Comparative Competition Law 1849804192, 9781849804196

Comparative Competition Law examines the key global issues facing competition law and policy. This volume s specially co

798 35 4MB

English Pages 528 [529] Year 2016

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
Contents
List of contributors
PART I: INTRODUCTION AND OVERVIEW
1. Themes • John Duns, Arlen Duke and Brendan Sweeney
2. The globalization of competition law: Yes or no? • Leela Cejnar and Rachel Burgess
PART II: THE SUBSTANTIVE LAW
3. Defining and proving markets and market power • Rhonda L. Smith
4. Anti-competitive agreements: The meaning of ‘agreement’ • George A. Hay
5. Anti-competitive agreements: The range of conduct caught • John Duns
6. Understanding market power • Alexandra Merrett
7. Antitrust treatment of intellectual property rights • Michael A. Carrier
8. Current issues in merger law • Julie Clarke
9. Vertical conduct: Non-price restraints • John Duns
10. Vertical conduct: Resale price maintenance • Eugène Buttigieg
PART III: ENFORCEMENT AND SANCTIONS
11. Public enforcement • Arlen Duke
12. Criminalizing cartels: A global trend? • Gregory C. Shaffer, Nathaniel H. Nesbitt and Spencer Weber Waller
13. International governance of competition and the problem of extraterritorial jurisdiction • Brendan Sweeney
14. Private antitrust enforcement: Comparative and policy considerations • Daniel A. Crane, Keith Klovers and Adam Speegle
PART IV: COMPETITION LAW IN SELECTED JURISDICTIONS
15. Competition lawin Japan • Simon Vande Walle and Tadashi Shiraishi
16. Competition lawin China • Wentong Zheng
17. Latin American antitrust lawand policy: An overview of three jurisdictions – Brazil, Chile and Colombia • Javier Tapia and Alexandre Ditzel Faraco
Index
Recommend Papers

Comparative Competition Law
 1849804192, 9781849804196

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

COMPARATIVE COMPETITION LAW

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

RESEARCH HANDBOOKS IN COMPARATIVE LAW Series Editors: Francesco Parisi, Oppenheimer Wolff and Donnelly Professor of Law, University of Minnesota, USA and Professor of Economics, University of Bologna, Italy and Tom Ginsburg, Professor of Law, University of Chicago, USA The volumes in this series offer high-level discussion and analysis on particular aspects of legal systems and the law. Well-known scholars edit each handbook and bring together accessible yet sophisticated contributions from an international cast of top researchers. The first series of its kind to cover a wide range of comparative issues so comprehensively, this is an indispensable resource for students and scholars alike. Titles in this series include: Comparative Administrative Law Edited by Susan Rose-Ackerman and Peter L. Lindseth Comparative Constitutional Law Edited by Tom Ginsburg and Rosalind Dixon Methods of Comparative Law Edited by Pier Giuseppe Monateri Comparative Law and Society Edited by David S. Clark Comparative Labor Law Edited by Matthew W. Finkin and Guy Mundlak Comparative Tort Law Edited by Mauro Bussani and Anthony Sebok Comparative Competition Law Edited by John Duns, Arlen Duke and Brendan Sweeney

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Comparative Competition Law

Edited by

John Duns Associate Professor, Monash University Faculty of Law, Australia

Arlen Duke Senior Lecturer, Melbourne Law School, Australia

Brendan Sweeney Adjunct Associate Professor of Law, Monash University, Australia

RESEARCH HANDBOOKS IN COMPARATIVE LAW

Cheltenham, UK

Columns Design XML Ltd

/

+

Northampton, MA, USA

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

© The Editors and Contributors Severally 2015 All rights reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical or photocopying, recording, or otherwise without the prior permission of the publisher. Published by Edward Elgar Publishing Limited The Lypiatts 15 Lansdown Road Cheltenham Glos GL50 2JA UK Edward Elgar Publishing, Inc. William Pratt House 9 Dewey Court Northampton Massachusetts 01060 USA

A catalogue record for this book is available from the British Library Library of Congress Control Number: 2015941464 This book is available electronically in the Law subject collection DOI 10.4337/9781785362576

ISBN 978 1 84980 419 6 (cased) ISBN 978 1 78536 257 6 (eBook)

02

Typeset by Columns Design XML Ltd, Reading

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Contents

List of contributors PART I

vii

INTRODUCTION AND OVERVIEW

1. Themes John Duns, Arlen Duke and Brendan Sweeney 2. The globalization of competition law:Yes or no? Leela Cejnar and Rachel Burgess PART II

3 9

THE SUBSTANTIVE LAW

3. Defining and proving markets and market power Rhonda L. Smith 4. Anti-competitive agreements: The meaning of ‘agreement’ George A. Hay 5. Anti-competitive agreements: The range of conduct caught John Duns 6. Understanding market power Alexandra Merrett 7. Antitrust treatment of intellectual property rights Michael A. Carrier 8. Current issues in merger law Julie Clarke 9. Vertical conduct: Non-price restraints John Duns 10. Vertical conduct: Resale price maintenance Eugène Buttigieg PART III

27 56 79 109 141 171 219 245

ENFORCEMENT AND SANCTIONS

11. Public enforcement Arlen Duke 12. Criminalizing cartels: A global trend? Gregory C. Shaffer, Nathaniel H. Nesbitt and Spencer Weber Waller

271 301

v

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

vi Comparative competition law

13. International governance of competition and the problem of extraterritorial jurisdiction Brendan Sweeney 14. Private antitrust enforcement: Comparative and policy considerations Daniel A. Crane, Keith Klovers and Adam Speegle PART IV

345

384

COMPETITION LAW IN SELECTED JURISDICTIONS

15. Competition law in Japan Simon Vande Walle and Tadashi Shiraishi 16. Competition law in China Wentong Zheng 17. Latin American antitrust law and policy: An overview of three jurisdictions – Brazil, Chile and Colombia Javier Tapia and Alexandre Ditzel Faraco

415

Index

501

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

443

472

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Contributors

Rachel Burgess, Consultant and Director, Amicitia Sdn Bhd, Malaysia Eugène Buttigieg, Judge, General Court of the European Union, and Associate Professor, Department of European and Comparative Law, University of Malta, Malta Michael A. Carrier, Distinguished Professor, Rutgers School of Law, USA Leela Cejnar, Senior Lecturer, University of New South Wales Law School, Australia Julie Clarke, Associate Professor, Deakin University School of Law, Australia Daniel A. Crane, Frederick P Furth Sr Professor of Law, University of Michigan Law School, USA Alexandre Ditzel Faraco, Partner, Levy & Salomão, Brazil and Professor, Federal University of Paraná, Brazil Arlen Duke, Senior Lecturer, University of Melbourne Law School, Australia John Duns, Associate Professor, Monash University Faculty of Law, Australia George A. Hay, Charles Frank Reavis Sr Professor of Law and Professor of Economics, Cornell University, USA Keith Klovers, Attorney, Kirkland & Ellis LLP, Washington DC, USA Alexandra Merrett, Senior Fellow, University of Melbourne Law School, Australia Nathaniel H. Nesbitt, Attorney, Hogan Lovells, Denver, CO, USA Gregory C. Shaffer, Chancellor’s Professor of Law, University of California School of Law, USA Tadashi Shiraishi, Professor, University of Tokyo Graduate School of Law and Politics, Japan vii

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

viii Comparative competition law

Rhonda L. Smith, Senior Lecturer, University of Melbourne Faculty of Business and Economics Adam Speegle, Attorney, United States Department of Justice, USA Brendan Sweeney, Adjunct Associate Professor, Monash University Faculty of Business and Economics, Australia Javier Tapia, Senior Researcher, Centre for Law Economics and Society, University College, London, United Kingdom and Judge, Chilean Competition Tribunal, Chile Simon Vande Walle, Case Handler, European Commission D-Comp (Brussels), Belgium Spencer Weber Waller, Professor and Director, Institute for Consumer Antitrust Studies, Loyola University of Chicago School of Law, USA Wentong Zheng, Associate Professor, University of Florida Levin College of Law, USA

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Prelims

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

PART I INTRODUCTION AND OVERVIEW

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 1 /

Date: 21/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

1. Themes John Duns, Arlen Duke and Brendan Sweeney

The growth in the number of antitrust regimes in the past 25 years has been nothing short of phenomenal. In the 1980s approximately 20 countries had some form of antitrust regulation. Enforcement was by no means consistent or wholehearted. Now over 120 jurisdictions boast an antitrust regime.1 While enforcement is still inconsistent, the trend towards antitrust is, when viewed broadly, nothing less than an example of global convergence. The convergence, however, is to a set of ideas not necessarily a precise set of rules. Broadly speaking, the aim of this book is to take a look at that convergence from a comparative perspective. As so many antitrust regimes have had little time to settle their rules the comparison is heavily weighted towards the developed capitalist economies, particularly the United States and the European Union, although the rules and procedures of emerging jurisdictions receive close attention in the final Part of the book. As the book is designed to be a research handbook, all chapters present descriptive material aimed at making the issues discussed accessible while also engaging with contemporary debates and issues. Of course, contrasting views on various aspects of competition law is not solely the province of interstate analysis. Vigorous, internal debate continues to be the order of the day even (and perhaps particularly) within the United States which has had over a hundred years of antitrust experience. The different approaches adopted when assessing market power provide a good example of such contrasting views or approaches. The book is divided into four Parts. Chapter 2, in Part I, provides a view of competition law through the lens of globalisation. The chapter traces the history of ‘global’ competition rules, a story that demonstrates the vagaries of shifting, state imperatives. This raises the question whether global rules are either necessary or indeed feasible. Part II investigates the substantive provisions that are common to nearly all competition law regimes. Part III examines enforcement issues, in particular the criminalisation of cartel conduct and the growing use of 1 See the website of the International Competition Network, the membership of which has grown from 16 in 2001 to well over 100 today.

3

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 1 /

Date: 7/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

4

Comparative competition law

private enforcement. Part IV considers the emergence and growth of competition regimes outside the developed states.

PART II: THE SUBSTANTIVE LAW Most competition law regimes proscribe a range of conduct that includes anti-competitive horizontal and vertical agreements, abuses of market power and anti-competitive mergers. Common to understanding each of them is the notion of a ‘market’. Therefore this Part begins by examining the notion of market; it then proceeds to a discussion on each type of conduct typically prohibited by competition law regimes. Despite the growing sophistication of econometric modelling, determination of the market remains a cornerstone of competition regulation. It remains important for differentiating lawful from unlawful conduct in both horizontal and vertical agreements, in isolating abuses of market power and in determining the outcome of mergers. Chapter 3 examines the concept of market as it has developed in the more developed competition regimes. Thus, the author discusses the purpose of market definition, the nature of markets, the theories that surround the delineation of markets and the future of market definition as a central tool of competition analysis. The pre-eminent concern of competition regulation is the proscription of hard-core cartels. This is also the area in which there is most agreement between states. In fact, convergence towards the goal of eradicating hard-core cartels has been quite remarkable. Nevertheless while there is broad agreement that hard-core cartels do substantial economic harm, identifying such cartels can be problematic. Problems begin with the threshold notion of collusion or agreement: what constitutes collusion and how is it proved? How should parallel conduct, particularly in oligopolistic markets, be treated? Chapter 4 explores how the United States and to a lesser extent the European Union have approached this threshold issue. Chapter 5 examines the content of horizontal agreements. As the author points out the variety of horizontal agreements is infinite, running from the clearly anti-competitive to the clearly pro-competitive. So a system of categorisation is required. The object is to seek a workable balance between ‘the need for business certainty on the one hand [and] an effective assessment of whether the agreement is anti-competitive on the other’. The modern trend is to move away from a simple binary categorisation (per se/rule of reason) and to recognise that horizontal agreements exist along a competition continuum.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Themes 5

Competition law regimes invariably contain a provision aimed at prohibiting abuses of market power. However, the object or objects of such provisions are not always clear. Is the main concern with allocative efficiency or is it with some other outcome such as wealth transfers? This lack of a clear organising principle affects the definition of market power and its application. Drawing on the vast literature that has examined this issue of market power the author of Chapter 6 contrasts the two dominant approaches, the classical approach of the Chicago school and the strategic approach. The author concludes by examining how Australian courts have approached key examples of market power conduct, refusals to deal, bundling, predatory pricing and raising rivals’ costs. One of the most perplexing issues confronting competition policy is the intersection between competition law and intellectual property rights. At its core intellectual property law recognises a right to exclude. This right to exclude has obvious potential to conflict with competition law’s prohibition on anti-competitive exclusion. How this tension is being resolved is the subject of Chapter 7. The author examines the tension through a variety of activities – refusals to license, settlements, product hopping, standards setting and patent pools. In each case the author compares the situation in the United States with that in Europe. Merger regulation, which forms part of almost all competition law regimes, differs from other forms of competition regulation in a number of respects. First, it necessarily employs an ex ante approach. The speculative nature of ex ante analysis coupled with the possible longlasting structural consequences of merger remedies increases the likelihood and repercussions of regulatory errors. Secondly, the globalisation of commerce has resulted in a significant number of mergers being international in character. This has tended to highlight national policy differences. While most states have similarly worded merger laws, significant differences exist in how the laws are interpreted and applied. These differences extend not only to the basic objectives of a merger regime, but also to the types (and weight) of evidence used to inform decisions. There are also important procedural differences. Chapter 8 examines these issues. Chapter 9 examines non-price, vertical restraints. No other area of antitrust regulation has been subject to such ‘dramatic shifts’ in the way in which antitrust regimes approach the subject. Given the wide variety of possible restraints the author first sets out a taxonomy of restraints. While this creates problems of classification, it enables some consistency in decision making. States disagree over the manner in which various categories of restraint should be handled. For example, the European Union’s commitment to a single market has influenced how European

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 3 /

Date: 7/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

6

Comparative competition law

authorities have approached territorial restraints. As with other areas of antitrust concern, the trend in the developed states is away from per se illegality and towards a rule of reason analysis. How states tackle this issue is the subject of Chapter 9. Chapter 10 investigates price-related vertical restraints, often referred to as resale price maintenance. The notion of the resale price being set by the supplier clashes at a fundamental level with the antitrust notion of the market as price setter. For this reason historically a rigid approach was taken to the practice of resale price maintenance; it was proscribed absolutely. This view came under sustained attack by scholars from the so-called Chicago school of antitrust who argued that resale price maintenance, like other forms of vertical restraint, is invariably imposed for pro-competitive reasons. However, the claimed pro-competitive effects and efficiencies are not universally accepted. Thus, in the US it was not until the 1990s that maximum resale price maintenance was recognised as qualitatively different from minimum resale price maintenance. This chapter analyses the changing attitudes to, and sophistication in dealing with, resale price maintenance in the US and compares it with the situation in the European Union.

PART III: ENFORCEMENT AND SANCTIONS A notable feature of competition regulation has been the concern with appropriate penalties and remedies. In relation to cartels the issue tends to centre round the notion of deterrence. Thus, while other goals such as compensation are important, the key is deterring cartels. This Part begins with a discussion of public enforcement (Chapter 11). Two contemporary aspects of competition enforcement are then considered, namely the trend towards criminalising cartel conduct (Chapter 12) and the growth of private enforcement (Chapter 14). Another notable feature of competition regulation has been the growth in cases that have an international aspect. Many cartels, for example, are no longer purely domestic affairs inflicting purely domestic harm. Just how states respond to the jurisdictional issues this creates is discussed in Chapter 13. No matter what form competition laws take, the proper functioning of the institutions entrusted to enforce such laws is a (or in most jurisdictions the) key factor determining the law’s efficacy. Chapter 11 introduces the reader to the various institutional structures adopted to enforce competition laws and assesses the merits and weaknesses of such arrangements. Detection strategies adopted by, and investigatory powers

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 4 /

Date: 7/9

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Themes 7

given to, enforcement agencies are also examined. Drawing on the highly influential theory of responsive regulation, the chapter explains the various methods adopted to promote compliance with competition laws before examining political influences and the increased importance of international cooperation. Chapter 12 examines an important aspect of public enforcement, the trend towards criminalisation of hard-core cartel conduct. This trend is quite noticeable, particularly in many developed states. It owes much to the prosecutorial zeal and the proselytising efforts of the US Department of Justice. However, convergence is neither universal nor inevitable. As the authors point out, criminalising conduct lies close to the heart of state sovereignty: thus, while a trend towards criminalisation is apparent it may not be on a sound footing if it lacks bottom-up domestic support. The growth of domestic competition regimes coupled with the internationalisation of many markets and the multinational nature of many firms has given rise to problems of enforcement coordination across borders. A key aspect of this has been the search for an appropriate level or theory of extraterritorialism. This is the subject of Chapter 13. While states have adopted different rules, there is widespread recognition that state sovereignty no longer demands a strict territorial approach to jurisdiction. There is also a growing recognition that expansive extraterritorialism is not compatible with a world of competition regimes. This is most notable in the recent trend in the United States to constrain the extraterritorial application of the US antitrust laws. The role of private enforcement in competition regulation is an issue that has generated considerable debate both domestically and internationally. As with so many areas of competition regulation, the United States has been at the forefront of the push for greater private enforcement. Of course, this is not surprising having regard to the nature of antitrust law in the United States, where private parties have always had a significant role in enforcement. However, the practice in most other states has been an almost total reliance on public enforcement: indeed, private enforcement has been regarded with suspicion. Chapter 14 examines the history, objectives and consequences of private enforcement, and how the attitude to it is changing in many states.

