1 June 2000, http://www.ftc.gov/speeches/anthony/sfip000601.shtm

392  Merger Control in European and Global Perspective Arquit, Kevin, “Unilateral Effects Analysis in the 2010 Horizontal Merger Guidelines: Beginning of the End of the Age of Restraint?”, in Barry Hawk, ed., International Antitrust Law and Policy: Fordham Competition Law 2010, Juris Publishing, 2011, Chapter 6 Arrow, Kenneth, “Economic Welfare and the Allocation of Resources for Invention”, in The Rate and Direction of Inventive Activity: Economic and Social Factors, National Bureau of Economic Research, 1962, pp. 609 et seq. Auletta, Ken, World War 3.0: Microsoft and its Enemies, Random House, 2001 Baer, William et al., “A Study of the Commission’s Divestiture Process”, 1999, http://www.ftc.gov/os/1999/08/divestiture.pdf Baker, Jonathan and Carl Shapiro, “Reinvigorating Horizontal Merger Enforcement”, in Robert Pitosfsky, ed., How the Chicago School Overshot the Mark: The Effect of Conservative Economic Analysis on U.S. Antitrust, OUP, 2008, pp. 235 et seq. Baker, Jonathan and Carl Shapiro, “Detecting and Reversing the Decline in Horizontal Merger Enforcement”, 22 Antitrust 29 (2008) Baker, Jonathan and Carl Shapiro, “The Perils of Armchair Analysis: Evaluating Merger Enforcement During the Obama Administration” (2012), http://ssrn. com/abstract=2115273 Baldanza, Huy do and Douglas New, “Canada: Foreign Investments”, in Global Competition Review, The Asia-Pacific Antitrust Review 2012, Section 3, Country Chapter 5 Bardong, Andreas, “Die zweite Inlandsumsatzschwelle, kein Änderungsbedarf!”, (2011) WuW 350 Barthelmeß, Stephan and Lars-Peter Rudolf, “Extraterritoriale Anwendbarkeit der deutschen Fusionskontrolle und das völkerrechtliche Abwägungsgebot”, (2003) WuW 1176 Baumol, William, “Contestable Markets: An Uprising in the Theory of Industry Structure”, 72 American Economic Review 1 (1982) Baumol, William, John Panzar and Robert Willig, Contestable Markets and the Theory of Industry Structure, Harcourt Brace Jovanovich, 1982 Bellamy, Christopher, “ECHR and Competition Law Post Menarini: An Overview of EU and National Case Law”, e-Competitions No. 47946 (5 July 2012)

References  393   Benyon, Frank, Direct Investment, National Champions and EU Treaty Freedoms: from Maastricht to Lisbon, Hart Publishing, 2010 Bergman, Mats, Maria Jakobsson and Carlos Razo, “An Econometric Analysis of the European Commission’s Merger Decisions”, 23 International Journal of Industrial Organization 717 (2005) Bok, Derek, “Section 7 of the Clayton Act and the Merging of Law and Economics”, 74 Harvard Law Review 226 (1960) Boland, Sean et al., “Remarks on Mergers, Cartels, and Single Firm Conduct – Panel Discussion, George Mason Law Review Antitrust Symposium, Washington, D.C., September 13, 2006”, 14 George Mason Law Review 879 (2007) Bougie, Jacques, “Reflections on the Merger Task Force at the Turn of the Millennium: the Requirement for Convergence of Multijurisdictional Merger Review Systems”, European Commission Merger Task Force’s 10th Anniversary Conference (September 2000) Brady, Una and Robert Feinberg, “An Examination of Stock-price Effects of EU Merger Control Policy”, 18 International Journal of Industrial Organization 885 (2000) Brandenburger, Rachel, “Transatlantic Antitrust: Past and Present”, speech, St. Gallen, 21 May 2010, http://www.justice.gov/atr/public/speeches/260273.htm Brandenburger, Rachel, “International Competition Policy and Practice: New Perspectives?”, speech, London, 29 October 2010, http://www.justice.gov/atr/ public/speeches/270980.pdf Brandenburger, Rachel and Joseph Matelis, “The 2010 U.S. Horizontal Merger Guidelines: An Historical and International Perspective”, 25 Antitrust 48 (summer 2011) Brannon, Leah and Kathleen Bradish, “The Revised Horizontal Merger Guidelines: Can the Courts be Persuaded?”, 10 Antitrust Source (October 2010). Bresnahan, Timothy, “Post-entry Competition in the Plain Paper Copier Market”, 75 American Economic Review 15 (1985) Buchmann, Eva, Positive Comity im internationalen Kartellrecht, European Law Publications, 2004

394  Merger Control in European and Global Perspective Carlton, Dennis, “The Need to Measure the Effect of Merger Policy and How to Do It”, 22 Antitrust 39 (Summer 2008) Carlton, Dennis, “Revising the Horizontal Merger Guidelines”, 6 Journal of Competition Law and Economics 619 (2010) Carlton, Dennis, “Comment on Proposed Horizontal Merger Guidelines” (4 June 2010), http://www.ftc.gov/os/comments/hmgrevisedguides/548050-00034.pdf Carlton, Dennis and Mark Israel, “Will the New Guidelines Clarify or Obscure Antitrust Policy?”, 10 Antitrust Source (October 2010) Carstensen, Peter and Harry First, “Rambling Through Economic Theory: Topco’s Closer Look”, in Eleanor Fox and Dan Crane, eds., Antitrust Stories, Foundation Press, 2007, chapter 6 Clark, John, “Competition Law and Policy Developments in Brazil”, 2(3) OECD Journal of Competition law and Policy 181 (2000) Clougherty, Joseph and Jo Seldeslachts, “Deterrence of Horizontal Mergers: Empirical Evidence from U.S. Industries”, mimeo (2010) Coate, Malcolm, and Fred McChesney, “Empirical Evidence on FTC Enforcement of the Merger Guidelines”, 30 Economic Inquiry 277 (1992) Coate, Malcolm and Joe Simons, “Continuity and Change in the 2010 Merger Guidelines”, CPI Antitrust Journal (October 2010/2) Coate, Malcolm and Joe Simons, “In Defense of Market Definition” (forthcoming, Antitrust Bulletin, 2012), http://papers.ssrn.com/sol3/papers.cfm?abstract_ id=1967208 Crane, Daniel, “Has the Obama Justice Department Reinvigorated Antitrust Enforcement?”, 65 Stanford Law Review 13 (2012) Dabbah, Maher, Competition Law and Policy in the Middle East, Cambridge University Press, 2007 Dabbah, Maher, “Cross-border Merger Control: Challenges for Developing and Emerging Economies”, report for the OECD (2011), http://www.oecd. org/officialdocuments/publicdisplaydocumentpdf/?cote=DAF/COMP/ GF(2011)1&docLanguage=En

References  395   De Leeuw, Edith, Joop Hox and Don Dillman, “The Cornerstones of Survey Research”, in Edith De Leeuw, et al., eds., International Handbook of Survey Methodology, Psychology Press, 2008, Chapter 1 Devai, Rita et al., “Merger Case M.5421 Panasonic/Sanyo – Batteries Included or ‘Lost in Translation’?”, Competition Policy Newsletter, No. 1 – 2010, pp. 60 et seq. Deloitte and Touche LLP, “The Deterrent Effect of competition enforcement by the OFT”, OFT962 (November 2007), http://www.oft.gov.uk/shared_oft/ reports/Evaluating-OFTs-work/oft962.pdf Drauz, Götz and Christopher Jones, eds., EU Competition Law, Volume II – Mergers and Acquisitions, Claeys & Casteels, 2006 Duso, Tomaso, Damien Neven and Lars-Hendrik Röller, “The Political Economy of European Merger Control: Evidence Using Stock Market Data”, 50 The Journal of Law and Economics 455 (2007) Duso, Tomaso, Klaus Gugler and Florian Szücs, “An Empirical Assessment of the 2004 EU Merger Policy Reform”, working paper (2010) Duso, Tomaso, Klaus Gugler and Burcin Yurtoglu, “How Effective is European Merger Control?”, mimeo (2010) Eckbo, Espen (1983): “Horizontal Mergers, Collusion, and Stockholder Wealth”, 11 Journal of Financial Economics 241 (1983) Ehlermann, Claus-Dieter and Mel Marquis, eds., European Competition Law Annual 2008: Antitrust Settlements under EC Competition Law, Hart Publishing, 2010 Ehlermann, Claus-Dieter and Mel Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Ellert, James, “Mergers, Antitrust Law Enforcement and Stockholder Returns”, 31 Journal of Finance 715 (1976) Elman, Philip, “The Need for Certainty and Predictability in the Application of the Merger Law”, 40 New York University Law Review 613 (1965) Fama, Eugene, Lawrence Fisher, Michael Jensen and Richard Roll, “The Adjustment of Stock and Prices to New Information”, 10 International Economic Review 1 (1969)

396  Merger Control in European and Global Perspective Farrell, Joseph and Carl Shapiro, “Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition”, 10(1) B.E. Journal of Theoretical Economics, Article 9 (2010) First, Harry and Tadashi Shiraishi, “Concentrated Power: The Paradox of Antitrust in Japan”, in Daniel Foote, ed., Law in Japan. A Turning Point, University of Washington Press, 2007, chapter 21 Fletcher, Amelia, “UPP: Up and away? The Rising Prevalence of Pricing Pressure Indicators”, presentation, EUI Competition Workshop, 12 November 2010 Fontane, Theodore, Effi Briest, Penguin, 1896/2001 Forrester, Ian, “The Modernisation of EC Antitrust Policy: Compatibility, Efficiency, Legal Security”, in Claus-Dieter Ehlermann and Isabela Atanasiu, eds., European Competition Law Annual 2000: The Modernisation of EC Antitrust Policy, Hart Publishing, 2001 Forrester, Ian, “A Bush in Need of Pruning: The Luxuriant Growth of ‘Light Judicial Review’”, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Forwood, Nicholas, “The Commission’s ‘More Economic Approach’ – Implications for the Role of the EU Courts, the Treatment of Economic Evidence and the Scope of Judicial Review”, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Fox, Eleanor, “GE/Honeywell: The U.S. Merger that Europe Stopped – A Story of the Politics of Convergence”, in Eleanor Fox and Dan Crane, eds., Antitrust Stories, Foundation Press, 2007, chapter 12 Galloway, Jonathan, “Convergence in International Merger Control”, 5(2) The Competition Law Review 179 (2009) Gavil, Andrew, “A First Look at the Powell Papers: Sylvania and the Process of Change in the Supreme Court”, 17 Antitrust 8 (fall 2002) Georgi, Alexandra, Die Aufgreifkriterien des § 35 GWB: Eine Effizienzanalyse nach Einführung der doppelten Inlandsumsatzschwelle, Nomos, 2012 Hawk, Barry, “EU ‘Modernisation’: A Latter-day Reformation”, Global Competition Review (August/September 1999), p. 12

References  397   Hemphill, C. Scott and Mark Lemley, “Earning Exclusivity: Generic Drug Incentives and the Hatch-Waxman Act”, 77 Antitrust Law Journal 947 (2011) Dane Holbrook, “International Merger Control Convergence: Resolving Multijurisdictional Review Problems”, 7 UCLA Journal of International Law & Foreign Affairs 345 (2002) ICPAC Final Report (2000), http://www.justice.gov/atr/icpac/finalreport.html International Bar Association, Report of the Task Force on Extraterritorial Jurisdiction, http://tinyurl.com/taskforce-etj-pdf Italianer, Alexander, “Safeguarding Due Process in Antitrust Proceedings”, Fordham speech, 23 September 2010 Jaeger, Marc, “Can the General Court Increase Coherence in the European Union Judicial System?”, in Tristan Baumé, Edmon Oude Elferink, Pauline Phoa and Dominique Thiaville, eds., Today’s Multilayered Legal Order: Current Issues and Perspectives. Liber Amicorum Arjen W.H. Meij, Paris Legal Publishers, 2011, Chapter 8 James, Charles, “International Antitrust in the Bush administration”, speech, Ottawa, 21 September 2001, http://www.justice.gov/atr/public/speeches/9100. htm Juster, F. Thomas, “Prediction and Consumer Buying Intentions”, 50 American Economic Review 604 (1960) Kaplow, Louis, “Why (Ever) Define Markets?”, The Harvard John M. Olin Discussion Paper Series, Discussion Paper No. 666, March 2010 Katz, Michael and Howard Shelanski, “Mergers and Innovation”, 74 Antitrust Law Journal 1 (2007) Kay, John, Foundations of Corporate Success: How Business Strategies Add Value, OUP, 1993/1995 Khemani, Shyam and Daniel Shapiro, “An Empirical Analysis of Canadian Merger Policy”, 41 The Journal of Industrial Economics 161 (1993) Kolasky, William, “International Comity in Antitrust: Advances and Challenges”, Legal Backgrounder, Washington Legal Foundation, 22 (2007), No. 16 (May 25), http://www.wlf.org/upload/05-25-07kolasky.pdf