PART IV: COMPETITION LAW IN SELECTED JURISDICTIONS As already noted, the growth of competition law regimes has been quite remarkable, from 20 in the 1980s to over 120 today. Countries as diverse

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 5 /

Date: 7/9

JOBNAME: Duns PAGE: 8 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

8

Comparative competition law

in their political, social and economic structures and institutions as China, India and Brazil now have competition regimes. Even among the 20 countries that had a competition regime in the 1980s, some existed more on paper than in practice. Japan is an example that readily comes to mind. So far for obvious reasons this study has tended to concentrate on the United States and the European Union. Part IV is designed to address this imbalance by examining the competition regimes in Japan, China, and three South American countries (Brazil, Chile and Colombia). Chapter 15 investigates competition law in Japan. Although Japan has had a competition regime since 1947, it is only in the last two decades that it has flourished. According to the authors, while there have been significant advances in the development of substantive competition standards, the real change in Japan’s competition regime has been in the area of enforcement. Undoubtedly the Japanese experience has lessons for developing Asian countries. The evolution of competition law in China is examined in Chapter 16. That evolution has been relatively speedy and shows little sign of abating. Given the economic, political and social differences between China and the Western developed economies, China presents as an interesting study in legal transplantation. The result is what the author refers to as a legal transplant with Chinese characteristics. The result is a competition regime making steady progress in the face of some significant limitations and distortions. Competition laws have been widely adopted within South America, but they have not been widely enforced. The South American experience demonstrates the importance of political, economic and social realities to the implementation of competition regimes. Without a commitment to free markets, competition regulation struggles to gain any effective purchase. When it does gain purchase it will necessarily reflect local conditions. In Chapter 17 the authors compare the situation in three South American countries (Brazil, Chile and Colombia) from the perspective of goals, administration, sanctions and substantive standards. In each of these countries competition law has moved from a shaky beginning to a reasonably robust presence.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 02_Chapter1

/Pg. Position: 6 /

Date: 13/10

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

2. The globalization of competition law: Yes or no? Leela Cejnar and Rachel Burgess

1. INTRODUCTION In recent years there has been increasing global recognition of the importance and significance of competition law to business and commercial conduct. Over 120 jurisdictions have now adopted a system of competition law1 with an increasing number of others currently in the process of developing some form of competition law framework. Many of these jurisdictions seek to regulate anti-competitive behaviour for the reason that competition is primarily intended to increase a market’s allocative, productive and dynamic efficiencies, thereby increasing innovation, offering consumers better prices, services and choices and improving economic welfare. Others do so at the behest of international organizations (such as the World Bank) or large powerful industrialized nations. Whichever reason applies, most of these competition law regimes share common characteristics and features,2 including prohibitions on certain types of behaviour such as horizontal agreements between firms (for example, cartels aimed at market sharing, price fixing, limiting production and collusive tendering), vertical restraints between firms operating at different levels of the market and excessive aggregation of market power. However, there are many economic, social, cultural and political differences between these jurisdictions, making it difficult to reconcile the benefits of removing hindrances to competition with the need for a set of ‘global’ competition laws and policies. In addition, the uncertainty of how competition law should apply across jurisdictions remains the subject of debate. One complication challenging the need for a ‘global’ competition law is that while many countries have adopted some form of competition law 1

MM Dabbah, International and Comparative Competition Law (Cambridge University Press 2010) 3. 2 Ibid 13.

9

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 1 /

Date: 7/9

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

10 Comparative competition law

system or framework, a large number of countries have opted to not legislate to protect competition,3 relying either on the market itself (such as by promoting free trade to bring about desired economic benefits) or on other types of laws to protect competition or on state control and planning. On the other hand, some countries have looked at unilateral, regional or bilateral arrangements or a mixture of these, to create some common order with regard to the application and enforcement of competition law. To what extent, therefore, can competition law be said to be ‘global’? The purpose of this chapter is to consider what attempts have been made to develop a set of ‘global’ competition laws, the problems that might be encountered in respect of other political or economic policies or priorities, what the scope of such laws might be and how they might be implemented.

2. THE ‘GLOBALIZATION’ OF COMPETITION LAW: THE HISTORICAL BACKGROUND In the years between the First and Second World Wars, cartels were employed by several countries, including Germany, Italy and Japan, primarily to achieve political ends in the period leading up to and during the Second World War.4 At the end of the Second World War, efforts to establish an International Trade Organization (ITO) included negotiations for the inclusion of competition rules in the Draft Havana Charter. A key aim of the Draft Charter included a proposal to introduce provisions dealing with restrictive business practices.5 In particular, where such restrictive practices interfered with the trade-liberalizing aims of the Charter, the Draft Charter sought to impose obligations on member countries of the proposed ITO to prevent firms engaging in activities which may ‘restrain competition, limit access to markets or foster

3

Ibid 5. MM Dabbah, The Internationalization of Antitrust Policy (Cambridge University Press 2003) 247–248. 5 Ibid 248. See also Havana Charter for an International Trade Organization UN Doc. E/Conf.2/78 1948 (hereafter ‘Havana Charter’) in CA Wilcox, A Charter for World Trade (Macmillan 1949); P Muchlinski, Multinational Enterprises and the Law (Blackwell 1995) 403. 4

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 11

monopolistic control in international trade’.6 The Charter also provided for complaints about such restraints to be brought before the ITO. However, the United States objected to these attempts to internationalize competition (antitrust) policy and so the Charter failed and the ITO never actually materialized.7 The United States’ objection was especially surprising given that former President Franklin Roosevelt, in a letter to former Secretary of State Cordell Hull, declared that: ‘The defeat of the Nazi armies will have to be followed by the eradication of … weapons of economic welfare. But more than the elimination of the political activities of German cartels will be required. Cartel practices which restrict the flow of goods in foreign commerce will have to be curbed.’8 A further attempt to internationalize competition law, after the unsuccessful attempt of the Havana Charter, occurred when the United Nations Economic and Social Council (ECOSOC) recommended the inclusion of a draft convention that would have established an international agency with responsibility for receiving and investigating complaints about restrictive business practices. The United States again rejected this convention, however, because of concerns that disparities in domestic laws and policies were so significant that the role and function of any such international organization would have been redundant.9 The number of competition regimes grew in the years that followed. In 1958 the European Economic Community was established. Among its core rules were competition rules prohibiting agreements between Member States that had as their object or effect the prevention, restriction or distortion of competition10 and prohibiting the abuse of a dominant position within the internal market.11 Competition rules subsequently appeared on the agenda at the United Nations Conference on Trade and Development (UNCTAD) and at the Organisation for Economic Co-operation and Development (OECD). 6 See Dabbah, Internationalization (n 4) 248. See also Havana Charter (n 5) Art 46. 7 See Dabbah, Internationalization (n 4) 248. See also A Lowenfeld, Public Controls on International Trade (Matthew Bender 1983). 8 Dabbah, Internationalization (n 4) 249. 9 D Wood, ‘The Impossible Dream: Real International Antitrust’ [1992] University of Chicago Legal Forum 277, 284–285. 10 Art 101 (formerly Art 85) Treaty Establishing the European Economic Community (TFEU), entered into force 1 January 1958 (hereafter ‘TFEU’). 11 TFEU (n 10) Art 102 (formerly Art 86).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 3 /

Date: 7/9

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

12 Comparative competition law

In 1973, interestingly, at the instigation of developing countries, UNCTAD began negotiations on the control of restrictive business practices. These negotiations occurred within three different groups, one group made of industrialized countries, one primarily comprising socialist countries and one containing developing and less-developed countries.12 In 1981 the United Nations General Assembly adopted UNCTAD’s Set of Multilaterally Agreed Principles and Rules for the Control of Restrictive Business Practices (Set of Principles), which aims to ensure favourable treatment towards developing countries by offering them protection from the restrictive business practices of multinational firms. It requires multinational firms to respect the domestic laws on restrictive business practices of the countries in which they operate. It also provides that countries should improve and enforce their laws on restrictive business practices and that they should consult and cooperate with competent authorities of countries adversely affected by restrictive business practices. UNCTAD’s Set of Principles, however, is voluntary and not binding. In 1996 a significant development occurred with respect to the ‘globalization’ of competition policy, with its inclusion on the agenda of the then newly formed World Trade Organization (WTO). A WTO Working Group was set up to ‘… study issues raised by Members relating to the interaction between trade and competition policy, including anti-competitive practices, in order to identify any areas that may merit further consideration in the WTO framework’.13 However, although the European Union, Canada and Japan supported a WTO competition agreement, the United States again remained opposed to a multilateral solution, particularly one involving supranational dispute resolution.14 The United States, concerned that multilateral rules would be too interventionist, sought a non-binding, bilateral solution.15 As such, attempts to include competition rules on the WTO’s agenda failed. However, this was as much because of opposition from developing states as it was due to US opposition. Many developing states were disappointed at the failure of the WTO to deliver better access to developed markets. Accordingly, competition policy was one of the early 12

See Dabbah, Internationalization (n 4) 254. Singapore Ministerial Declaration, WTO Doc WT/MIN(96)/DEC (13 December 1996) [20]. 14 B Sweeney, ‘International Competition Law and Policy: A Work in Progress’ (2009) 10 Melbourne Journal of International Law 58, 60. 15 See Dabbah, Internationalization (n 4). 13

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 4 /

Date: 10/9

JOBNAME: Duns PAGE: 5 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 13

casualties of the Doha Round of negotiations and in 2003, at the Ministerial Conference at Cancun in Mexico, competition policy was removed from the WTO agenda.16 Yet again, attempts to establish a multilateral competition agreement were thwarted. In 2000 a Resolution was adopted by the 4th UN Conference, dealing with the issue of cooperation between antitrust authorities. The Resolution recognizes the importance of bilateral agreements and multilateral initiatives and also asked UNCTAD to examine the possibility of developing a model cooperation agreement on competition law and policy, based on UNCTAD’s Set of Principles. Accordingly, UNCTAD’s work includes the development of a Model Law on Competition, which states that the main objectives of national competition law and policy are ‘… to control or eliminate restrictive agreements or arrangements among enterprises, or mergers and acquisitions or abuse of dominant positions of market power, which limit access to markets or otherwise unduly restrain competition, adversely affecting domestic or international trade or economic development’.17 The most up-to-date Model Law was released in 2010 and is available on the OECD website. Through its Intergovernmental Group of Experts on Competition Law and Policy, UNCTAD has also been considering ways to improve worldwide cooperation on competition policy and how to spread a competition-based culture.18 Among the measures adopted by UNCTAD are training activities developed in collaboration with competition authorities in different countries. These training activities highlight the need to formulate and enforce competition and consumer protection laws in developing countries in order to increase efficient allocation of resources and, in the longer term, reduce poverty. The OECD has also been playing an important role both with regard to countries with competition law systems, as well as countries requiring assistance as they introduce or aim to introduce competition law and policy into their domestic legal systems. In particular, the OECD has issued several non-binding recommendations, including in 1986 and 1995 on international cooperation among domestic competition law authorities and a further recommendation in 1998 condemning hard-core cartels.19 16 B Sweeney, ‘Global Competition: Searching for a Rational Basis for Global Competition Rules’ (2008) 30 Sydney Law Review 209, 211. 17 UNCTAD Secretariat, ‘Objectives of competition law and policy: Towards a coherent strategy for promoting competition and development’ (UNCTAD Homepage, 2014) . 18 See Dabbah, International and Comparative Competition Law (n 1) 143. 19 See Dabbah, Internationalization (n 4) 252–253.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 5 /

Date: 13/10

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

14 Comparative competition law

The OECD Competition Law and Policy Committee (CLPC) and Joint Group on Trade and Competition (JGTC) are both engaged in programmes dealing with competition policy. The aim of the CLPC is primarily to promote common understanding and cooperation among competition authorities, while the JGTC focuses on fostering the understanding of member countries on issues relevant to the interface between competition and trade policy. The International Competition Network (ICN) is an independent body that was launched in 2001. A multilateral initiative, it focuses exclusively on competition law enforcement and on the development of domestic competition law regimes around the world. Its three areas of ‘priority’ are multi-jurisdictional merger control, competition advocacy and cartels. While each of the international organizations mentioned above (OECD, UNCTAD, WTO and ICN) has made a substantial contribution to the development of competition law and policy, more progress is needed to ensure greater international coordination and consistency in the development, application and enforcement of competition laws and policies.

3. THE GLOBAL NATURE OF COMPETITION With the shift towards trade liberalization and deregulation in the 1980s, the range of markets in which businesses could operate significantly broadened and businesses began to compete globally for customers, greater market share and increased profits. Globalization has increased multinational corporate expansion and greater international competition has in turn increased the risk of cross-border anti-competitive conduct. Competition problems have therefore become increasingly ‘global’ as firms compete in international markets in response to greater international competition.20 As discussed above, past attempts to develop multilateral competition rules have not been successful. However, as business is now conducted, almost as a matter of course, in globally integrated markets (particularly with the increasing aid of technology), the need for the development of some form of ‘global’ competition law has arguably become more essential. In considering the goals of competition law and policy it would appear that there are some common goals across the various competition law 20 M Taylor, International Competition Law: A New Dimension for the WTO? (Cambridge University Press 2006) 36–37.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 6 /

Date: 7/9

JOBNAME: Duns PAGE: 7 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 15

systems, even if the approach adopted in each, in trying to implement and achieve these goals, is different. For example, countries that have adopted a competition law framework recognize the fact that competition enhances economic efficiency for the benefit of the business or individual consumer. Similarly, protecting the consumer (and small and mediumsized firms) from any misuse of market power and ensuring a ‘level playing field’ across markets underlies another fundamental interest in and commitment to the development of a competition framework. For some, there is also the wider political agenda, such as the commitment to free trade. For example, the recognition of the importance of competition policy is evident in the WTO’s Most-Favoured Nation21 and National Treatment22 principles, as well as in several WTO agreements.23 Similarly, the Asia-Pacific Economic Cooperation Forum (APEC) recognizes that countries may be prepared to agree on areas of cooperation and facilitate harmonization with regard to aspects of competition policy even if they do not have any properly developed competition system or framework. Within APEC, this recognition has come through acceptance of APEC’s trade liberalization measures, outside a framework of legally binding measures. The fact remains, however, that in spite of a smorgasbord of competition laws and policies across jurisdictions and different approaches by the different competition authorities to the enforcement of competition laws, certain types of agreements and conduct are universally recognized as raising competition law concerns. These include hard-core cartels (particularly involving price fixing), misuse of market power, certain 21

Under the WTO’s Most-Favoured Nation (MFN) principle, WTO members must treat other WTO members equally and cannot discriminate between their trading partners. In other words, if one country is granted a special favour (such as a lower customs duty rate for one of their products) then the same ‘favour’ should be offered to all other WTO members: see WTO website: ‘Understanding the WTO: Basic Principles of the Trading System’ . 22 Under the National Treatment (NT) principle, WTO members must give others the same treatment as is given to their own nationals. In other words, imported and locally produced goods should be treated equally: World Trade Organisation, ‘Principles of the Trading System’ (World Trade Organization Website, 2014) . 23 See for example the WTO’s General Agreement on Trade in Services, Arts VIII, IX and IX:2; Agreement on Trade-Related Investment Measures, Art 9; Agreement on Trade-Related Aspects of Intellectual Property Rights, Art 41. Available at: .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 7 /

Date: 13/10

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

16 Comparative competition law

types of exclusionary conduct and international mergers. It is in these areas that the law may most benefit from a global approach. Firms doing business internationally might be faced with similar issues globally, however even where a competition law system exists, the practice of competition law and policy is likely to develop differently, if for no other reason than the differences in the political, social and judicial structures of different jurisdictions.24 So a lack of legal certainty exists. For example, the US approach to the implementation and enforcement of antitrust laws and policies is based on the free market principle: that is, every firm is free to compete, including dominant firms, with the exception of cases where specific behaviour is seen as anti-competitive.25 Section 1 of the US Sherman Act, 1890 states that a ‘… contract, combination … conspiracy, in restraint of trade … is declared illegal’. However, the term ‘restraint of trade’ is not defined in the legislation. Instead, US courts have provided a common law interpretation of the term, namely that only ‘unreasonable’ restraints of trade should be covered. The EU also adopted broad competition prohibitions. In other jurisdictions, for example Australia, the law is far more proscriptive.

4. A GLOBAL COMPETITION LAW FRAMEWORK: THE CHALLENGES The history of global competition law set out above illustrates the lack of progress made in agreeing a common set of competition rules. The appetite for doing so seems to have waned, at least since the Ministerial Conference at Cancun in 2003. The ability to reach agreement will no doubt be hampered by the established positions of the countries that have applied competition laws, independently, for more than 50 years. There are marked differences in approach even between these well-established competition regimes. For example, in the United States, the rules that apply to agreements dealing with resale price maintenance used to be fairly consistent with those that applied in Australia, that is, that agreements that control a resale price are strictly prohibited. Yet, the United States has recently changed its view on this, deciding that resale

24

See Singapore Ministerial Declaration (n 13) 64. See Dabbah, Internationalization (n 4) 273.

25

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 8 /

Date: 7/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 17

price maintenance agreements should be assessed using a ‘rule of reason’ approach.26 In regional areas like the EU, a common competition law has been relatively successful. However, the EU Member States also have consistent and coordinated economic policies which make the application of a common competition policy more realistic. While Member States have maintained or introduced their own domestic competition laws, they are required to interpret those laws in a manner consistent with the overarching EU law.27 It is difficult to envisage a similar approach being agreeable in a global context. Although the introduction of competition law and policy has spread remarkably quickly in recent years, it is developing at vastly different paces and in different ways. In 2007 India’s Competition Act came into force. The Competition Commission of India (CCI) has been reasonably aggressive in its application of the new legislation. For example, it has already imposed record fines of US$1.13 billion on 10 cement manufacturers and the Cement Manufacturers Association for engaging in cartel activity. By contrast, the People’s Republic of China introduced its Anti-Monopoly Law in 2008. Its various regulators have had a slower start with enforcement, with its initial investigations being restricted to domestic companies and only small fines being imposed. This changed in 2013 when larger fines started to be imposed and investigations extended to multinational companies. The ASEAN region agreed non-binding Regional Guidelines on Competition Policy in 2010. The ASEAN members have committed to the introduction of competition laws by 2015. To date, Indonesia, Singapore, Thailand, Vietnam and Malaysia have laws in place. There are some significant differences between the laws, despite the common policy and origins. For example, Singapore expressly excludes all vertical agreements from its Act (so resale price maintenance agreements are permitted in Singapore) while Malaysia’s legislation currently does not include a merger regime. The development of laws in all jurisdictions is inevitably driven by political or economic policies or priorities. Lack of resources means that not all potentially anti-competitive agreements and conduct are investigated and decisions as to which matters should be investigated are often a matter of policy. What will be important for the well-established competition law regimes may not even be recognized as an issue for 26 27

Columns Design XML Ltd

Leegin Creative Leather Products, Inc. v PSKS, Inc., 551 US 877 (2007). See, for example, section 60 Competition Act 1998 (UK).

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 9 /

Date: 7/9

JOBNAME: Duns PAGE: 10 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

18 Comparative competition law

some of the newer regimes. An obvious example is the criminalization of cartel conduct. Although many of the established regimes now impose criminal sanctions, some of the newer regimes are not ready to tackle this challenge. It is taking time to build expertise and resources in the countries with new competition regimes. While they are still grappling with some of the basics of implementing competition law, it is difficult to imagine these ‘newer’ competition law countries being prepared to relinquish sovereignty over competition law to a global regime. In fact, much of the opposition to developing global competition rules by the WTO came from ‘small and developing countries that feared intrusive competition law enforcement designed by developed countries’.28 The Working Paper argues ‘parachuting competition laws from developed countries on developing countries would mean that they will not be able to develop competition rules to suit their own legal, economic and political conditions’.29

5. A GLOBAL COMPETITION LAW FRAMEWORK: THE SCOPE Against this chequered background, it is difficult to envisage what the scope of a truly global competition law could realistically be. As certain types of agreements and conduct (such as hard-core cartels) are prohibited by all competition law regimes, it is possible that these types of competition breaches could be governed by a global law. However, would a global law make these types of arrangements ‘per se’ offences? If not, what legal tests would be applied? Would those involved be subjected to criminal sanction, and if so, what would be the standard of proof required? Without a central body to enforce these laws, could the same tests be applied across all jurisdictions? Other breaches of competition law, such as exclusive arrangements or abuse of dominance, that normally require an assessment of the effect of an agreement or conduct on a particular market, will be much harder to assess on an international level. Knowledge of the structure of local 28 Dr J Malinauskaite, ‘International Competition Law Harmonisation and the WTO: Past, Present and Future’ (2008) Working Paper presented during Workshop ‘Theory and Practices of Harmonisation’ held at the IALS, 24–26 June 2008, 6. 29 Ibid 12.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 10 /

Date: 13/10

JOBNAME: Duns PAGE: 11 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 19

markets, including the suppliers and customers, is essential for these types of analyses. Merger control rules exist in many competition regimes, but not all, so could a global competition law cover merger control? Even where merger control exists, different regimes apply different tests. For example, Europe looks at whether effective competition will be significantly impeded as a result of the creation or strengthening of a ‘dominant position’, whereas the United States and Australia assess whether the merger will ‘substantially lessen competition’. Arguably this is an area which would benefit the most from global rules. Most large mergers now involve multinational corporations, therefore requiring approval from multiple competition authorities worldwide. Increased coordination of the development, application and enforcement of merger laws would reduce transaction costs and lessen the burden for business of multiple jurisdictional reviews of the same transaction. It would also reduce the potential for the adoption of inconsistent procedures or conflicting decisions by different competition authorities30 and would ensure a more transparent decision-making process. Perhaps global rules could apply to multijurisdictional mergers meeting certain thresholds (as is the case in the EU), while domestic mergers could be governed by domestic rules. Would there be some benefit in creating an international competition law that only applies to international companies, so that the international body need only apply the international competition law rules to those that fall within its jurisdiction, leaving domestic competition law to apply to those companies that only operate domestically? The debate on which body (or bodies) would enforce the global rules will no doubt be significant. The establishment of an international competition body (in whatever form), responsible for enforcing a global competition law, seems unlikely to gain the support of the many countries that have their own established and respected competition authorities, as they will be reluctant to surrender their sovereignty. The alternative, of having national authorities and courts implement the global law, gives rise to different problems. Inevitably, the law will develop differently in each jurisdiction, thus quickly losing the benefit of consistency.31 30

See Jones and Sufrin for a discussion of the well-publicized merger cases of Boeing/McDonnell Douglas and GE/Honeywell where the EC and US merger authorities were of different opinions: A Jones and B Sufrin, EC Competition Law Text: Cases and Materials (Oxford University Press 2011) 1254–1255. 31 A Geiger and W Von Meibom, ‘A World Competition Law as an Ultima Ratio’ [2002] European Competition Law Review 445, 448.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 11 /

Date: 13/10

JOBNAME: Duns PAGE: 12 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

20 Comparative competition law

The WTO has been put forward as a suitable body, given its almost universal membership32 and the fact that it already has a dispute resolution system in place that is regarded as successful. However, the WTO is a body that ‘concerns itself with public restraints of trade’ whereas competition law ‘concerns itself with private restraints of trade’.33 The stalled Doha round of trade liberalization talks demonstrates that the WTO cannot reach agreement on a range of trade liberalization issues; reaching an agreement on an acceptable competition policy agenda seems even more unlikely. The failure to agree trade issues at an international level has been augmented by increased regional cooperation on trade e.g. the Trans-Pacific Partnership and the Transatlantic Free Trade Area. Perhaps agreement on competition law policies could develop in the same way, that is, by increased cooperation on a regional basis. Agreement should be easier to reach on a regional basis between ‘states with similar economic and legal structures (such as the OECD) or with special geographical links (such as the EC, EEA and NAFTA)’.34 What is clear from this discussion is that there are a number of fundamental questions that need to be resolved before a global competition law can gain any real momentum. The complexity of these issues makes it seem unlikely that they will be resolved quickly, if at all.

6. THE WAY FORWARD 6.1 Multilateral or Bilateral Agreements The negotiation of a single multilateral agreement setting out an agreed set of competition law rules presents obvious difficulties, illustrated by the failure to achieve one to date. The only multilateral agreement currently in place in relation to competition law is the UNCTAD Set of Principles referred to above, which is only voluntary. As more than 120 countries currently have competition laws in place, reaching agreement on a global competition regime is likely to be extremely contentious and time consuming. There is a real risk that any 32

Malinauskaite, ‘International Competition Law Harmonisation and the WTO’ (n 28) 7. 33 KR Fisher, ‘Transnational Competition Law and the WTO’ [2006] Journal of International Law and Policy 42, 44. 34 C Cocuzza and M Massimiliano, ‘International Antitrust Co-operation in a Global Economy’ [1998] European Competition Law Review 156, 159.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 12 /

Date: 7/9

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 21

final agreement reached would be the ‘lowest common denominator’35 as countries will be unwilling to surrender responsibility over the more important issues. By contrast, there are a number of bilateral agreements already in place around the world that deal either directly or indirectly with competition law. The EU has cooperation agreements dedicated to competition law with competition authorities in 10 countries around the world;36 the United States has antitrust cooperation agreements with 12 competition authorities37 and Australia has agreements with 10 competition authorities.38 China has entered into cooperation agreements with five competition authorities39 and India has formal cooperation agreements with the EU, the United States and Australia. Perhaps the best-known cooperation agreements in the area of competition law are those in existence between the EU and the United States.40 Despite being the subject of quite heavy criticism, they have been relatively successful. The problematic mergers of Boeing/ McDonnell Douglas and GE/Honeywell could be used as examples of why an international competition law is required, at least in the area of mergers between multinational companies. However, as Professor Whish suggests, perhaps we should applaud the fact that there are so many

35

Geiger and Von Meibom (n 31) 447. Bosnia and Herzegovina, Brazil, Canada, China, India, Japan, Republic of Korea, Russia, Switzerland and the United States. See . 37 Australia, Brazil, Canada, Chile, China, European Union, Germany, India, Israel, Japan, Mexico, Russia: The United States Department of Justice, ‘Antitrust Cooperation Agreements’ (Department of Justice Website, 2014) . 38 Canada, China, Fiji, India, New Zealand, Papua New Guinea, Republic of Korea, Taipei, United Kingdom and United States. There are further cooperation agreements in place with other countries on consumer policy issues only, e.g. European Commission: Australian Competition and Consumer Commission, ‘Treaties and Agreements’ (ACCC Website, 2014) . 39 Australia, Brazil, European Commission, Republic of Korea, United Kingdom. 40 1991 EU/US Competition Cooperation Agreement; 1998 EU/US Positive Comity Agreement; 2011 EU/US Best Practices on Cooperation in Merger Investigations: European Commission, ‘Bilateral Relations: United States of America’ (European Commission Website, 2014) . 36

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 13 /

Date: 7/9

JOBNAME: Duns PAGE: 14 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

22 Comparative competition law

cases, including mergers, where there is no friction between the EU and the United States.41 Although bilateral agreements may arguably be slightly easier to achieve from a policy perspective, it is unlikely that a series of bilateral agreements would be consistent. Negotiations between countries will inevitably lead to different agreements being reached, thus not moving the world any closer to an ‘international’ competition law. 6.2 Enhanced International Cooperation The challenges associated with negotiating agreements (even non-binding ones) between governments, whether bilateral or multilateral, may make progress in this area unattainable. Perhaps a set of global competition rules is a step too far, and effort and attention in this area should be focused instead on improving international cooperation. Through the ICN, much progress has been made with regard to informal cooperation between countries with competition laws. The ICN Statement of Achievements 2001–2013 explains the numerous achievements of the ICN since its inception in October 2001.42 The OECD has also issued a number of documents encouraging international cooperation including the OECD Recommendation on international cooperation (1995) and Best Practices on the exchange of information in cartel investigations (2005). In 2013 the OECD published its report on International Enforcement Cooperation43 setting out its findings on how the countries that responded to the survey are experiencing international cooperation. The respondents emphasised that greater cooperation in enforcement is needed, given the globalization of markets and the resulting increase in anti-competitive activity. Informal cooperation between agencies has worked well and has been valuable, with bilateral agreements and confidentiality waivers being the most widely available instruments to assist in cooperation. However, legal limitations based on different legal systems and, in particular, restrictions 41

R Whish and D Bailey, Competition Law (Oxford University Press 2012)

511. 42 International Competition Network, ‘ICN Statement of Achievements 2001–2013’ (International Competition Network Annual Conference, Warsaw, Poland, April 2013) . 43 OECD, ‘Secretariat Report on the OECD/ICN Survey on International Enforcement Co-operation’ (OECD 2013) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 14 /

Date: 7/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The globalization of competition law 23

on sharing confidential information were found to be the greatest impediments to cooperation. Although the ability to share confidential information is often achieved through confidentiality waivers, this is not ideal as it is dependent on obtaining parties’ consent.44 As such, the system for exchanging sensitive information was identified as a key area for improvement. The respondents requested that the OECD undertake projects to develop, inter alia: (a) (b)

a model multilateral cooperation and exchange of information agreement; and/or a model bilateral cooperation and exchange of information agreement.

7. CONCLUSION There is no doubt that the global nature of business presents significant legal challenges for companies operating on the international stage. In the context of competition law, those challenges have come into sharper focus as the prevalence (and perhaps publicity) of international mergers and cross-border cartel activity has increased. The number of countries with domestic competition laws has also increased, creating yet more rules requiring compliance. Those in favour of a global competition regime will argue that many of these challenges would be substantially reduced if one set of international competition rules was adopted. A common set of rules should, in theory, provide a certainty that does not otherwise exist. But would it? Establishing a common set of rules will itself be challenging and ensuring that a common set of rules is applied in the same way in each country will be almost impossible. The creation of an international body to administer those rules is unlikely to proceed as countries are concerned about losing their sovereignty over these issues. As competition laws around the world develop alongside each other, largely seeking to achieve the same broad competition policy objectives, perhaps there is no longer a need for an international law. Although problems will continue to exist in implementing laws in a consistent manner, especially where the merger or cartel is multi-jurisdictional, 44

Microsoft waived their confidentiality in respect of the investigations by the European Commission and the US Department of Justice, which made the process more efficient.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 15 /

Date: 7/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

24 Comparative competition law

perhaps the focus should be on enhancing and improving cooperation to achieve a more unified outcome, without any need for countries to relinquish control over their own laws and markets.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 03_Chapter2

/Pg. Position: 16 /

Date: 7/9

JOBNAME: Duns PAGE: 1 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

PART II THE SUBSTANTIVE LAW

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 1 /

Date: 21/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

3. Defining and proving markets and market power Rhonda L. Smith

1. THE PURPOSE OF MARKET DEFINITION Ask a lawyer why market definition is necessary for competition/antitrust cases and the answer is likely to be along the lines of: the statute requires it. In the European Union (EU) Article 102 of the Treaty on the Functioning of the European Union (TFEU), for example, relates to abuse of dominance. A firm may be dominant in a market and so indirectly the statute requires identification of that market. Of more recent origins, Chinese competition law is modelled on EU law and so it too addresses dominance and hence requires identification of a relevant market. This, however, is not always the case. For example the United States Sherman Antitrust Act 1890 makes no specific reference to market in either of the sections outlining the core prohibitions.1 Even so, the United States (US) courts and regulatory agencies routinely identify the relevant market or markets affected by the impugned conduct to enable the competitive effects of that conduct to be analysed. For the non-lawyer (or for the lawyer experienced in competition cases) the answer is rather different. Market definition facilitates the competition analysis by enabling key analytical measures to be assessed: clearly market shares and measures of market concentration require specification of the relevant market; import penetration must be determined for the relevant group of products; and barriers to entry are obstacles to entry into a market and cannot be identified in the abstract. Thus, market definition assists in assessing whether conduct has, or is likely to have, anti-competitive consequences and enables an assessment

1 Section 1 of the Act states that every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations is illegal. Section 2 prohibits monopolisation or attempted monopolisation any part of the trade or commerce among the several States, or with foreign nations.