398  Merger Control in European and Global Perspective Kopke, Alex et al., “Merger Remedies Study”, 2005, http://ec.europa.eu/ competition/mergers/legislation/remedies_study.pdf Kovacic, William, “Antitrust Decision Making and the Supreme Court: Perspectives from the Thurgood Marshall Papers”, 42 Antitrust Bulletin 93 (1997) Kovacic, William, “Evaluating Antitrust Experiments: Using Ex Post Assessments of Government Enforcement Decisions to Inform Competition Policy,” 9 George Mason Law Review 843 (2001) Kovacic, William, “Toward a Domestic Competition Network” in Richard Epstein and Michael Greve, eds., Competition Laws in Conflict, American Enterprise Institute, 2004, chapter 13 Kovacic, William, “Antitrust in the O’Connor-Rehnquist Era. A View from inside the Supreme Court”, 20 Antitrust 21 (summer 2006) Kovacic, William, “Review of Antitrust Stories”, 4 Competition Policy International 241 (2008) Kovacic, William, “The Institutions of Antitrust Law: How Structure Shapes Substance”, 110 Michigan Law Review 1019 (2012) Kovacic, William, Robert Marshall, Leslie Marx and Steven Schulenberg, “Quantitative Analysis of Coordinated Effects”, 76 Antitrust Law Journal 397 (2009) Krattenmaker, Thomas and Robert Pitofsky, “Antitrust Merger Policy and the Reagan Administration”, 33 Antitrust Bulletin 211 (1988) Kraus, Elizabeth and Maria Coppola, “The International Competition Network: A Virtual Reality” (2004), http://www.ftc.gov/bc/international/docs/ krauscoppolaicn04.pdf Langen/Bunte, GWB, 11th edition (2010) Levy, Nicholas, “EU Merger Control: From Birth to Adolescence”, 26 World Competition 201 (2003) Levy, Nicholas, “The EU’s SIEC Test Five Years On: Has it Made a Difference?”, 6 European Competition Journal 211 (2010)

References  399   Levy, Nicholas, European Merger Control Law: A Guide to the Merger Regulation, 8th edition, LexisNexis Matthew Bender, 2011 Lu, Z. John and William Comanor, “Strategic Pricing of New Pharmaceuticals”, 80 Review of Economics and Statistics 108 (1998) Loewenheim/Meessen/Riesenkampff, GWB, 2nd edition (2009) Lübking, Johannes, “Some Reflections on the Application of the Substantive Test and of the Substantive Merger Guidelines”, presentation, EUI Competition Workshop, 12 November 2010 Lusk, Jayson, “Effects of Cheap Talk on Consumer Willingness-to-Pay for Golden Rice”, 85 American Journal of Agricultural Economics 840 (2003) Maillo, Jerónimo, “Understanding China’s Competition Law and Policy: Merger Control as a Case Study”, CEU San Pablo University Institute for European Studies Competition Policy Series Working Paper 30/2011, http://papers.ssrn. com/sol3/papers.cfm?abstract_id=1898190 Marquis, Mel, “Rules that Govern Rules: Evidence, Proof and Judicial Control in Competition Cases”, in Claus-Dieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Martinez, Ana Paula and Mariana Tavares de Araújo, “Brazil´s New Competition Law: Promising but Challenging”, CPI Antitrust Chronicle, November 2011 (2) McAfee, Preston, “Can Event Studies Detect Anticompetitive Mergers?”, 28 Economics Letters 199 (1988) McFadden, Daniel et al., “Statistical Analysis of Choice Experiments and Surveys”, 16(3–4) Marketing Letters 183 (2005) Montag, Frank, “Konvergenz bei internationalen Fusionen”, FIW Heft 188, p. 50 Nakano, Kaori, “The Recent Developments in Merger Investigation”, 320 Rippo to Chosa [Legislation and Investigation] 79 (2011) Neven, Damien, Robin Nuttall and Paul Seabright, Merger in Daylight: The Economics and Politics of European Merger Control, Centre for Economic Policy Research, 1993

400  Merger Control in European and Global Perspective Noland, Marcus and Howard Pack, Industrial Policy in an Era of Globalization: Lessons from Asia, Institute for International Economics, 2003 OECD, Recommendation of the Council on Merger Review (2005), http://www. oecd.org/dataoecd/3/41/40537528.pdf OECD Competition Committee, Procedural Fairness and Transparency: Key Points. See http://www.oecd.org/dataoecd/16/34/50235955.pdf Oldale, Alison, “Market Definition Is Dead. Long Live Market Definition?”, presentation, EUI Competition Workshop, 12 November 2010 Parisi, John, “Cooperation Among Competition Authorities in Merger Regulation”, 43 Cornell International Law Journal 55 (2010) Parker, Richard, Michael Antalics and Bilal Sayyed, “Shrinking from the ‘Third Rail’: Avoiding Direct Effects Analysis in Lundbeck”, 25 Antitrust 14 (spring 2011) Philipsen, Niels, Background Note, in OECD, Competition and Regulation in Auditing and Related Professions, DAF/COMP(2009)19 Reimers, Thilo, Probleme und Perspektiven der Internationalisierung des Wettbewerbsrechts, Nomos, 2007 Renard, François, “A Practitioner’s Look at Merger Control Remedies in China”, CPI Antitrust Chronicle, January 2012 (2) Rosch, J. Thomas, “A Peek Inside: One Commissioner’s Perspective on the Commission’s Roles as Prosecutor and Judge”, speech, Santa Fe, 3 July 2008, http://www.ftc.gov/speeches/rosch/080703nera.pdf Rosch, J. Thomas, “Terra Incognita: Vertical and Conglomerate Merger and Interlocking Directorate Law Enforcement in the United States”, speech, Hong Kong, 11 September 2009, http://www.ftc.gov/speeches/rosch/090911roschspe echunivhongkong.pdf Salop, Steve and Serge Moresi, Hearings Statement, 3 December 2009, http:// www.ftc.gov/os/comments/horizontalmergerguides/545095-00032.pdf Schroeder, Dirk and Andrea Edeler, “Reformbedarf in der deutschen Fusionskontrolle”, ZWeR 364 (2008) Schumpeter, Joseph, Capitalism, Socialism and Democracy (first edition 1942), Routledge, 1994

References  401   Scott, Sheridan, “Canadian Perspectives on the Role of Comity in Competition Law Enforcement in a Globalized World – To Defer, or Not To Defer? Is that the question?”, speech, Washington, D.C., 29 March 2006, http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/02049.html29 Seldeslachts, Jo, Joseph Clougherty and Pedro Pita Barros, “Settle for Now but Block for Tomorrow: The Deterrence Effects of Merger Policy Tools”, 52 The Journal of Law and Economics 607 (2009) Shapiro, Carl, “Mergers with Differentiated Products”, 10 Antitrust 23 (spring 1996) Shapiro, Carl, “The 2010 Horizontal Merger Guidelines: From Hedgehog to Fox in Forty Years”, 77 Antitrust Law Journal 49 (2010) Shiraishi, Tadashi, Antitrust Law Lecture, 5th edition, Yuhikaku Publishing House, 2010 (in Japanese) Sims, Joe et al., “Merger Process Reform: A Sisyphean Journey?”, 23 Antitrust 60 (Spring 2009) Sims, Joe and Deborah Platt Majoras, “The Effect of Twenty Years of HartScott-Rodino on Merger Practice: A Case Study of Unintended Consequences Applied to Antitrust Legislation”, 65 Antitrust Law Journal 865 (1997) Smith, Blackwell, “Precedent, Public Policy and Predictability”, 46 Georgetown Law Journal 633 (1957) Sohn, Michael, “Pricing of Differentiated Products in the 2010 Horizontal Merger Guidelines: Some Unanswered Questions”, in Barry Hawk, ed., International Antitrust Law and Policy: Fordham Competition Law 2010, Juris Publishing, 2011, chapter 8 Soloway, Julie and Cassandra Brown, “Canada”, in Ilene Knable Gotts, ed., The Merger Control Review, 2010, pp. 68 et seq. Sørgard, Lars, “Optimal Merger Policy: Enforcement vs. Deterrence”, 57 Journal of Industrial Economics 438 (2009) Stark, Charles, “Improving Bilateral Antitrust Cooperation”, speech, 23 June 2000, http://www.justice.gov/atr/public/speeches/5075.pdf Stillman, Robert, “Examining antitrust policy towards horizontal mergers”, 11 Journal of Financial Economics 225 (1983)

402  Merger Control in European and Global Perspective Sunshine, Steven and David Wales, Statement at AMC Merger Enforcement Hearing (17 November 2005), http://govinfo.library.unt.edu/amc/commission_ hearings/pdf/Sunshine_Statement.pdf Sutton, John, Sunk Costs and Market Structure: Price Competition, Advertising and the Evolution of Concentration, MIT Press, 1991 Sutton, John, Technology and Market Structure – Theory and History, MIT Press, 1998 Takeshima,Kazuhiko, “Cooperation and Alliance between Japan and EU in competition policy”, Europe 257 (Spring 2009) Takeshima, Kazuhiko, “Toward more Effective and Efficient Cooperation in Merger Review”, speech, Rio de Janeiro, 19 April 2012, http://www.jftc.go.jp/ en/policy_enforcement/speeches/pdf/2MWG_Plenary_Script.pdf Takigawa, Toshiaki, “The Prospect of Antitrust Law and Policy in the TwentyFirst Century: In Reference to the Japanese Antimonopoly Law and Japan Fair Trade Commission”, 1 Washington University Global Legal Studies Law Review 275 (2002) Ulen, Thomas, “The Uneasy Case for Competition Law and Regulation as Decisive Factors in Development: Some Lessons from China”, in Michael Faure and Xinzhu Zhang, eds., Competition Policy and Regulation: Recent Developments in China, the US and Europe, Edward Elgar Publishing, 2011 Varney, Christine, Answers to Sen. Schumer before Senate Judiciary Committee (10 March 2009), http://judiciary.senate.gov/nominations/111thCongressExecu tiveNominations/upload/Varney-QFRs.pdf Varney, Christine, “Procedural Fairness”, speech, Fiesole, 12 September 2009, http://www.justice.gov/atr/public/speeches/249974.pdf Varney, Christine, “An Update on the Review of the Horizontal Merger Guidelines”, speech, 26 January 2010, Washington, D.C., http://www.justice. gov/atr/public/speeches/254577.pdf Varney, Christine, “Coordinated Remedies: Convergence, Cooperation, and the Role of Transparency”, speech, Paris, 15 February 2010, http://www.justice. gov/atr/public/speeches/255189.htm Varney, Christine, “International Cooperation: Preparing for the Future”, speech, Washington, D.C., 21 September 2010, p. 16, http://www.justice.gov/atr/ public/speeches/262606.pdf

References  403   Venit, James, “Human All Too Human: The Gathering and Assessment of Evidence and the Appropriate Standard of Proof and Judicial Review in Commission Enforcement Proceedings Applying Articles 81 and 82”, in ClausDieter Ehlermann and Mel Marquis, eds., European Competition Law Annual 2009: Evaluation of Evidence and its Judicial Review, Hart Publishing, 2011 Vesterdorf, Bo, “Judicial Review in EC Commission Law: Reflections on the Role of the Community Courts in the EC System of Competition Law Enforcement,” 1 Global Competition Policy 1 (2005) Vesterdorf, Bo, “Standard of Proof in Merger Cases: Reflections in the Light of Recent Case Law of the Community Courts”, 1 European Competition Journal 3 (2005) Vesterdorf, Bo, “The European Courts’ Case Law in Merger Control”, in Abel Mateus and Teresa Moreira, eds., Competition Law and Economics: Advances in Competition Policy and Antitrust Enforcement, Kluwer Law International, 2007, chapter 13 Walters, Chris, “Diversion Ahead! Approximating diversion ratios for retail chain mergers”, Competition Commission Occasional Paper (November 2007), http:// www.competition-commission.org.uk/our_role/analysis/diversion_ratios.pdf Wiedemann, Gerhard, Handbuch des Kartellrechts, 2nd edition (2008) Weitbrecht, Andreas, “Völkerrecht und Kollisionsrecht in der deutschen Fusionskontrolle, Zur Untersagung von Auslandszusammenschlüssen durch das Bundeskartellamt”, in Festschrift Birk (70. Geburtstag), (Tübingen, 2008) Whish, Richard, Competition Law, 6th edition, OUP, 2009 Wils, Wouter, “The Role of the Hearing Officer in Competition Proceedings before the European Commission”, 35 World Competition 431 (2012) Zhang, Xinzhu and Vanessa Yanhua Zhang, “Chinese Merger Control: Patterns and Implications,” 6 Journal of Competition Law and Economics 477 (2010) Zhang, Xinzhu and Vanessa Yanhua Zhang, “Revisiting China’s Merger Control: Where Are We Going after Three-Year Milestone?” CASS Working Paper, 2012