27

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 1 /

Date: 7/9

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

28 Comparative competition law

of the extent of the market power of particular firms.2 Recently Werden has explained that market definition separates the closest substitutes and hence the most immediate constraint from more distant substitutes, which may be no more than a weak or ultimate constraint but which still allow consumers to be harmed by the exercise of market power.3 The econometrician may have yet another answer, namely that market definition is no longer necessary. It may be explained that new quantitative techniques allow the analyst to bypass market definition and proceed directly to assessing the competition effects of the alleged conduct. This view is discussed further in Section 10. Leaving aside this last view for the present, generally it is accepted as necessary to define a market in relation to the particular issue to be examined so that the most immediate constraints on the firm’s decision making are captured. In this sense market definition is purposive. As a consequence, different market definitions may be used in relation to the same industry because the alleged conduct to be considered is different, or because the competitive environment has changed. However, purposive market definition does not mean that the market must be coextensive with the alleged conduct – it may be broader or narrower4 and the means by which this is determined is discussed below.

2. THE NATURE OF MARKETS An economic market could be described as existing where buyers and sellers interact to engage in transactions. The simplest representation of a market is as a place where buyers and sellers actually meet. For example, the local fresh produce market. Today, however, buyers and sellers may not be in the same place, instead product and other relevant information may be obtained, and the transaction conducted, electronically, as in the case of purchases on eBay or Amazon. An interesting issue arises as to whether a market exists if there is a seller but no buyer or vice versa. The

2 Maureen Brunt, Economic Essays on Australian and New Zealand Competition Law (Kluwer Law International 2003) 194. 3 Gregory Werden, ‘Why (Ever) Define Markets: An Answer To Professor Kaplow’ (2013) 78 Antitrust Law Journal, 729. 4 Assume, for example, that a competition issue arises when a grocery wholesaler wants to acquire a rival wholesaler. The product of concern is likely to be groceries and associated wholesale services. However, if the competition issue relates to a grocery chain’s supply arrangements for a particular product, such as bread, the relevant market definition is unlikely to be groceries.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 29

answer seems to depend on the likelihood that the missing element will emerge within a reasonable time.5 An antitrust or competition market, as distinct from an economic market, comprises a set of close constraints with respect to the firm or firms whose proposed merger or other conduct has given rise to the competition inquiry.6 Assuming firms maximise profits, a firm’s pricing and production decisions will be constrained to the extent that buyers are willing to switch to other products and/or other producers are willing to offer an alternative product, if the original supplier increases prices by a relatively small amount.7 In the alternative, discretionary power rests upon an absence of close substitutes for the firm’s output, either actually or potentially. This means that market power ultimately rests upon two factors that act as constraints upon a firm’s business behaviour: the numbers of (independent) firms and patterns of substitution for their products within an industry, and the extent to which there are barriers to entry of new firms, which would produce close substitute products, from ‘outside’ an industry (including the limbo of unborn firms) … Thus, substitutability becomes the basic concept – in the economist’s sense.8

Currently some firms in a particular market may not impose much of a competitive constraint on other market participants because they are niche operators. Nevertheless, given an incentive, they may become an effective competitive constraint. Consequently, firms included in a market are those that impose close actual or potential competitive constraints on one another.9 Market is a multi-dimensional concept. Although exactly how many dimensions it has seems to vary from one jurisdiction to another, all accept that there is a product dimension and a geographic dimension to a market. A joint publication of the UK Competition Commission/Office of 5

Rhonda L Smith and David K Round, ‘When is a Market a Market?’ (2003) 31 Australian Business Law Review, 412–422. 6 Edward S Mason, Economic Concentration and the Monopoly Problem (Harvard University Press 1959) 65; Jeffrey Church and Roger Ware, Industrial Organization: A Strategic Approach (Irwin McGraw Hill 2000) 601ff. 7 Rather than an increase in the nominal price of the product, the supplier may offer a poorer quality product at the same price, may fail to lower prices in response to reduced production costs or may alter other terms such as credit terms. All of these can be conveniently addressed as an increase in price. 8 Brunt (n 2) 194–195. 9 Other entities may influence the operation of markets, including the government, which may put restrictions in place that may influence the willingness and/or ability of buyers to switch suppliers; and intermediaries such as brokers or insurance companies.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 3 /

Date: 7/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

30 Comparative competition law

Fair Trading, for example, states that ‘[m]arkets may also be defined by reference to customer group or temporal factors’.10 The product dimension of a market identifies the products which, if given a price incentive to do so, will be substituted for one another by buyers. The geographic dimension of the market includes those areas from which consumers can source supply without too much additional cost or inconvenience in response to a price incentive. In addition, the competition analysis may be assisted by identifying the relevant steps or functions in the supply chain for the product (the functional dimension). 2.1 Debate about Supply-side Substitutability The pricing and other decisions of the firm or firms whose conduct has aroused competition concerns may be constrained by the ability of buyers to switch to other products. However, in addition, they may be constrained by the ability of some firms to alter their product mix quickly and without significant investment, to supply the relevant product and so take customers away.11 Baker suggests that one of the main disputes in relation to market definition is the role of supply-side constraints. He acknowledges that the debate is about when, rather than whether, supply-side should be introduced, and comes down strongly in favour of defining markets based on demand-side substitutability and then including firms in the market if they are able to supply the relevant product/s – this is the approach adopted for mergers in the US, although in Canada supply-side may be considered as part of the market definition process.12 The basis for this conclusion is that it can be ‘both difficult and confusing to ask one analytical step, market definition, to account for two

10

Competition Commission & Office of Fair Trading, Merger Assessment Guidelines (September 2010) para 5.2.5. Identifying a customer dimension to a market may be appropriate when there is potential for price discrimination between customers (para 5.2.5(c)). 11 Supply-side substitutability refers to the ability of a producer, in response to a SSNIP, to alter its production mix in favour of the item whose price has increased. For example a market gardener producing strawberries could quickly switch to producing lettuces if the price of lettuces increased by a SSNIP. However, it is unlikely that under the same circumstances a strawberry producer could switch into cereal production, as investment in land and new equipment, and possibly establishing new marketing arrangements, would be required. 12 US Department of Justice and Federal Trade Commission, US Horizontal Merger Guidelines (2010); Competition Bureau Canada, Merger Enforcement Guidelines (2011), para 5.1.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 4 /

Date: 7/9

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 31

economic forces, demand and supply substitution’.13 Although supplyside constraints may not form part of the market definition when assessing mergers, courts in the US tend to take it into account when assessing other forms of potentially anti-competitive conduct. Baker suggests that this is to avoid having to balance the pro-competitive and anti-competitive effects of the conduct required by an unstructured rule of reason analysis.14 This contrasts with the position in Europe and in countries such as Australia. For example, the Office of Fair Trading (OFT) states: ‘undertakings that do not currently supply a product might be able to supply it at short notice and without incurring substantial sunk costs. This may prevent the hypothetical monopolist profitably sustaining prices 5 to 10 per cent above competitive levels.’15 Padilla argues that it is appropriate to distinguish between supply-side substitutability (including it as part of market definition) and new entry (addressed as part of the competition analysis).16 First, generally additional supply becomes available more quickly via supply-side substitutability than as a result of entry. Second, supply-side substitutability is uncommitted entry in that it is low cost and does not require irreversible investment, unlike new entry. Third, supplyside substitutability impacts both pre- and post-entry prices, whereas if new entry involves substantial sunk cost, it will be based on expectations about retaliatory actions and post-entry prices. Padilla concludes that ‘the constraint imposed by supply-side substitution on the pricing incentives of incumbent firms is equivalent to that created by readily available demand substitutes’.17 Arguably whether the supply-side constraint is considered as part of the exercise of defining the market or as part of the competition analysis should be of little consequence. It should neither narrow nor broaden markets. However, occasionally it does matter, as for example when

13

Jonathan B Baker, ‘Market Definition: An Analytical Overview’ (2007) 74 Antitrust Law Journal, 129–173. 14 This concept refers to the position of the US Supreme Court that only combinations, including contracts, that unreasonably restrain trade are subject to the antitrust laws, and possession of monopoly power is not illegal per se. 15 Office of Fair Trading, Market Definition: Understanding Competition Law, 2004, para 3.13. 16 Atilano Jorge Padilla, The Role of Supply-side Substitution in the Definition of the Relevant Market in Merger Control (A Report to DG Enterprise A/4, European Commission, June 2001). 17 Ibid 20.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 5 /

Date: 7/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

32 Comparative competition law

demand-side substitutability is unlikely to render a price increase unprofitable but supply-side substitutability would be likely to do so.

3. THE PROCESS FOR DEFINING MARKETS – THE PRODUCT DIMENSION Competition regulators often issue guidelines, especially in relation to mergers, and these provide detailed information concerning how the relevant market should be identified in a particular matter. On the other hand, the relevant competition statute generally has little, if anything, to say about the process for defining markets. The process has been described as partly descriptive and partly purposive.18 However, if the alleged anti-competitive conduct impacts on substitution possibilities then arguably it should not be taken into account when defining the market. In addition, if the market defines away the ‘area’ where the investigation of the conduct would occur, this is unlikely to be helpful. Imagine, for example, that a competition issue arises in relation to the supply of sports channels for subscription television. Parties related to a key supplier of this product have entered into exclusive contracts with the owners of various key sports rights which the only other sports channel supplier claims has forced it out of business. Whether or not this actually raises a competition issue is not relevant for the present purpose. More significantly, if it is concluded that rights owners could integrate forward into channel supply and/or subscription television service providers could integrate backward into channel supply and hence there is no channel supply market, this would obviate the need for a competition analysis, an outcome that seems unhelpful. Thus, Market definition is not an exact physical exercise to identify a physical feature of the world; nor is it the enquiry after the nature of some form of essential existence. Rather, it is the recognition and use of an economic tool or instrumental concept related to market power, constraints on power and the competitive process which is best adapted to analyse the asserted anticompetitive conduct.19

It follows that markets are constructs to facilitate analysis of particular alleged conduct rather than facts in their own right. 18 Re Singapore Airlines Limited v Taprobane Tours WA Pty Ltd [1991] FCA 621; (1992) 14 ATPR 41-159, [36]. 19 Australian Competition and Consumer Commission v Liquorland (Aust) Pty Ltd [2006] FCA 826, [429].

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 6 /

Date: 7/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 33

A product market consists of those goods or services which are readily substitutable in demand or, depending on the jurisdiction, in supply, within an appropriate time frame, for the product(s) of the firm(s) whose conduct is at issue. Similarly, identification of the geographic dimension of the market commences from the area supplied by the firm. However, it is convenient to focus only on the product market for the present. As most firms do not produce just one product, it will be necessary to determine which of the products supplied are relevant to the matter at hand. Although a supermarket sells many thousands of items, if the competition issue relates to its dealing with bread suppliers, the starting point may be narrowed to bread; whereas if the competition issue involved the acquisition of another supermarket in a particular location, the starting point may be the goods and associated services (such as car parking, and the ambience of the store), supplied by the supermarket as grocery prices reflect this bundled offer.20 Economic markets generally consist of products that are close substitutes, but sometimes competition markets are comprised of products that are not substitutable or that are not even interchangeable in use. Assume that a competition issue arises in relation to the transport of a product to a port for export. Although road and rail transport are not close economic substitutes (rail is considerably cheaper over long distances), the competition issue relates to both rail and road. Thus, the starting point for market definition is ‘transport (both rail and road)’ and as there are unlikely to be any close substitutes for this, the product dimension of the market is likely to be transport. There may be interdependencies between the products/services supplied from the buyers’ perspective which determine the product dimension of the market, even though consumers do not regard the components as substitutable. This will be the case if it is more convenient and cheaper for consumers to acquire these products from a single supplier (the one-stop retail shop) – for example, banking products.21 A collection of

20 Some bundled products can be unbundled in whole or in part, for example generally supermarkets offer a bundle of grocery products and services such as car parking but the latter need not be supplied by the supermarket. 21 The cluster market concept was spelt out in United States v Pittsburgh National Bank & Trust Co: ‘The clustering of financial products or services in banks facilitates convenient access to them for all banking customers. For some customers, full-service banking makes possible access to certain products or services that would otherwise be unavailable to them. In short, the cluster of products and services termed commercial banking has economic significance well beyond the various products and services involved’ 306 F.Supp 645, 648–51

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 7 /

Date: 7/9

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

34 Comparative competition law

different but related products may be found to form a single market if an increase in their price would be unlikely to be constrained by another supplier offering only some of the products/services because consumers do not wish to acquire the components separately from different suppliers (such a market is described as a cluster market). 3.1 The SSNIP Test/Hypothetical Monopolist Test Next, to provide for the practical application of the price elevation test, substitutability is assessed by reference to the response of demand (and, depending on the jurisdiction, supply) to a relatively small, sustained, not insignificant increase in price from the competitive level (a SSNIP) by a hypothetical monopolist, while holding the other determinants of demand constant.22 There are two opposing outcomes from a SSNIP – a loss of revenue and profit as the quantity demanded reduces in response to the price increase and higher revenue and profit from retained sales. Thus ‘[t]he key issue is whether the loss of sales would be sufficient to offset the increased profits that would be made from retained sales following the price increase’.23 The starting point is to delineate a candidate market based on the qualitative and quantitative information available. If the SSNIP is unprofitable, it means that the hypothetical monopolist is constrained by the loss of sales to other products outside of the proposed ‘test’ market. The test market then needs to be broadened to include these products and the SSNIP test reapplied. The iterative process continues until the SSNIP is found to be profitable, thereby identifying the outer boundaries of the market. The focus is on the smallest market where the hypothetical monopolist could profitably maintain a SSNIP, taking into account all relevant sources of substitution. This means that weaker, but relevant, competitive constraints may exist outside of the market and should be considered when undertaking the competition analysis rather than as part of the market definition exercise. It is often assumed incorrectly that the SSNIP test is an objective test. The SSNIP is applied to a candidate market which must be decided based (D.N.J. 1969) reviewed, 399 US 350 (1970), quoted by Gregory J Werden, ‘History of Antitrust Market Definition’ (1992) 76 Marquette Law Review, 123, 166. 22 This is often referred to as the hypothetical monopolist test. 23 UK Office of Fair Trading, The role of market definition in monopoly and dominance inquiries: Economic Discussion Paper 2 (prepared by NERA, July 2001) 7.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 8 /

Date: 7/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 35

upon knowledge of the industry and given the nature of the alleged conduct, and this starting point may be determinative of the market definition. Second, if the candidate market is expanded by adding the next closest substitute, this too involves judgement. It requires at least a qualitative assessment of the degree of substitutability – this then determines the sequence of closest substitutes and different views about this sequence may alter the market identified. Third, the extent of the response to a SSNIP depends on the amount of time allowed for responses. While theory of the firm defines time on the supply-side in terms of technology, demand-side responses are not categorised in the same way. Generally response is greater the longer the period allowed for buyers to investigate their options. The period over which substitutability is tested should be long enough for consumers (and alternative suppliers) to respond to the incentive to switch resulting from the price increase. This will differ according to the type of product and may be different for different type of consumers. However, if too long a period is allowed, price responses tend to become confused with responses to other changes in the market, changes that shift the demand curve and are not relevant for market definition, which is based on movement along the demand (or supply) curve. The consensus is to allow a period of up to 12 months for assessing substitutability, although in reality often most of the response occurs within a much shorter period. The SSNIP should be applied to the competitive price rather than to the market price. This is because if the price of the product under consideration is already very high due to existing market power, a further price rise might cause consumers to buy less and to replace it with other products that consumers regard as poor alternatives even though if price was set at the competitive level little or no substitutability would be observed. This is referred to as the ‘cellophane fallacy’.24 However, the competitive price may be difficult to identify (for example, the product is new) or empirical data concerning consumer responses may not be available (for example, because the market has never been competitive). For mergers, however, the guidelines of the 24 Du Pont was the sole producer of cellophane but argued that it operated in a wrappings market that included various other forms of wrapping (such as foil, wax paper and polythene) because there was a high cross price elasticity of demand between cellophane and these other products. However, this high elasticity of demand was caused by the fact that Du Pont, because of its market power, was able to charge monopoly prices for cellophane. See United States v E.I. DuPont & Co, 351 US 377 (1956).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 9 /

Date: 7/9

JOBNAME: Duns PAGE: 12 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

36 Comparative competition law

various regulatory bodies generally state that the SSNIP is applied to the market price. The implicit assumption is that the competition test relates to a change in competition as a result of the merger. The SSNIP is generally taken to be an increase in price of 5 per cent, which, unlike the 1 per cent used as the basis for measuring elasticity in economics textbooks, is large enough to potentially cause a response. However, this is arbitrary and so sometimes the SSNIP is 10 per cent; occasionally it is less than 5 per cent. An interesting issue arises where the conduct under investigation is at the wholesale level of the supply chain, for example in relation to groceries. Assume a SSNIP of 5 per cent is applied by a hypothetical monopolist wholesale supplier of groceries. Even if the full price increase was passed on, arguably such an increase would not be noticed by consumers because wholesale costs are only one part of retail price, which also includes retail costs. If this is the case, then retailers would not respond as their demand is derived from that of their customers. It may be necessary to increase the wholesale price by much more than 5 per cent for it to be noticeable by retail customers. While such an increase would not be regarded as ‘small’ in an absolute sense, in the relative sense of the SSNIP test such an increase may be regarded as appropriately small. As Geroski and Griffith observe, ‘what is considered to be “small but significant” will vary across markets and over time and will depend on product characteristics, past price increases, current inflation rates and a number of other factors’.25 3.2 When is the SSNIP Test not Useful? There are situations where a SSNIP test is not helpful, or may not appear to be helpful, in delineating the market. Probably the most wellrecognised situation is where a new product has been introduced or where there is ongoing technological change (as in the case of telecommunications and related products) as this can ‘change the nature of products, the possible substitutes for those products, and the identity of the suppliers of the products and their substitutes’.26 Assume that a new product such as subscription television is introduced, that is, there is discrete product innovation. In the period following its introduction several things will happen. First, some consumers will take up the new 25 Paul Geroski and Rachel Griffith, Identifying Anti-trust Markets (Institute for Fiscal Studies Working Paper 3/01, 2001) 8. 26 William Baxter, ‘The Definition and Measurement of Market Power in Industries Characterized by Rapidly Developing and Changing Technologies’ (1984) 53 Antitrust Law Journal 717.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 10 /

Date: 7/9

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 37

product but not all will continue to subscribe either because it does not meet their expectations or because it is too expensive. Likewise from the supply-side, it is not necessarily easy to determine the appropriate price for a new product and so the price may be adjusted from time to time following the initial product launch. Suppliers of competing products may also adjust their offerings. The result is that there may appear to be considerable ‘substitution’ occurring. However, this is not substitution in response to a SSNIP but instead is the market adapting to the new product. A SSNIP for the purpose of identifying a relevant market should not be applied until the market has ‘settled down’ and how long that takes depends on the nature of the new product and market conditions more generally. This will mean that if a competition issue arises before this process has been completed, a SSNIP test will not assist the process of market definition. As competition is more likely to be based on product attributes than on price, in these industries, Jorde and Teece suggest recasting the SSNIP test to inquire ‘whether a change in the performance attributes of one commodity would induce substitution to or from another … [if so] the differentiated products, even if based on alternative technologies, should be included in the relevant product market’.27 However, it seems likely that such a test would be difficult to apply – for example, the meaning of ‘performance attribute’ is uncertain. In their Guidelines for the Licensing of Intellectual Property, issued in 1995, the Department of Justice (DOJ) and Federal Trade Commission (FTC) introduced the concept of ‘innovation markets’, a market that consists of the research and development (R&D) directed to particular new and improved goods and processes, and the close substitutes for that R&D.28 An innovation market includes ‘all firms with the capability and incentive to undertake research and development closely substitutable for’ the R&D at issue,29 even if the firms are not competitors in the relevant markets for related goods. Close substitutes may include other R&D, technologies and goods that significantly constrain the market power associated with the relevant R&D. The Guidelines specify that innovation markets are only to be employed when the competitive effects of licensing arrangements cannot be adequately assessed within the 27

Thomas M Jorde and David J Teece, ‘Rule of Reason Analysis of Horizontal Arrangements: Agreements Designed to Advance Innovation and Commercialize Technology’ (1993) 61 Antitrust Law Journal 579. 28 Department of Justice and Fair Trading Commission, Guidelines for the Licensing of Intellectual Property (1995) [3.2.3]. 29 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 11 /

Date: 7/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

38 Comparative competition law

relevant market for the goods affected by the arrangements (eg if the goods do not exist as yet); and when the capabilities to engage in the relevant R&D can be associated with the specialised assets or characteristics of specific firms. Rather than apply a SSNIP, Gilbert and Sunshine suggest that in order to identify an innovation market, it is appropriate to assess ‘the set of research and development activities for which a hypothetical monopolist would profit by a small but significant and non transitory reduction in R&D’.30 Relevant constraints may come from alternative sources of R&D, or from downstream products. A two-year time period is suggested. However, some writers have rejected the need for innovation markets.31 Rapp, for example, argues that the relationship between the number of competitors and the extent of R&D is too ambiguous to allow meaningful conclusions about the effects of horizontal arrangements between innovation market competitors.32 More recently, with enormous technological changes in communications technology, attention has turned to the treatment of futuregeneration products. The issue is the ability of the hypothetical monopolist to slow the pace or limit the scope of R&D in relation to an envisaged product. However, this creates a problem for market definition and competition analysis in that it is only present-generation goods, existing assets and R&D in progress that are currently observable. Yet, for example, for R&D alliances formed to facilitate innovation of new products or production processes, current products and markets are largely irrelevant.33 One possible approach is to attempt to ‘forecast’ future goods markets based on what is known of the R&D, that is, working through innovation markets.34 Clearly, this is likely to be speculative. An alternative approach is to focus on the nature of the assets used in the R&D, especially when the firms merging or entering into

30 Richard J Gilbert and Steven C Sunshine, ‘Incorporating Dynamic Efficiency Concerns in Merger Analysis: The Use of Innovation Markets’ (1995) 63 Antitrust Law Journal, 569, 594. 31 See vol 64(1) of the Antitrust Law Journal for a number of articles addressing the issues. 32 Richard T Rapp, ‘Misapplication of the Innovation Market Approach to Merger Analysis’ (1995) 64 Antitrust Law Journal, 19, 20. 33 Richard Clapes, ‘Blinded by the Light: Antitrust Analysis of Computer Industry Alliances’ (1993) 61 Antitrust Law Journal, 899, 904. 34 Janusz A Ordover and Robert D Willig, ‘Antitrust for High Technology Industries: Assessing Research Joint Ventures and Mergers’ (1985) 28 Journal of Law and Economics, 311.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 12 /

Date: 7/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 39

agreements have large shares of substitute R&D assets that also require a large share of market-specific assets for effective commercialisation.35

4. GEOGRAPHIC DIMENSION OF MARKET The geographic dimension of the market refers to the area over which consumers could source the relevant product without undue cost and inconvenience and similarly into which alternative suppliers could supply. Like the product dimension of the market, it is assessed based on substitutability in response to a SSNIP. A threshold consideration is the extent to which the product can be moved from one location to another in response to a price increase. Thus, if the relevant product is particularly fragile or highly perishable, this may limit the distance over which it can be supplied or sourced. Likewise, government regulation may restrict the area over which substitution can occur. For example, the government may issue only one licence for supply of a particular product into a given area. While many services must be supplied at the point of sale, technological changes in some cases allow delivery of service by a supplier located elsewhere, even overseas. For example, diagnostic services may be provided by a medical expert located overseas who receives information electronically and then provides the diagnosis back to the referring doctor. Clearly, a prime determinant of the geographic boundaries of a market is transport costs because substitution will not occur if the additional cost involved in sourcing from further afield outweighs all or most of the price increase by the hypothetical monopolist. As a result, the geographic market for bulky, low-value products (such as sterile water or cement) is generally smaller than for small light and/or high-value products (such as jewellery). Additional costs associated with sourcing from/supplying to more distant locations include the need to hold more stocks of a product to avoid shortfalls, insurance costs, and in the case of supply from another country, hedging costs against exchange rate changes. Another relevant consideration is whether there are government restrictions on ‘importing’ a product from one area to another (whether from overseas or within a particular country). Such restrictions may include tariffs, the need to obtain a licence to be able to supply the product (such as registration of pharmaceutical products with the industry regulator) and labelling requirements (such as those required on tobacco products). 35 Janusz Ordover and William Baumol, ‘Antitrust Policy and HighTechnology Industries’ (1988) 4 Oxford Review of Economic Policy, 13.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 13 /

Date: 7/9

JOBNAME: Duns PAGE: 16 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

40 Comparative competition law

Globalisation and e-commerce pose a variety of challenges for competition analysis, including for determining the appropriate geographic dimension of the market. This is illustrated by an issue relating to re-importation of pharmaceutical drugs from Canada to the US. In early 2005 various pharmaceutical companies in the US threatened to reduce supply to Canadian distributors if those distributors continued to sell their product to US consumers, known as re-importing.36 This occurred because Canadian government price regulations often meant that the same prescription pharmaceuticals were 30 to 50 per cent cheaper in Canada than in the US. Given that the Canadian market is quite small relative to the US market, ‘re-importation’ meant that the pharmaceutical companies were losing millions of dollars’ worth of sales. For the alleged threat to be susceptible to the Sherman Antitrust Act 1890, it was necessary that the relevant geographic dimension of the market was Canada and the US, not simply the latter. Excluding Canada from the geographic market would separate the drug companies’ practices into two distinct markets, precluding an antitrust violation because the disputed action would occur in the Canadian market and would therefore have no antitrust effect on the US market and would remain unreachable by US oversight.37 The basis for concluding that the relevant market included both Canada and the US was said to be, first, that the pharmaceutical companies operate throughout the world but supply Canada from the US. Arguably, the internet facilitates re-importation and integrates geographic markets,38 and government barriers to trade have not prevented trade.39 Second, sales data revealed that as US prices for pharmaceuticals increased, re-importation from Canada increased. The practice is to identify separate markets in circumstances where price discrimination occurs,40 which

36

The following is based on Maryan M Chirayath, ‘Oh Canada! Antitrust Geographic Market Definition and The Reimportation of Prescription Drugs’ (2005) 46 Boston College Law Review, 1027. 37 Ibid 1055. 38 Presumably sales could occur using the postal service but would occur on a relatively small scale. 39 Canada allows pharmacies to fill patient prescriptions without first visiting a Canadian doctor. Although the US government’s stance is that re-importation is illegal, laws against re-importation are not being enforced, and public pressure for more affordable pharmaceuticals could result in a change of policy. 40 See US Horizontal Merger Guidelines (n 12) s 4.1.4.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 14 /

Date: 13/10

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 41

would place Canada and the US in separate geographic markets. However, in this case it was argued that the price differentials reflected cost differences and so were not discriminatory.41 It is sometimes said that the spatial dimension of the market can be determined from where transactions occur. This may be true in some cases, but not if the product relates to a transport service. In such matters, the transaction may occur in one place but this may or may not be where service delivery occurs. For example, a taxi ride is paid for at the conclusion of the trip but the relevant taxi market would generally be taken to be wider even than the area between the origin and the destination. Once conduct occurs beyond national boundaries there is the potential for jurisdictional issues to arise. Most countries have provisions in their competition laws that address these. Possibly as a result of an overzealous draftsperson at some earlier time, recently some countries have discovered that the specific wording of their legislation has raised concerns about the ability of their competition regulator to address certain issues. Both Australia and New Zealand have sections in their Competition Acts that describe a market as ‘… a market in’ Australia or New Zealand as the case may be. Generally, courts have taken the view that if the conduct occurred in a global market and had some impact within the particular country, then there was a market ‘in’ the relevant country. Nevertheless, in the court cases in each country relating to the air cargo cartels (the fixing of fuel surcharges) this approach was challenged by the airlines. Essentially the airlines argued that transactions occur within a market and where they occur identifies the market. Air cargo is sold at the point of origin for air cargo going to Australia or New Zealand and the airlines argued that this meant that the market was at the origin, not at the destination. In New Zealand, the court rejected the airlines’ argument. It found that the service provided by the airlines was the delivery of air cargo from origin to destination and so the product dimension of the market did include delivery and this occurred in New Zealand.42 This creates the interesting situation that the product dimension has a geographic element to it, which then is relevant for the geographic dimension of the market. While this issue is peculiar to the 41 The price differences may reflect different approval processes for new drugs and different government regulations. Price discrimination occurs if different customers are charged different prices when the cost of supplying them is the same or where prices are the same despite different supply costs. 42 Commerce Commission v Air New Zealand Ltd [2012] NZHC 271.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 15 /

Date: 7/9

JOBNAME: Duns PAGE: 18 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

42 Comparative competition law

extremely prescriptive competition law in Australia and New Zealand, it does draw attention to the fact that in relation to transport industries, and possibly also in other industries, where the transaction is located may not reveal anything useful about the geographic dimension of market boundaries, particularly in global markets.