404  Merger Control in European and Global Perspective II. Legislation, proposed legislation and soft law A. The European Union European legislation Commission Proposal for a Regulation of the Council of Ministers on the Control of Concentrations between Undertakings, 1973 OJ C92/1 Council Regulation (EC) No. 139/2004 of 20 January 2004 on the control of concentrations between undertakings, 2004 OJ L24/1 Commission Regulation (EC) No. 802/2004 of 7 April 2004 implementing Council Regulation (EC) No. 139/2004, 2004 OJ L133/1 Commission Decision of 23 May 2001 on the terms of reference of hearing officers in certain competition proceedings, 2001 OJ L162/21 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 Decision 2011/695/EU of the President of the European Commission of 13 October 2011 on the function and terms of reference of the hearing officer in certain competition proceedings, 2011 OJ L275/29 European soft law 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 assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings, 2004 OJ C31/5 Commission, Guidelines on the method of setting fines imposed pursuant to Article 23(2)(a) of Regulation No 1/2003, 2006 OJ C210/02 Commission, Memorandum on the concentration of enterprises in the common market, EEC Competition Series Study No 3, 1966; 2 Revue trimestrielle de droit européen 651 (1966) Commission Notice on best practices for the conduct of proceedings concerning Articles 101 and 102 TFEU, 2011 OJ C308/6

References  405   Commission, Notice on Case Referral in respect of concentrations, 2005 OJ C56/02 DG Competition, Best Practices on the conduct of EC merger control proceedings (2004), http://ec.europa.eu/competition/mergers/legislation/proceedings.pdf B. National jurisdictions Australia ACCC, Merger review process guidelines (2006, addendum 2011), http://www. accc.gov.au/content/index.phtml/itemId/740765 Canada Competition Bureau, Merger Review Process Guidelines (2009), http://www. competitionbureau.gc.ca/eic/site/cb-bc.nsf/eng/03128.html Competition Bureau, Fees and Service Standards Handbook for Mergers and MergerRelated Matters (“Merger Handbook”) (2010), http://www.competitionbureau. gc.ca/eic/site/cb-bc.nsf/vwapj/Final-Fees-Service-Handbook-Oct-2010-e. pdf/$FILE/Final-Fees-Service-Handbook-Oct-2010-e.pdf France Lignes directrices relatives au contrôle des concentrations – Procédure et analyse”, http://www.economie.gouv.fr/directions_services/dgccrf/concurrence/ concentrations/lignesdirectrices_2007_anglais.pdf Lignes directrices relatives au contrôle des concentrations, 16 December 2009, http://www.autoritedelaconcurrence.fr/doc/ld_mergers_final.pdf(English-language version) Germany Bundeskartellamt, Guidance on Substantive Merger Control (2012), http://www. bundeskartellamt.de/wEnglisch/download/pdf/Merkblaetter/2012-03-29_ Guidance_final_neu.pdf Ireland Ireland Competition Authority, Mergers and Acquisition Procedures, http://www. tca.ie/images/uploaded/documents/Merger_Procedures.pdf

406  Merger Control in European and Global Perspective Ireland Competition Authority, Access to the File in Merger Cases, http://www. tca.ie/images/uploaded/documents/Merger_File_Access_Procedures.pdf Japan Japan Fair Trade Commission, Guidelines to Application of the Antimonopoly Act Concerning Review of Business Combination (28 March 2007), http://www. jftc.go.jp/e-page/pressreleases/2007/March/RevisedMergerGuidelines.pdf Japan Fair Trade Commission, proposed Guidelines to Application of The Antimonopoly Act Concerning Review of Business Combination (2011), http:// www.jftc.go.jp/e-page/legislation/ama/RevisedMergerGuidelines.pdf Korea KFTC, 2010 annual report to the OECD Competition Committee (2011), DAF/ COMP/AR(2011)29, http://www.oecd.org/officialdocuments/publicdisplaydoc umentpdf/?cote=DAF/COMP/AR(2011)29&docLanguage=En Monopoly Regulation and Fair Trade Act (2011), http://eng.ftc.go.kr/files/static/ Legal_Authority/Monopoly%20Regulation%20and%20Fair%20Trade%20 Act_mar%2014%202012.pdf Netherlands NMa, Best Practices on the Conduct of Concentration Control Proceedings, http://www.nmanet.nl/Images/90106%20Spelregels%20voor%20 concentratiezaken%20-%20vertaling_tcm16-135884.pdf Norway Konkurransetilsynet, Best Practices on the conduct of merger control proceedings (2009), http://www.konkurransetilsynet.no/Global/English/Merger_Control_ Proceedings_Best_Practices.pdf Romania Competition Council, Guidelines on the rules for access to the Competition Council file in cases pursuant to Art. 5 and 6 of Competition Law No. 21/1996 and in the economic concentration cases”, http://www.competition.ro/documente/ Instructiuni%20acces%20dosar_14852en.pdf

References  407   Singapore Competition Commission, revised Guidelines on Merger Procedures (2012), http://www.ccs.gov.sg/content/dam/ccs/PDFs/CCSGuidelines/Revised%20 Merger%20Guidelines%20Procedures%202012.pdf United Kingdom UK CC and OFT, Good practice in the design and presentation of consumer survey evidence in merger inquiries, CC2com1 and OFT1230 (March 2011), http:// www.oft.gov.uk/shared_oft/consultations/merger-inquiries/Good-practiceguide.pdf UK CC and OFT, Merger Assessment Guidelines, CC2 (revised) and OFT1254 (September 2010), http://www.oft.gov.uk/shared_oft/mergers/642749/ OFT1254.pdf Competition Commission, Rules of Procedure (2006), http://www.competitioncommission.org.uk/rep_pub/rules_and_guide/pdf/cc1.pdf United States U.S. DOJ and FTC, Antitrust Enforcement Guidelines for International Operations (April 1995), http://www.justice.gov/atr/public/guidelines/internat.htm US DOJ and FTC, Horizontal Merger Guidelines (August 2010), http://www. justice.gov/atr/public/guidelines/hmg-2010.html US-EU Merger Working Group, Best Practices on Cooperation in Merger Investigations (2011), http://ec.europa.eu/competition/mergers/legislation/ best_practices_2011_en.pdf III. Official and unofficial reports, newsletters, studies, papers, etc. A. The European Union Commission, Competition Policy Newsletter Devai, Rita et al., “Merger Case M.5421 Panasonic/Sanyo – Batteries included or ‘lost in translation’?”, Competition Policy Newsletter, No. 1 – 2010, pp. 60 et seq.