5. THE SUPPLY CHAIN AND THE FUNCTIONAL DIMENSION OF MARKETS A further aspect of the market which needs to be considered is the supply chain associated with the relevant product. A traditional supply chain might start with farmers growing wheat; that product is then milled and converted into flour; the flour is used by bakers to bake bread; and the bread is then sold to supermarkets and other retailers where it is purchased by consumers. Other supply chains are shorter. For example, in advertising the advertisement is conceptualised, created and published. For some services, creation and distribution occur simultaneously, as when supplying a haircut. Mostly not much thought needs to be given to which step or steps in the supply chain are relevant for the competition analysis and hence for market definition; they are obvious. Typically, an implicit decision in relation to the supply chain is included in the specification of the product dimension of the market, such as ‘the market for the production of steel’ or ‘the retail supply of bread’. From time to time, however, the relationship between these functional activities is significant for the competition analysis, and so identifying which of these activities is relevant and how they relate is necessary. Are there separate wholesale and retail markets, for example, or should they be considered jointly? If the latter, why is this the case? The issue becomes clearer, perhaps, when it is recognised that often at least some of the major participants in an industry are vertically integrated, but the competition issue to be analysed arises in relation to only one of these activities. Alternatively, the decision making of a firm may be so constrained by the competition that its customers face, that even though it is not vertically integrated, it must take that competition into account in making its own business decisions. As each of the functional activities in a supply chain are complementary, it is not appropriate to identify the relevant functions for the competition analysis and hence which of these needs to be included in the market definition by considering substitutability. First, starting from the functional activity at which the firm operates or at which the issue has arisen, it is appropriate to consider whether and to what extent firms

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 16 /

Date: 7/9

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 43

operating in this activity are involved at other functional stages, that is, whether they are vertically integrated.43 If there are significant efficiencies (such as savings in transaction costs) from vertical integration, then it would be inappropriate in undertaking the competition analysis to create a functional division between the activities. Nevertheless, the vertical integration observed within an industry may not be a consequence of economic considerations, but may be caused by other factors such as regulation or past government ownership.44 Even if this is not the case, frequently some firms in an industry are vertically integrated and some are not. This suggests that the synergies from vertical integration are not so great that a firm operating on a stand-alone basis cannot be commercially viable and so the functional activities might be considered separately. However, even if firms are not vertically integrated, it is appropriate to combine two or more functional activities when it is not possible to explain adequately the competitive process at one functional stage without knowledge of the role or influence of individual firms who also operate at adjacent stages. This is illustrated as follows. Assume a merger takes place between the only two independent grocery wholesalers in a particular geographic market. Independent wholesalers are assumed only to supply independent grocery retailers and independent grocery retailers only acquire supply from independent grocery wholesalers. However, independent grocery retailers are in competition with the retail outlets of the vertically integrated supermarket chains. Thus, the merged entity will be constrained by the loss of business at the retail level that would result from an exercise of market power. To understand the competitive constraint on the merged entity, it is necessary to understand the competitive process at the retail level. Consequently, the two stages or levels in the supply chain should be addressed jointly given the purpose of market definition. Economies of scale and scope are critical to the profitability of both wholesalers and retailers, and failure to fully exploit them, for example due to loss of business to another retailer, raises unit costs and creates a downward spiral not just for the retailer, but also for the wholesaler. If 43

Both vertical integration by ownership and by long-term contracts aligning interests are relevant for this purpose. Firms at different functional levels may be regarded as vertically integrated where they are associated via a long-term one-on-one contract even though there is no common ownership because the contract creates an incentive to act as though there was common ownership. 44 Governments traditionally took the view that if there was a natural monopoly element within a supply chain, it was appropriate for government to operate at all levels.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 17 /

Date: 7/9

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

44 Comparative competition law

this is the case, then wholesalers are closely constrained by competition at the retail level and it seems appropriate to treat wholesale and retail jointly for the purpose of facilitating a competition analysis. Arguably this could form part of the competition analysis, but given the closeness of the constraint it seems more appropriate to specifically recognise the constraint in defining the market.

6. THE EVIDENTIARY BASIS FOR MARKET DEFINITION Although the SSNIP test is formulated in a quantitative way, its value is in providing a conceptual framework within which evidence about substitutability can be assessed by reference to qualitative and quantitative industry information. At a practical level, the process of market definition requires identifying the information required and then sourcing that information. Essentially the former concerns the responsiveness of buyers and sellers to a SSNIP, and various elasticities measure this.45 However, mostly these measures are not readily available and so the factors that determine the elasticities are assessed instead to determine whether responses are likely to be large or small. Thus, in relation to the product dimension of the market, demand-side substitutability will depend on whether buyers accept that alternative products are functionally substitutable (capable of fulfilling the same role). Given this, consideration will be given to the relative prices of the initial product and substitute products as substitution will not occur if the substitutes are more expensive than the original product post SSNIP. In addition, any costs involved in switching must also be taken into account. For example, changing a fuel source from electricity to gas or vice versa. On the supply-side, the threshold issue will be whether production (or other) processes can be readily adapted to supply the product at issue or its close substitutes. If so, then the extent of any switching costs (such as identifying and negotiating alternative distribution arrangements), and

45 Demand-side substitutability is measured by the cross elasticity of demand, that is, the responsiveness of the demand for product Y to a relatively small sustained change in the price of product X (X being the product of the firm whose conduct is the subject of the competition inquiry). This provides information about the extent of substitution.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 18 /

Date: 7/9

JOBNAME: Duns PAGE: 21 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 45

regulatory requirements (such as licensing or other approvals) must be weighed against the impact on prices of the SSNIP.46 Having identified the information required, the next issue is to source that information. The focus is on determining the likely extent of substitution in response to a SSNIP and so the focus is not on core consumers47 but on marginal consumers. There are numerous sources of information, both quantitative and qualitative, that may assist in identifying markets.48 These include: (a)

(b)

(c) (d)

(e)

(f)

(g)

documents that indicate the considerations taken into account when setting the prices of the firm’s products – which other product prices and firms are considered in this process; market research reports commissioned by the firm/s which assess the factors that influence demand for the product/s and changes in the purchase patterns of buyers; sales reports, data such as scan data from stores that link price and sales volume; accounting data from which variable profit margins can be derived which may allow inferences concerning the price elasticity of demand as well as the relative profitability of different products; documents that report the firm’s assessment of its performance because generally these reports involve comparisons against those firms regarded as the firm’s closest rivals; strategy documents (management and board reports) that indicate how the firm assesses the commercial opportunities available to it, as well as business analyses prepared for investors; and information concerning the flow of product between different geographic regions.

46 Correspondingly, in relation to the geographic dimension of the market, the threshold issue concerns the ability of product to be transported from outside the original supply area. This might be affected by perishability, fragility and the bulky, low unit value of the product. Assuming that movement is possible, the key consideration is likely to be transport costs relative to the SSNIP. Other considerations relate to costs such as insurance and regulatory requirements (such as access to licences and peculiar packaging/labelling requirements). 47 These consumers are quite unresponsive to price increases, having a strong preference for the particular product. For example a supporter of a particular football club is very unlikely to buy the supporter wear of another club because it is cheaper. 48 Caron Beaton-Wells, Proof of Antitrust Markets in Australia (The Federation Press 2003) chapters 3–6.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 19 /

Date: 13/10

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

46 Comparative competition law

However, care must be taken in interpreting these documents because the market responses referred to may be due to a change in a non-price variable, especially when the product at issue is a relatively new product. Further, where responses do relate to a price change, that change may be considerably larger than a SSNIP and so is likely to overstate the degree of responsiveness. Alternatively, both price and other supply terms may be changed at the same time so that the price effect cannot be separately distinguished. In addition, it is important to interpret business documents in context. For example, recognition of the position of a product in its product life cycle (whether it is a new product, a mature product or a declining product) may assist in interpreting observed switching between products. Market research documents may be helpful in this respect. Often submissions to public inquiries relevant to the particular industry provide information about who the firms in the industry see as their competitors and their views of how ‘markets’ work. These reports/ submissions have the advantage of being independent of the particular competition issue being considered; however, care should be taken when interpreting the statements as submissions are generally focussed in ways dictated by the nature of the inquiry and may not coincide with the issue at hand. Other useful sources include submissions to and findings from previous competition inquiries and government inquiries into the relevant industry. Official and industry data sets may be useful but given the purposive nature of market definition, these may be of limited assistance.49 Useful information often comes from views expressed by industry participants about who they regard as competitors, especially in relation to decision making within the business. However, care needs to be taken when interpreting this material. First, often business people are concerned about competitiveness rather than competition, which may narrow their focus to the exclusion of other competitors. Second, business documents are created for a particular purpose and that may widen or narrow the range of other firms discussed.

49

Witness Statements prepared for the purpose of litigation may assist with factual information and context but often views about the market must be formed before these are available.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 20 /

Date: 7/9

JOBNAME: Duns PAGE: 23 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 47

7. QUANTIFYING THE RESPONSE TO A SSNIP The extent to which quantitative techniques are used for identifying markets varies. A key determinant is whether appropriate data is available. This is especially so in smaller jurisdictions where data is often unavailable. Three of the more common quantitative approaches are considered below. 7.1 Natural Experiments This process is probably the simplest and least demanding of data. It essentially uses differences between products or locations or through time to provide insights into the relevant market. Other things being similar, it could be concluded that if prices for a particular product in two locations were the same when in one area a second product was not present, then the product was not a close substitute for the other product and should not be included in the market. One of the difficulties with this approach is that the differences may reflect other unrecognised differences. Natural experiments are perhaps more often used to assess the competition effects of alleged conduct or proposed mergers than to determine markets. As is noted elsewhere it is an approach which has been used to justify avoiding market definition and proceeding direct to a competition analysis. In giving evidence to the Antitrust Modernization Commission, Willig rejected such an approach.50 Coate and Fischer offer an interesting perspective on this issue based on the Staples case.51 The issue at hand was whether the merger of two of the three office supply superstores (Staples and Office Depot) would lessen competition. They start with the information that prices for the products supplied by Staples and Office Depot increased as the number of independent superstores in a geographic area declined. Given that the market was defined as an ‘office superstore market’, this supported concern about the competition effects of the proposed merger. However, they suggest that without this market definition it would be dangerous to read too much into pricing information. This is because: 50

Robert Willig, ‘Testimony Before the Antitrust Modernization Commission’ (17 November 2005) . 51 Malcolm B Coate and Jeffrey H Fischer, ‘A Practical Guide to the Hypothetical Monopolist Test for Market Definition’ (2008) 44 Journal of Competition Law and Economics, 1031, 1047.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 21 /

Date: 13/10

JOBNAME: Duns PAGE: 24 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

48 Comparative competition law

(a) (b)

pricing could be influenced by promotional pricing; prices were higher in smaller stores because they were unable to exploit fully economies of scale; and customers valued the one-stop shop such that they were prepared to pay more for products sourced in that way.

(c)

7.2 Price Correlations Stigler and Sherwin adopt a quite basic approach.52 They suggest that where substitution possibilities are strong, while the price levels of different products may differ, the pattern of price movements will be very similar (highly correlated). Thus, if the price movements of two products are highly correlated, this indicates that they are highly substitutable (have high cross price elasticities of demand) and the products are deemed to be in the same market. Conversely, if the movements in product prices are not highly correlated, then the products are deemed to be in different markets. However, there are problems associated with the first of these conclusions. This is because the prices of different products may be correlated for reasons other than substitutability,53 and so there is no basis for concluding that products are in the same market – the correlation of price changes is not necessarily the result of substitution. Nevertheless, if the changes in the prices of two products are not highly correlated, it is very probable that they are not in the same market. It is in this sense that the Stigler-Sherwin test may assist with market definition. 7.3 Critical Loss Analysis Possibly the first empirical method used to assist in delineating markets was that of Elzinga and Hogarty54 and this can be seen as the forerunner of modern critical loss analysis as developed by Harris and Simons.55 The critical loss ‘identifies for any given price increase the amount of sales (as a percentage) that can be lost before the price increase becomes 52 GJ Stigler and RA Sherwin, ‘The Extent of the Market’ (1985) 28 Journal of Law and Economics, 555. 53 For a discussion of the use of price trend data and the problems associated with this, see Office of Fair Trading, Market Definition in UK Competition Policy (Research Paper 1, February 1993) 35. 54 KG Elzinga and TF Hogarty, ‘The Problem of Geographic Market Delineation of Antitrust Suits’ (1973) 18 Antitrust Bulletin, 45. 55 Barry C Harris and Joseph J Simons, ‘Focusing Market Definition: How Much Substitution is Necessary’ (1989) 12 Research in Law and Economics, 207; see also Jeffrey Church and Roger Ware, Industrial Organization: A

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 22 /

Date: 13/10

JOBNAME: Duns PAGE: 25 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 49

unprofitable’.56 In terms of market definition, it identifies the loss of sales revenue by a hypothetical monopolist that renders the SSNIP unprofitable – if the actual loss exceeds the critical loss, the SSNIP will not be profitable. The residual elasticity of demand measures the extent to which demand for the products of a particular firm responds to a price increase, given consumer reaction and the reaction of other firms.57 Applied to a hypothetical monopolist, the residual elasticity of demand determines whether a critical loss actually occurs, that is the price increase is unprofitable. The higher the residual elasticity of demand for the products of the hypothetical monopolist, the more sales it will lose and the more likely the actual sales loss will exceed the critical loss. While this approach identifies the constraints on a firm (or hypothetical monopolist), it does not identify the source of those constraints. Yet this is the essence of market definition. Further, suitable and accurate price and cost data may not be available.

8. TESTING THE MARKET DEFINITION: COMMERCIALITY The product and geographic dimensions of markets are defined on the basis of substitutability, although this is not the case in relation to the functional dimension. Nevertheless, it is appropriate to test the market definition/s resulting from this process. One way of doing this is to consider whether the definition is commercially realistic, that is whether it is consistent with the way the industry operates and the way consumers behave.58

Strategic Approach (Irwin McGraw Hill 2000) 606–607; Deidre L Hay, ‘Multipurpose Critical Loss and Critical Demand Elasticity’ (2004) 11 Competition and Consumer Law Journal, 236. 56 Ibid 239. 57 Sumunth Addanki, ‘Market Definition Using Econometrics: An Apparent Paradox Explained’ in Lawrence Wu (ed), Economics of Antitrust (NERA 2004) 18–19. Addanki argues that Marshallian rather than residual demand elasticities should be used for identifying markets and the two may, under certain circumstances, provide quite different results. 58 Brunt (n 2) 209–211; David Round, ‘A Matter of Fact, Commercial Commonsense or Economic Principles?’ in A Bollard (ed), The Economics of the Commerce Act (New Zealand Institute of Economic Research, Research Monologue No 52, 1989).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 23 /

Date: 13/10

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

50 Comparative competition law

The relevance of such considerations is illustrated by an issue involving recording companies and retailers in New Zealand.59 The latter claimed that the imposition of maximum retail prices on them by Festival Records substantially lessened competition. In relation to market definition, it was argued that a significant proportion of customers shopping for a particular album would not be prepared to substitute an alternative album in response to a SSNIP for the first album (ie the cross price elasticity of demand between albums was low). In addition, supply-side substitutability was low because copyright controls and artists’ contracts limited or prevented the supply of alternative versions of the same album. Thus, it was argued that there was a separate market for each album. However, commercial considerations indicate that the approach was not realistic. The popularity of any album is generally quite short lived and within months or even weeks demand is likely to shift to some other album. At the same time, recording companies can easily switch from supplying one album to another. They respond to the latest hit by trying to supply similar types of songs/music by similar types of artists and they may also attempt to change consumer tastes in favour of their own product. Judged only on substitutability, a single album market may have been adopted. However, in the light of commercial reality, this was rejected as no record company could ever run a business on the basis of one album. It is inappropriate to consider only a ‘snapshot’ of the responses at a point in time as a basis for assessing market boundaries. By testing the market definition arrived at on the basis of substitutability against the way the industry operates and the way consumers behave, it is possible to arrive at market definition(s) that reflect not only industry structure, but also strategic conduct.

9. MORE COMPLEX MARKETS: TWO-SIDED PLATFORMS OR MARKETS Competition inquiries and cases relating to interchange fees for the use of credit cards drew attention to a structure which had previously received little direct recognition. However, that does not mean that firms with such structures had not been considered previously, rather simply that this particular feature was not specifically acknowledged. What is distinctive

59 Tru Tone v Festival Records Retail Marketing Ltd (1988) 2 NZBLC 99–135.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 24 /

Date: 7/9

JOBNAME: Duns PAGE: 27 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 51

about these firms is that they offer a platform that brings together two groups, each of which is then a customer of the platform. The structure to which the interchange fee cases around the world drew attention was described as a two-sided market, that is, firms supplied two distinctive groups of customers with services, namely cardholders and merchants. It quickly becomes apparent that there are many firms associated with the supply of different products that have a similar structure. Perhaps the most obvious are media firms that sell advertising space – newspapers and television are obvious examples. The second group of customers are those supplied with the vehicle for the advertising – readers of the newspaper and television viewers.60 Similarly, software platforms bring together various groups of customers – hardware producers, applications developers and computer users. Finally, there are exchanges of various sorts that bring together different groups of customers. These include shopping centres, auction houses, internet sites, brokerage services, real estate agents and dating services. What distinguishes a two-sided platform is the mutuality of the two groups. Two-sided platforms have three key characteristics:61 first, there are two distinct groups of customers who need each other and rely on the platform as an intermediary to bring them together – the platform provides goods and/or services to the two groups simultaneously. Second, there are indirect externalities across customer groups – the value that the customers on one side obtain depends on the number of customers on the other side. Third, the price structure is non-neutral – the price structure of the platform affects the level of transactions. Given this, the price structure must be designed to induce both groups of customers to join the platform – one group of customers may pay nothing (as in free-to-air television) or each group may pay (as in newspapers where readers and advertisers both pay) but they may not contribute proportionately. In markets such as these if a competition issue arises in relation to customers on one side of the market, the assumed SSNIP used to identify the product dimension of the market will have implications for customers on the other side of the market and hence for the platform operator’s profitability. Consequently, it is not appropriate to define the market only in relation to the customers on one side of the market. Yet this is

60 OECD, ‘Two Sided Markets, Best Practice Roundtable’ (2009) . Whilst the OECD Country Report for Australia refers to a number of media cases, at the time when those cases were considered there was no specific recognition of the two-sided market. 61 Ibid 11–15.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 25 /

Date: 7/9

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

52 Comparative competition law

precisely what competition regulators appear to be doing. As Evans and Noel state: In some cases, the fact that a business could be considered a 2SP may be irrelevant, either because the indirect network effects, though present, are small or because nothing in the analysis of the practices really hinges on the interlinked demand. In other cases, the fact that a business is a 2SP will prove important both for identifying the real dimensions of competition and focusing on sources of constraints.62

It quickly becomes apparent that two-sided platforms create problems in applying the hypothetical monopolist test, that is, the SSNIP test. First, the SSNIP could be applied to the overall price, or to one price only – one price may be increased, while the other remains unchanged or is reduced.63 Second, it is not readily apparent what the appropriate competitive price is because it cannot be assumed to simply reflect the marginal cost of supply.64 Perhaps these difficulties explain the tendency for competition authorities to ignore the existence of two-sided platforms when defining markets. Instead, the additional customer group is taken into account when undertaking the competition analysis. This may result in markets being defined too broadly or too narrowly – group externalities mean that the loss of one customer may lead to the loss of other customers.