408  Merger Control in European and Global Perspective Commission, staff reports Staff Working Paper accompanying the Communication from the Commission to the Council, Report on the Functioning of Regulation No 139/2004 (COM(2009) 281 final), 30 June 2009 B. National jurisdictions United Kingdom Accent and Rand Europe, Review of Stated Preference and Willingness to Pay Methods (April 2010), http://www.competition-commission.org.uk/our_role/ analysis/summary_and_report_combined.pdf Deloitte and Touche LLP, The deterrent effect of competition enforcement by the OFT, OFT962 (November 2007), http://www.oft.gov.uk/shared_oft/reports/ Evaluating-OFTs-work/oft962.pdf United States Antitrust Modernization Commission, Report and Recommendations, April 2007, http://govinfo.library.unt.edu/amc/report_recommendation/amc_final_report.pdf ICPAC Final Report to the DOJ, (2000), http://www.justice.gov/atr/icpac/ finalreport.html Kovacic, William et al., The Federal Trade Commission at 100: Into Our 2d Century (January 2009), http://www.ftc.gov/os/2009/01/ftc100rpt.pdf C. International organizations International Competition Network ICN, A Statement of Achievements Through April 2010, http://www.icn-istanbul. org/Upload/Materials/Others/StatementOfAchievements.pdf ICN, Information Requirements for Merger Notification, Report for the ICN Annual Conference Zurich, Switzerland (2009), http://www. internationalcompetitionnetwork.org/uploads/library/doc328.pdf ICN Recommended Practices for Merger Notification Procedures, http://www. internationalcompetitionnetwork.org/uploads/library/doc588.pdf

References  409   ICN, Setting Notification Thresholds for Merger Review, Report to the ICN Annual Conference Kyoto, Japan (2008), http://www.internationalcompetitionnetwork. org/uploads/library/doc326.pdf ICN, Merger Working Group, Guiding Principles for Merger Notification and Review, http://www.internationalcompetitionnetwork.org/uploads/library/ doc591.pdf Organisation for Economic Co-operation and Development Business and Industry Advisory Committee to the OECD, “Recommended Framework for Best Practices in International Merger Control Procedures” (4 October 2001), http://www.biac.org/statements/comp/BIAC-ICCMergerPaper. pdf OECD, Competition and Regulation in Auditing and Related Professions (2009), DAF/COMP(2009)19, http://www.oecd.org/dataoecd/55/46/44284186.pdf OCED, Global Forum on Competition, 10th edition, Paris, 17–18 February 2011, http://www.oecd.org/document/52/0,3746,en_40382599_40393118_46076788 _1_1_1_1,00.html OECD, Policy Brief, “Competition Law and Policy in Brazil”, September 2005 OECD, Recommendation of the Council on Merger Review (2005), http://www. oecd.org/dataoecd/3/41/40537528.pdf

Table of Cases I. Judgments of the European Courts European Court of Human Rights Kokkinakis v Greece, (1993) 17 EHRR 397

153

Courts of the European Union Joined Cases 89, 104, 114, 116, 117 and 125 to 129/85, xxxvii, 197, 275 A. Ahlström Osakeyhtiö and others v Commission (Wood Pulp) [1988] ECR 5193 Joined Cases C-89/85, C-104/85, C-114/85, C-116/85, C-117/85 119 and C-125/85 to C-129/85 A. Ahlström Osakeyhtiö and others v Commission (Wood Pulp II) [1993] ECR I-1307 C-342/99, Airtours plc v Commission [2002] ECR II-2585 5, 15, 81, 88, 100, 118, 123, 126, 224, 376 Case C-441/07 P, Alrosa v Commission [2010] ECR I-5949

xxix, xxxi, 123

Case T-321/05 Astra Zeneca v Commission [2010] ECR II-2805, xxix, 116 appeals dismissed: Case C-457/10 P, judgment of the ECJ of 6 December 2012, not yet reported Case 142/84, BAT and RJ Reynolds v Commission [1987] ECR 4487

144

Case C-413/06 P, Bertelsmann AG and Sony Corporation xxxii, 5, 85, 122, 138 of America v Independent Music Publishers and Labels Association (Impala) [2008] ECR I-4951 Case T-282/02, Cementbouw Handel & Industrie BV v Commission 140 [2006] ECR II-319 Case C-282/04, Commission v Netherlands [2006] ECR I-9141

148

Case C-112/05, Commission v Germany [2007] ECR I-8995

148

Case C-196/07, Commission v Spain [2008] ECR I-41

149

Case C-12/03 P, Commission v Tetra Laval [2005] ECR I-987 xxix, xxxii, 81, 83, 117, 122, 124, 125, 126, 127, 139

412  Merger Control in European and Global Perspective Case C-13/03 P, Commission v Tetra Laval [2005] ECR I-1113

338

Case C-98/01, Commission v UK [2003] ECR I-4641

148

Joined Cases 29/83 and 30/83, Compagnie Royale Asturienne des xxix, 83 Mines SA and Rheinzink GmbH v Commission [1984] ECR 1679 Case 6/72, Continental Can v Commission [1973] ECR 125

144, 146

Case C-280/08 P, Deutsche Telekom AG v Commission xxix, 202 [2010] ECR I-9555 Case T-87/05, EDP v Commission [2005] ECR II-3745

139

Case T-417/05, Endesa v Commission [2006] ECR II-2533

154

Case T-102/96, Gencor v Commission [1999] ECR II-753

150, 215, 275

Case T-210/01, General Electric v Commission [2005] ECR II-5575 xxix, 139 Case T-209/01, Honeywell v Commission [2005] ECR II-5527

119

Case T-290/94, Kaysersberg v Commission [1997] II-2137

119

Case T-33/97, Kesko Oy v Commission [1999] ECR II-3775

100

Case C-272/09 P, KME Germany AG and others v Commission, xxix, 77 judgment of the ECJ of 8 December 2011, not yet reported Case T-310/00, MCI v Commission [2004] ECR II-3253