10. A DIMINISHING ROLE FOR MARKET DEFINITION Inability to convince a court of a particular market definition often is seen to be determinative of antitrust cases, and consequently it is usually contentious. However, in recent times, particularly in relation to mergers, regulatory authorities have reduced their reliance on market definition, opting instead for directly assessing the competition effects of the merger or other conduct. In August 2010 the FTC and the DOJ jointly issued updated Horizontal Merger Guidelines. These Guidelines indicate that ‘[t]he Agencies’ analysis need not start with market definition. Some of the analytical tools used by the Agencies to assess competitive effects do 62

David S Evans and Michael Noel, ‘Defining Antitrust Markets When Firms Operate Two-Sided Platforms’ [2005] Columbia Business Law Review, 101, 130. 63 See OECD (n 60) 149–156. 64 Similarly, applying critical loss analysis to two-sided platforms may be problematic: Evans and Noel, (n 62) 133.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 26 /

Date: 7/9

JOBNAME: Duns PAGE: 29 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 53

not rely on market definition, although evaluation of competitive alternatives available to customers is always necessary at some point.’65 Similarly, in guidelines issues at around the same time in the UK, the OFT and the Competition Commission took a similar position.66 There are a number of reasons for the fall from grace of market definition. They include: (1) recognition that structural factors do not necessarily enable forecasts of conduct or performance; (2) the problem of markets definition when products are differentiated and unilateral effects are the issue; (3) the difficulty of addressing certain business structures using quantitative techniques to define markets (such as two-sided markets or situations where prices are stable or relatively so); and (4) development of direct measures of the impact of conduct. Of these, without doubt, the last is the most significant. Nevertheless, attempts to bypass market definition raise a number of concerns. The first is that rather than avoiding market definition, the definition is implicit and hence the process is less transparent. Competition agencies argue that consideration of substitutability results in markets that are too broadly defined when products are differentiated and there are unilateral effects. However, ignoring substitutability results in markets that are defined unrealistically. Several US merger assessments illustrate this. For example, in relation to the proposed acquisition by Nestle Holdings, Inc. of Dreyer’s Grand Ice Cream, Inc., the direct effects of the proposed merger were considered only in terms of super premium ice cream because the parties are major suppliers of this product, and so the implied market definition was super premium ice cream.67 Second, in markets with differentiated products, if there is simply a focus on the relationship between the two firms associated with a proposed merger and without broader inquiry about other competitive constraints which would be captured by the market definition, there is significant potential for false positives. The propensity for narrow markets can be illustrated by several recent US merger assessments such as

65

US Horizontal Merger Guidelines (n 12) 7. Competition Commission & Office of Fair Trading, Merger Assessment Guidelines (n 10). 67 Fair Trade Commission, ‘Nestle-Dreyer’s Settle FTC Charges’ (FTC Media Release, 25 June 2003) . 66

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 27 /

Date: 13/10

JOBNAME: Duns PAGE: 30 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

54 Comparative competition law

Staples (a merger of office supply superstores); and Whole Foods and Wild Oats (a merger supplying premium and natural organic groceries).68 Third, frequently the data available for assessing the direct effects of the alleged conduct is inadequate and/or unsuitable. Rill draws attention to the use of historical data in econometric models which fail to incorporate dynamic market responses to the merger, yet these responses supply the competitive constraint that prevents anti-competitive conduct post merger.69 Carlton,70 while supportive of the use of new econometric tools designed to measure demand-side responses for antitrust purposes, suggests that there are four problems associated with their current use. These are: (a)

use of the wrong variable – for example, often retail demand is used as a basis for estimating wholesale demand; ‘using a combination of demand elasticities and an assumption about competition’ to estimate costs rather than use of actual marginal cost;71 failure to take into account non-price as well as price; and to predict post-merger equilibrium prices, the type of competition before and after the merger must be specified but generally it is assumed that this remains unchanged.

(b)

(c) (d)

Fourth, market definition provides the basis for a rigorous consideration of the constraints and the strength of those constraints on the firm/s whose conduct may damage the competitive process. Dispensing with market definition (or deriving it from the econometric analysis of effects) may result in less rigorous analysis and less consistency in decision making. Fifth, assessment of direct effects shifts the focus from harm to the competitive process to the impact on competitors and consumers. For example, the Gillette case related to an attempt by the Gillette Company to acquire Parker Pen Holdings. The concern related to premium fountain pens in the US as the merged entity would account for 40 per cent of

68

For a discussion of these cases see Neal R Stoll and Shepard Goldfein, ‘We Don’t Need No Stinking Markets’ [2004] The Antitrust Bulletin, 593. 69 James F Rill, ‘Practicing What They Preach: One Lawyer’s View of Econometric Models in Differentiated Products Mergers’ (1997) 5 George Mason Law Review, 397, 399. 70 Dennis W Carlton, ‘Market Definition: Use and Abuse’ (2007) 3 Competition Policy International, 52, 60–61. 71 Ibid 60.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 28 /

Date: 7/9

JOBNAME: Duns PAGE: 31 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Defining and proving markets and market power 55

such sales. The DOJ argued that pens within a given price range were not interchangeable with pens in other price ranges because customers viewed and treated them differently. Given this, consumers, who could have switched to other fountain pens and other writing implements but who chose not to do so, were afforded protection. However, at trial the court did not accept this very narrow market definition.72 In addition, in emphasising only the most immediate existing substitutes rather than defining the market, the regulator may fail to take into account firms that currently do not supply close constraints but that could do so relatively easily in response to the incentive provided by an exercise of market power. Market definition and competition analysis are part of the same process, separated for convenience – a means of managing the often large amount of relevant information. The market defined is likely to have fuzzy edges and it may not capture all the factors influencing competition. It may be just one of a number of candidate markets but it is the one that facilitates the particular competition analysis.

72

U.S. v The Gillette Company, et al, 828 F.Supp. 78 (D.D.C. 1993) 84–85. In the US, demand-side substitutability is the sole basis for market definition. Supply-side substitutability is taken into account in identifying those suppliers in the market. In Gillette, on review, it was found that barriers to entry into fountain pen supply were relatively low, a factor that is of fundamental importance to the assessment of the competition effects of the merger.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 04_Chapter3

/Pg. Position: 29 /

Date: 7/9

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

4. Anti-competitive agreements: The meaning of ‘agreement’ George A. Hay*

1. INTRODUCTION In the classic cartel, supposed competitors meet in the proverbial smoke-filled hotel room and agree to fix prices at supra-competitive levels. Even though the ‘agreement’ is unlikely to be legally binding on the parties (i.e., the agreement could not be enforced against one of the cartelists that began to ‘cheat’ by offering lower prices), virtually all modern economies would treat such a cartel as unlawful under their national antitrust laws. For the United States, this negative attitude toward cartels is both long standing (dating virtually to the passage of the Sherman Act in 1890) and harsh, with the potential not only for very large fines (in 2012, the Department of Justice sought a fine of $1 billion against just one of several participants in the LCD cartel – AU Optronics – and the Court granted a fine of $500 million), but also for incarceration for key individual participants (the Department of Justice (DOJ) asked for 10-year prison sentences for two individual executives of AUO and the Court awarded three-year sentences).1 And, of course, there is the very real possibility of follow-on private actions seeking damages that are automatically trebled. A recent report indicated that the total of the settlements reached thus far on behalf of US consumers with members of the LCD cartel was $1.1 billion.2 Other jurisdictions have come more recently to treat cartels harshly, but prosecution of such cartels is now pursued vigorously on a global * The author is grateful to several Cornell Law School students, especially Colin McKeon and Briana Serano, for assistance in preparing this chapter. 1 See Department of Justice (US), ‘Press Release: Taiwan-Based AU Optronics Corporation Sentenced to Pay $500 Million Criminal Fine for Role in LCD Price-Fixing Conspiracy’ (Department of Justice US Website, 20 September 2012) . 2 Stephanie Mlot, ‘Buy an LCD? Sign Up for Piece of $1.1B Price-Fixing Deal’ (PCMag.com, 23 October 2012) .

56

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 1 /

Date: 7/9

JOBNAME: Duns PAGE: 2 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 57

scale. Outside the US, the imposition of large fines is now commonplace, in part because, unlike the US, in many jurisdictions, such as the European Union, a fine is not regarded as a ‘criminal’ penalty. The EU recently leveled a fine of €1.47billion (US$1.92 billion) on participants in a cathode-ray tube cartel.3 Prison sentences for involved individuals are far less common, but a number of jurisdictions, including Australia for example, have moved to criminalize cartel behavior and thereby hold out the possibility of incarceration, and there is likely to be further movement in that direction.4 Private damage actions are not as prevalent outside the US for a variety of reasons, including the lack of a trebling provision in the relevant statute and the difficulty of effectively maintaining a class action, but even here there seems to be some movement toward the US model.5 In sum, while the penalties for the cartels that are detected and prosecuted differ across jurisdictions, the basic attitude that such ‘gardenvariety’ cartels are and should be unlawful is remarkably consistent. Moreover, the ability to detect (and therefore prosecute) such cartels has evolved rapidly over the past 20 or so years, in part because of a remarkable degree of cooperation among enforcement authorities around the world, but in even larger part because of the effectiveness of various leniency programs maintained by multiple jurisdictions around the world (the US and the EU being the most notable examples), which offer some diminution in adverse consequences for the first member of a cartel to 3 Lucian Constantin, ‘EU Fines CRT Makers €1.47 Billion for Price-fixing’ (PC Advisor, 5 December 2012) . 4 See for example Caron Beaton-Wells, ‘Australia’s Criminalization of Cartels: Will It Be Contagious?’ in Josef Drexl and other (eds), More Common Ground for Competition Law? (Edward Elgar 2011) 148–73. 5 See for example Russell discussing, among other developments, ‘recent calls within the European Union for greater private enforcement of competition law and [outlining] steps the European Commission has taken to address that need, including the recently published White Paper on Damages for Breach of EC Antitrust Rules’: Tiana Leia Russell, ‘Exporting Class Actions to the European Union’ (2010) 28 Boston University International Law Journal 141. European Competition Commissioner Almunia announced that he intends to submit to the College of Commissioners a legislative proposal on private antitrust damage actions that would, among other things, make it easier for victims of cartels to gain access to evidence uncovered by the Commission: Joaquin Almunia, ‘The Role of Competition Policy in Times of Crisis’ (Address Before the 29th Annual AmCham EU Competition Policy Conference, Brussels, 6 December 2012) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 2 /

Date: 13/10

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

58 Comparative competition law

come forward, confess its own role, and provide evidence implicating other members of the cartel.6 Indeed, the frequency of new cartel discoveries that are reported around the world in various antitrust newsletters suggest that cartelists are almost tripping over themselves in a rush to be the first to confess. Yet the suspicion remains that, despite the effectiveness of the leniency programs in encouraging confessions, there are still cartels (both domestic and international) that operate under a cloak of secrecy, and therefore governments must continue to use more conventional tools to seek them out and prosecute them. Sometimes these traditional investigatory tools will yield the confession or the hot documents that will make prosecution easy, but not always. So the question arises: in the absence of a video-taped cartel meeting, a cooperating participant, or incriminating documents, can a cartel be successfully prosecuted based primarily on ‘circumstantial’ evidence such as the charging of a common price by competitors? However, it is not just a question of what kind of circumstantial evidence can be used to establish the existence of a hard-core cartel – that question might best be described in these circumstances as akin to an economic detective story. A more complicated substantive question underlies the effort to describe the type and quantity of the circumstantial evidence necessary to obtain a successful prosecution, namely, what precisely do we mean to include under the category of an illegal ‘agreement’ or ‘conspiracy’ or ‘concerted practice?’ The contract lawyer’s reaction to such a question might be one of puzzlement as to why it is being asked. After all, every good contract lawyer is trained to identify when parties have reached an ‘agreement.’ But not so fast. The contract lawyer’s assignment is to determine when the parties have reached an agreement that can be enforced in court against those who attempt to violate its terms. But we already know that, in almost all jurisdictions, the ‘garden-variety’ cartel, as described above, does not produce a legally binding agreement. In almost any jurisdiction, any such ‘agreement’ would be void and unenforceable as against public policy. So the cartels we are most interested in prosecuting are not cartels that have reached ‘agreement’ in the technical legal sense. And if the antitrust ‘agreement’ is different from and in some ways less than the 6 Details of the Department of Justice’s Antitrust Leniency Program are provided on the Antitrust Division’s website at . Details of the EU Leniency Program can be found on the EU website at .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 3 /

Date: 7/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 59

contract lawyer’s agreement, then we need to establish exactly what counts as an agreement or conspiracy or concerted practice (depending on the language of the relevant statute) in order to know how to prove its existence. In other words, to know how to establish the existence of an unlawful agreement under the antitrust laws, we must first develop a sense of what ‘counts’ as an unlawful agreement and how far our concept of an unlawful agreement can depart from the contract lawyer’s notion of a legally binding contract.7 The plan of this chapter is to begin by describing how the US treats classic cartel behavior when proof of the existence of a cartel agreement is not an issue. Then we turn to the task of establishing the existence of an illegal agreement primarily or entirely through circumstantial evidence. This will quickly get us into the question of what constitutes an unlawful agreement under US law and in particular, the recently renewed debate about whether classic oligopoly behavior can be prosecuted as an unlawful agreement. In the process, we will refer to how similar issues are dealt with in other jurisdictions, most notably the European Union.