117

Case C-42/01, Portugal v Commission [2004] ECR I-6079

149

Case 42/84, Remia BV and others v Commission [1985] ECR 2545

xxix, 86

Case T-342/07, Ryanair v Commission [2010] ECR II-3457 xxix, 6, 34, 81, 116 Case T-310/01, Schneider Electric v Commission 81, 118, 119, 337, 376 [2002] ECR II-4071 Case T-282/06, Sun Chemical Group v Commission [2007] ECR II-2149 6, 81 Case T-5/02, Tetra Laval v Commission [2002] ECR II-4381

81, 117, 376

Table of Cases  413   Case T-83/91, Tetra Pak International SA v Commission 78 [1994] ECR-II 755, upheld: Case 333/94 P, [1996] ECR I-5951 Case T-155/06, Tomra Systems ASA v Commission [2010] ECR II-4361, 102 upheld: Case C-549/10, judgment of the ECJ of 19 April 2012, not yet reported II. Decisions of the European Commission Merger decisions M.4980, ABF/GBI Business (23 September 2008) M.4985, BHP Billiton/Rio Tinto (26 November 2008)

5 15, 69, 179, 200, 206, 211, 237, 301

M.3669, Blackstone Group/NHP Plc (2 February 2005)

162

M.1879, Boeing/Hughes (27 September 2000)

389

M.5741, CDC/Veolia Environment/Transdev/Veolia Transport 240 (12 August 2010) M.5669, Cisco/Tandberg (29 March 2010)

xxxiii, 184, 239

M.5860, Cooper/Danaher/JV (2 June 2010)

389

M.5814, CVC/Univar Europe/Eurochem (16 July 2010)

240

M.3868, DONG/Elsam/Energi E2 (14 March 2006)

66, 70

M.1140, Dresser/Halliburton (6 July 1998)

134, 137, 138

M.5549, EDF/Segebel (12 November 2009)

5

M.4994, Electrabel /Compagnie Nationale du Rhône (10 June 2009)

152

M.5518, Fiat/Chrysler (24 July 2009)

181

M.4844, Fortis/ABN Amro Assets (3 October 2007) M.5046, Friesland/Campina (17 December 2008)

5 65, 66, 69

414  Merger Control in European and Global Perspective M.2220, General Electric/Honeywell (3 July 2001)

200, 385

M.4731, Google/Double Click (11 March 2008)

5, 69

M.5747, Iberia/British Airways (14 July 2010)

330

M.5406, IPIC/MAN Ferrostaal AG (13 March 2009)

6

M. 4874, Itema Holding/Barcovision Division (4 August 2008) M.5141, KLM/MartinAir (17 December 2008)

66, 69 11, 33, 36, 69

M.5644, Kraft Foods/Cadbury (6 January 2010)

4, 338

M.5502, Merck/Schering-Plough (22 October 2009)

263

M.5727, Microsoft/Yahoo! Search Business (18 February 2010)

181, 239

M.2876, Newscorp/Telepiù (2 April 2003)

198

M.4942, Nokia/Navteq (2 July 2008)

5, 69

M.3216, Oracle/PeopleSoft (26 October 2004) M.5529, Oracle/Sun Microsystems (21 January 2010)

138 139, 181, 236, 386

M.5421, Panasonic Corporation/Sanyo Electric (29 September 2009) M.5748, Prisa/Telefonica/Telecinco/Digital+ (11 March 2010)

198, 206, 237 240

M.5828, Procter & Gamble/Sara Lee (Air Care) (17 June 2010)

240

M.5881, ProSiebenSat.1 Media/RTL Interactive/JV (24 September 2010)

240

M.5535, Renesas Technology/NEC Electronics (2 December 2009)

290

M.4439, Ryanair/Aer Lingus (27 June 2007)

12, 69

M.1543, Sanofi/Synthélabo (2 July 1999)

139

M.4691, Schering-Plough/Organon Biosciences (11 October 2007)

389

M.5969, SCJ/Sara Lee (9 May 2011)

240

Table of Cases  415   M.6214, Seagate Technology/The HDD business of Samsung Electronics 294 (19 November 2011) M.4139, Sony/NEC/JV (31 March 2006)

290

M.5675, Syngenta/Monsanto (Sunflower Seeds Business) 240 (17 November 2011) M.5776, Telecinco/Cuatro (11 March 2010)

240

M.4465, Thrane & Thrane/Nera (21 March 2007)

163

M.3916, T-mobile Austria/Tele.ring (26 April 2006) M.5650, T-Mobile/Orange (1 March 2010) M.4854, TomTom/Tele Atlas (14 May 2008)

4, 70 181, 210, 241 5, 69

M.6203, Western Digital Ireland/Viviti Technologies (23 November 2011) 294 Non-merger decisions Commission Decision of 28 January 2009 in Case COMP/39.406 – 280 Marine Hoses Commission Decision of 16 December 2009 in Case COMP/39.530 – 184 Microsoft (Tying)

Commission Decision of 29 March 2006 in Case COMP/38.113 — 102 Prokent/Tomra III. US cases (courts and the agencies) American Banana Co. v. United Fruit Co., 213 U.S. 347 (1909)

xxxvii, 275

Boeing/McDonnell Douglas, FTC File No. 971-0051 (1 July 1997), 188 http://www.ftc.gov/opa/1997/07/boeingsta.shtm Cisco/Tandberg, DOJ Press Release of 29 March 2010, xxxiii, 134, 184 http://www.justice.gov/atr/public/press_releases/2010/257173.htm Daubert v. Merrell Dow Pharmaceuticals, 509 U.S. 579 (1993)

103

416  Merger Control in European and Global Perspective Empagran S.A. v. F. Hoffman-La Roche Ltd., 417 F. 3d 1267, 279 (D.C. Cir. 2005), cert. denied, 546 U.S. 1092 (2006)

Exxon/Mobile, FTC File No. 991-0077 (30 November 1999), 188, 200 http://www.ftc.gov/os/1999/11/exxonmobilmaj.pdf F. Hoffmann-La Roche Ltd. v. Empagran S.A., 542 U.S. 155 (2004)

193, 279

FTC v. Whole Foods Market, Inc., 533 F.3d 869 (D.C. Cir. 2008)

169

FTC v. Whole Foods Mkt., Inc., 548 F.3d 1028 (D.C. Cir. 2008)

19

FTC v. Lundbeck, No. 08-6379, 2010 WL 3810015 xx, 20, 46, 47 (D. Minn. 2010), aff’d, 650 F.3d 1236 (8th Cir. 2011) Malone/Discovery Holding, FTC File No. 0810219 (23 June 2009) 152 Minn-Chem, Inc. v. Agrium, Inc., No. 10-1712 (7th Cir. 2012) (en banc) 198 United States v. Aluminum Co. of America, 148 F.2d 416 (2d Cir. 1945)

275

United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001), 21 remanded to 231 F. Supp. 2d 144 (D.D.C. 2002), affd sub nom. Massachusetts v. Microsoft Corp., 373 F.3d 1199 (D.C. Cir. 2004) United States v. Oracle Corp., 331 F. Supp. 2d 1098 (N.D. Cal. 2004) xxii, 18, 56

United States v. Ticketmaster Entertainment, Inc., 184 No. 1-10-cv-00139 (D.D.C. 2010) Whirlpool/Maytag, DOJ Press Release of 29 March 2006, 188 http://www.justice.gov/atr/public/press-releases/2006/215326.htm IV. Other national jurisdictions China

MOFCOM, Alpha V/Savio Decision, Notice No. 73 [2011], 309 31 October 2011, http://fldj.mofcom.gov.cn/aarticle/zcfb/201111/ 20111107809156.html

Table of Cases  417   MOFCOM, Coca-Cola/Huiyuan Decision, Notice No. 22 xlix, 199, 310, 321 [2009], 18 March 2009, http://fldj.mofcom.gov.cn/aarticle/ ztxx/200903/20090306108494.html MOFCOM, GE/Shenhua Decision, Notice No. 74 [2011], li, 309, 310, 311 10 November 2011, http://fldj.mofcom.gov.cn/aarticle/zcfb/ 201111/20111107824342.html MOFCOM, GM/Delphi Decision, Notice No. 76 [2009], li, 309, 310, 311, 313 28 September 2009, http://fldj.mofcom.gov.cn/aarticle/ ztxx/200909/20090906540211.html MOFCOM, Novartis/Alcon Decision, Notice No. 53 269, 306, 308, 309, 313 [2010], 13 August 2010, http://fldj.mofcom.gov.cn/ aarticle/ztxx/201008/20100807080639.html MOFCOM, Panasonic/Sanyo Decision, Notice No. 82 [2009], xli, 306, 309 30 October 2009, http://fldj.mofcom.gov.cn/aarticle/ztxx/ 200910/20091006593175.html MOFCOM, Pfizer/Wyeth Decision, Notice No. 77 [2009], 306, 307, 309 29 September 2009, http://fldj.mofcom.gov.cn/aarticle/ztxx/ 200909/20090906541443.html

MOFCOM, Seagate/Samsung Decision, Notice No. 90 [2011], l, 313 12 December 2011, http://fldj.mofcom.gov.cn/aarticle/ zcfb/201112/20111207874274.html MOFCOM, Uralkali/Silvinit Decision, Notice No. 33 [2011], 308, 309, 313 2 June 2011, http://fldj.mofcom.gov.cn/aarticle/zcfb/ 201106/20110607583288.html Germany Bundeskartellamt Decision of 17 August 2004, B7-65/04, WuW DE-V 965 - 134, 223 General Electric/InVision Decision of 11 April 2007, WuW, DE-V 1365 - Phonak/ReSound 179, 215, 216, 217

418  Merger Control in European and Global Perspective Courts BGH, judgment of 11 November 2008, KVR 60/07, WuW, DE-R, 2451 - 224 E.ON/Stadtwerke Eschwege BGH, judgment of 12 July 1973, KRB 2/72, WuW/E BGH 1276, 215 1279 – Ölfeldrohre BGH, judgment of 29 May 1979, KVR 2/78, WuW/E BGH 1613, 1614 – Organische Pigmente

215



BGH, judgment of 20 April 2010, KVR 1/09, WuW, DE-R 2905 - 215 Phonak/GN Sto­re

BGH, judgment of 25 September 2007, KVR 19/07, WuW/E DE-R 2133 – 215 Sulzer/Kelmix OLG Düsseldorf, judgment of 26 November 2008, VI-Kart 8/07 179, 215, 218 (V), WuW, DE-R 2477 - Phonak II (reversed on other grounds: BGH, judgment of 20 April 2010, KVR 1/09, WuW, DE-R 2905 Phonak/GN Sto­re) Italy AGCM, Procedure n. 16809 - C 8094 LIDL ITALIA/ RAMI 152 D’AZIENDA, 10 March 2007 Japan BHP Billiton/Rio Tinto, JFTC Press Release of 18 October 2010, 206 http://www.jftc.go.jp/en/pressreleases/archives/individual-000016.html Cathode Ray Tubes, JFTC Press Release of 7 October 2009, 193 http://www.jftc.go.jp/en/pressreleases/archives/individual-000037.html and JFTC Press Release of 29 March 2010, http://www.jftc.go.jp/en/ pressreleases/archives/individual-000136.html Microsoft, Decision of 16 September 2008, Shinketsushu vol.55, p. 380

276

Nordion, Decision of 3 September 1998, Shinketsushu vol.45, p. 148

276

NSC/Sumitomo Metal, Decision of 14 December 2011, http://www.jftc.go.jp/pressrelease/11.december/11121403.pdf

291, 292, 293

Table of Cases  419   Qualcomm, Order of 28 September 2009, http://www.jftc.go.jp/en/ 276 pressreleases/archives/individual-000038.html Seagate Technology International/Samsung, Decision of 28 December 294 2011, http://www.jftc.go.jp/pressrelease/11.december/111228.pdf

Sony/NEC, (2005), http://www.jftc.go.jp/en/pressreleases/uploads/2007- 290 Mar-28_1_1.pdf

Western Digital/Viviti Technologies, Decision of 28 December 2011, 293 http://www.jftc.go.jp/pressrelease/11.december/111228.pdf Korea eBay/Gmarket, KFTC Decision 2009-145, 25 June 2009

301

United Kingdom Handyside v UK, [1974] 17 YB 228

153

Competition Commission NEG/Thameslink, CC Press Release 84/05 of 22 December 2005

12

Somerfield/Morrison, CC Press Release 53/05 of 2 September 2005 33, 34, 37, 38 Office of Fair Trading ME/1544/05 – Blackstone Group/NHP plc, http://www.oft.gov.uk/ 162 OFTwork/mergers/decisions/2005/blackstone ME/1024/04 – General Electric/Invision Technologies, 134 http://www.oft.gov.uk/OFTwork/mergers/decisions/2005/general-electric

ME/2684/06 – Thrane and Thrane A/S /Nera ASA, http://www.oft.gov.uk/ 163 OFTwork/mergers/decisions/2006/thrane