2. ANTITRUST TREATMENT OF ‘HARD-CORE’ CARTELS Section 1 of the Sherman Act prohibits every ‘contract, combination … or conspiracy, in restraint of trade or commerce …’ It is rare for courts to assign separate meanings to the three categories of offense, and so the statute is usually paraphrased as prohibiting any ‘agreement’ in restraint of trade. In the EU, Article 101 prohibits ‘all agreements between undertakings, decisions by associations of undertakings and concerted practices’ which directly or indirectly fix prices. More than a century ago, US courts determined that price-fixing, as represented by the classic cartel, belonged in the prohibited category,8 and that such cartel agreements fixing prices could not be defended on the ground that they were 7 To be more concrete, if the only kind of ‘agreement’ that the Sherman Act recognized were to be one in which there were direct, face-to-face communication, as in the classic smoke-filled room, that would have significant implications for the kind of circumstantial evidence that might support a finding of agreement. At the opposite end of the spectrum, if the law treated simple oligopolistic interdependence as also constituting an agreement, the range of relevant circumstantial evidence would be much wider. 8 United States v Addyston Pipe & Steel Co., 85 F 271, 293 (6th Cir 1898), aff’d, 175 US 211 (1899).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 4 /

Date: 7/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

60 Comparative competition law

‘reasonable.’9 The phrase ‘illegal per se’ was formally added to the antitrust lexicon involving price-fixing in 1940, when the Supreme Court declared that ‘a combination formed for the purpose and with the effect of raising, depressing, fixing, pegging, or stabilizing the price of a commodity in interstate or foreign commerce is illegal per se.’10 While the EU has not specifically adopted the ‘per se’ designation for traditional price-fixing, the practical effect is the same.11 With the per se rule firmly established for classic cartels, the Court turned its attention to questions of whether the rule would apply to all kinds of defendants in all kinds of circumstances. In Goldfarb the Court made it clear that the per se rule against price-fixing applied to professions, such as lawyers, though in a famous footnote the Court decreed that other kinds of agreements that would merit per se treatment for traditional commercial entities, such as an agreement not to engage in certain kinds of advertising, for example, might be evaluated under the rule of reason if the restraint could be characterized as an ethical norm aimed at regulating and thereby promoting competition.12 In 1979 in Broadcast Music (BMI), the Court determined that a rule-of-reason analysis should be applied where the price-fixing was ancillary to the establishment of a joint venture that created a new product.13 In 1984 the Court found the NCAA (National Collegiate Athletic Association) liable for what it characterized as price-fixing although the Court modified its 9

The classic ‘reasonableness defense’ had one or more of the following elements: (a) the cartel does not have the power to impose unreasonable prices; (b) the cartel had a reasonable purpose (i.e., a purpose other than to impose unreasonable prices); (c) the actual prices agreed to were not unreasonable. 10 United States v Socony-Vacuum Oil Co., 310 US 150, 223 (1940). 11 See for example European Commission, ‘Antitrust Overview’ (European Commission, 13 August 2012) . 12 Goldfarb v Va State Bar, 421 US 773, 778 n.17 (1975). Almost immediately thereafter, the Court made it clear that switching to a rule-of-reason analysis did not open up every possible defense argument and that the only inquiry under the rule of reason would be whether the restrictive agreement on balance promoted or restricted competition. National Society of Professional Engineers v United States, 435 US 679, 696 (1978). 13 Broad. Music, Inc. v Columbia Broad Sys, 441 US 1, 24 (1979). While the BMI opinion contained language that may have suggested a much more substantial narrowing of the reach of the per se rule, the Court in 1982 seemed to reaffirm its commitment to the rule when price-fixing occurred outside the context of a joint venture in holding that it applied even to an agreement setting maximum prices. Arizona v Maricopa County Medical Society, 457 US 332, 348 (1982).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 5 /

Date: 7/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 61

analysis to take a ‘quick look’ at the NCAA’s alleged procompetitive justification on the grounds that the NCAA was a ‘league’ and that in this context, some horizontal agreements are necessary for the league to function. The quick look was very quick, however, as the Court found no basis for the NCAA’s claim that the price-fixing was necessary. Therefore, while there are circumstances where the per se rule may not apply to price-fixing, these tend to be quite limited and generally involve some sort of (legitimate) joint venture where the price-fixing is seen as at least facially ancillary to the functioning of the joint venture.14 As applied to the ‘garden-variety’ cartel, however, the per se rule seems firmly established.15

3. PROVING THE EXISTENCE OF A CARTEL AGREEMENT Of course, while the per se rule gives the plaintiff a great litigating advantage once the existence of a cartel agreement is established, there is still the task of establishing the existence of the cartel agreement in the first place. In an ideal plaintiff’s world, that task would be facilitated by the existence of one or more kinds of ‘hard’ evidence, such as a videotape of the cartel meeting,16 confession and cooperation by a key 14

Where a court finds in its ‘quick look’ that the price-fixing is not genuinely ancillary to the functioning of the joint venture, it essentially reverts to per se analysis. 15 The most interesting ‘relaxation’ of the rule was by an appellate court which held that agreements on financial aid by Ivy League and other colleges were properly characterized as price-fixing but should nevertheless be assessed under the rule of reason because the schools were non-profit entities without a profit-maximizing purpose. Moreover, the court directed the trial court (on remand) to take into account the social welfare justifications as a possible offset to the anti-competitive effect. The case then settled. United States v Brown University, 5 F 3d 658, 678 (3d Cir 1993). 16 This is rare but not impossible. In the case of the worldwide lysine cartel, the Justice Department persuaded a cooperating witness to arrange for a cartel meeting to be held in the US at a resort hotel (with the incentive being a very nice golf course on the premises for post-meeting enjoyment). The meeting room was equipped with a secret camera and microphone, so the entire proceedings were videotaped: Scott D Hammond, ‘Caught in the Act: Inside an International Cartel’ (Address Before OECD Competition Committee, 18 October 2005) . Highlights of the full recording were distributed on request by the Justice Department; ‘Copies of the tape and transcripts are available at no charge by mailing … your request to the

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 6 /

Date: 7/9

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

62 Comparative competition law

participant in the cartel meetings, or an incriminating document that can be described as a ‘smoking gun’. Confessions have been a frequent source of evidence in the last 10 or so years, as the US and other jurisdictions have introduced amnesty programs for the first participant in a cartel to come forward with a confession and a promise of cooperation aimed at implicating other participants.17 But what happens when there is no ‘hard’ evidence (or at least not enough by itself to ensure successful prosecution)? Can the plaintiff win a price-fixing case when the only evidence is circumstantial? And if the answer to that question is ‘yes,’ what exactly is the nature of the evidence that a court will entertain and find persuasive? Or perhaps more critically, since most cartel cases in the US, if they are litigated to a conclusion, will be decided by a lay jury,18 what evidence is sufficient to allow the case to go to the jury?19 United States Department of Justice, Antitrust Division, Freedom of Information Act Unit, 325 Seventh Street, NW, Suite 200, Washington, D.C., 20530.’: ibid. 17 See Scott D Hammond, ‘Cornerstones of an Effective Leniency Program’ (Address Before ICN Workshop On Leniency Programs, 22–23 November 2004) . For an overview of the EU leniency policy, see European Commission, ‘Cartels: Leniency’ (European Commission Website, 16 April 2012) . To date, there is no formal reward program for whistleblowers (such as that employed by the Internal Revenue Service to detect tax evasion, or by the SEC under Dodd-Frank to uncover securities fraud) but one suspects that a statute creating such a program is not too far off. 18 In criminal cases, the defendant is entitled to a jury. In private treble damage cases, if the amount in question is greater than a de minimis threshold, either party can ask for a lay jury. Typically it is the plaintiff who requests the jury, presumably out of a perception, not necessarily supported by empirical data, that juries will be more favorable to plaintiffs than judges in price-fixing cases: See for example Kevin M Clermont and Theodore Eisenberg, ‘Trial by Jury or Judge: Transcending Empiricism’ (1992) 77 Cornell Law Review 1124. 19 The defendant has various opportunities to ask the judge to dismiss the case either before a jury has begun to deliberate or even after a jury has announced its verdict. While there are occasional permutations the basic scheme is as follows. After the complaint has been filed but before broad discovery has commenced (there may be some limited and finely tailored discovery on a specific issue), the defendant can move to dismiss the complaint. This kind of dismissal used to be quite rare in the absence of an incompetently drafted complaint, but in the wake of Twombly, such dismissals are not uncommon: See Bell Atlantic Corp. v Twombly, 550 US 544 (2007). See also, Kevin M Clermont, ‘Inventing Tests, Destabilizing Systems’ (2010) 95 Iowa Law Review 821; Colleen McMahon, ‘The Law of Unintended Consequences: Shockwaves in the Lower Courts After Bell Atlantic Corp. v. Twombly’ (2008) 41 Suffolk University

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 7 /

Date: 7/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 63

4. USE OF CIRCUMSTANTIAL EVIDENCE A general framework for answering these questions was provided by the Court in the classic Interstate Circuit case from 1939.20 Interstate Circuit was a chain of so-called ‘first-run’ movie theaters and had a dominant share of the first-run theaters in the relevant geographic market.21 Despite its ‘monopoly,’ Interstate was concerned about competition from ‘secondrun’ theaters that did not have the latest releases but charged lower admission prices. Interstate persuaded each of the eight major movie distributors, as a condition of being able to access Interstate’s first-run theaters, to impose onerous licensing terms on the second-run theaters that would result in their charging higher prices and otherwise limiting their competitiveness. At the time, it was not certain that the eight individual agreements, each involving Interstate and one distributor, would be found to be unlawful, so the Justice Department alleged that the eight distributors had agreed among themselves to go along with Interstate’s demands and effectively increase prices to the second-run theaters. The case went to the Supreme Court primarily on the issue of whether there was evidence to support the trial court’s conclusion that the eight had agreed among themselves. Law Review 851. Before the formal courtroom trial begins but after a lot of evidence in the form of documents and oral depositions has been amassed, a defendant can seek summary judgment on the grounds that there is no relevant dispute about the facts so that the only issues are legal. Once the courtroom trial has concluded, either just before the case is turned over to the jury or right after the jury has reached a verdict, the defendant can seek judgment as a matter of law (formerly called, respectively, a motion for directed verdict or a motion for judgment notwithstanding the verdict (JNOV)). Since juries do not publish opinions, virtually all of the significant ‘law’ involving cartels and circumstantial evidence comes in response to one of these motions. 20 Interstate Circuit, Inc. v United States, 306 US 208 (1939) (hereafter Interstate Circuit). 21 First-run theaters were a critical part of the movie business before the emergence of television advertising of newly released films. Today, the latest blockbuster-to-be is released simultaneously in many theaters in all major metropolitan areas, accompanied by massive television advertising to generate demand to see the film. But back before television, a film would be released very selectively – typically in only a single theater (the first-run theater) in most markets – and demand would be created by word of mouth from those who saw the film in the first weeks or months. Only much later would it be released more widely. For some history of this and other practices, see Michael Conant, Antitrust in the Motion Picture Industry: Economic and Legal Analysis (University of California Press 1960).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 8 /

Date: 7/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

64 Comparative competition law

The Court, in upholding the trial court’s finding of a horizontal agreement, emphasized two related aspects of the emergence of the eight essentially identical agreements between Interstate and each of the distributors. The first was that the new contracts represented a ‘radical departure’ from the previous arrangements between Interstate and the distributors and the distributors and the second-run theaters.22 Just like a buyer receiving eight sealed bids identical to the second decimal point, the Court was skeptical that this could have happened without the eight having been in communication (and ‘agreement’) with one another. Second, the Court remarked that, in part because of the radical departure, there was great risk to any distributor that went along with Interstate’s request without some assurance that the other distributors would take the same path.23 So while the parallel conduct presented a mystery, for the trial court (and ultimately the Supreme Court), the hypothesis of agreement was the most plausible answer to the mystery. One might imagine the Justice Department’s closing argument to the Court to have been something along the lines: ‘How could this possibly have happened unless the distributors had reached some prior agreement?’24 22 Interstate Circuit, 306 US at 222: ‘There was risk, too, that, without agreement, diversity of action would follow. Compliance with the proposals involved a radical departure from the previous business practices of the industry and a drastic increase in admission prices of most of the subsequent-run theaters.’ 23 ‘Each was aware that all were in active competition, and that, without substantially unanimous action with respect to the restrictions for any given territory, there was risk of a substantial loss of the business and goodwill of the subsequent-run and independent exhibitors, but that, with it, there was the prospect of increased profits. There was therefore strong motive for concerted action.’: ibid. 24 A potentially complicating factor was the fact that Interstate had sent identical letters, containing its demands, to each of the eight distributors, and using the ‘cc:’ designation, made each of the eight aware that its competitors were receiving the identical letter. This is a complicating factor because it might provide an alternative explanation for the parallel conduct by eliminating the coincidence and alleviating the risk at least to some extent. However, the letter proved not to be an obstacle to the Supreme Court’s finding of an illegal agreement for two reasons. First, the actual terms of the contracts that eventually came into being differed in significant ways from those proposed in Interstate’s letter, suggesting that there had to have been some discussions subsequent to the letter. Second, even if the letter provided a complete explanation for the ‘coincidence,’ the Court was prepared to find an unlawful conspiracy stemming from the letter itself. ‘It is elementary that an unlawful conspiracy may be and often is formed without simultaneous action or agreement on the part of the

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 9 /

Date: 7/9

JOBNAME: Duns PAGE: 10 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 65

A European case with some similarities to Interstate Circuit is the Dyestuffs case.25 In an industry described as an oligopoly, there was a series of near identical, near simultaneous price increases. The circumstances suggested that while there was an element of price leader/price follower in connection with (at least some of) the increases, some of the actions seemed hard to reconcile with simple price leadership or traditional oligopolistic interdependence. The Court did not state unambiguously that there must have been a traditional agreement, stating that a concerted practice did not have to have all of the elements of a contract. Indeed, in parts, the Court seemed almost to echo the Supreme Court’s dictum in Interstate Circuit regarding invitation followed by acceptance,26 which would not provide a basis to exonerate classic price leadership in an oligopolistic industry. A case often paired in law school casebooks with Interstate Circuit is Theatre Enterprises27 because it demonstrates a situation in which parallel conduct has a perfectly innocent explanation. The Crest Theater was a movie theater in suburban Baltimore that approached a number of major film distributors (basically the same cast of characters as those in Interstate Circuit) to be allowed to show ‘first-run’ films. The tradition in the industry, not really challenged in the case, was that only one theater in a metropolitan area such as Baltimore would get any new film; i.e., there was area-wide exclusivity. Each of the distributors rebuffed The Crest in favor of awarding its first-run films to one of the large theaters in downtown Baltimore. The Crest claimed that the parallel conduct was best explained as the result of an agreement among the distributors. When the jury returned a verdict for the distributor defendants, The Crest sought a judgment notwithstanding the verdict, arguing that the trial judge should have directed a verdict for the plaintiff. Here the Court thought that the ‘coincidence’ could easily be explained as the result of each distributor independently reaching the same conclusion; i.e., if it had to choose between licensing a downtown theater and one in suburban conspirators . … Acceptance by competitors, without previous agreement, of an invitation to participate in a plan, the necessary consequence of which, if carried out, is restraint of interstate commerce, is sufficient to establish an unlawful conspiracy under the Sherman Act.’ Interstate Circuit, 306 US at 227. Later in this chapter we will discuss what use might be made of the Court’s rather sweeping language in the context of parallelism based solely on oligopolistic interdependence. 25 Case 48/69 Imperial Chemical Indus. v Commission [1972] ECR 619. 26 Ibid. 27 Theatre Enterprises v Paramount Film Distributing Corp., 346 US 537.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 10 /

Date: 13/10

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

66 Comparative competition law

Baltimore, the downtown theater was the obvious choice regardless of what the other distributors decided.28 Therefore, no prior agreement was needed to explain each distributor’s conduct. The ‘coincidence’ was the result of each distributor independently reacting to the same basic market forces. It would appear, therefore, that the framework is complete. There is widespread agreement that simply parallel conduct is not enough to permit an inference of agreement. While parallel conduct can be interdependent and the result of a prior agreement, it can also be rational, independent, ‘competitive’ behavior. Each firm would behave the same way regardless of what it anticipated its competitors might do, and hence no prior agreement is required to explain the parallelism. The task, then, it might seem, is simply to determine when parallel conduct is interdependent and when it is independent. The former suggests the presence of an agreement; the latter does not.

5. OLIGOPOLY COMPLICATES THE ANALYSIS/PROOF Unfortunately, parallel conduct that results from agreement on one hand and parallel independent (and ‘competitive’) conduct on the other are not the only possibilities. Consider the now well-used example of the only two gas stations for miles around (i.e., a separate geographic ‘market’) located right across the road from one another. Assume that, to this point, prices have been parallel (i.e., identical) but competitive, at $1 per gallon for ‘regular’ gas.29 Call the two owners A and B, respectively. Owner A begins pondering whether his station can be made more profitable and consults an economist who, after doing some fancy econometric research (and charging appropriately), announces that she has good news. It turns out that market demand for gasoline is highly inelastic. If the market price were to double to $2, overall market demand would decrease by less than 50 percent, and hence overall revenues (and a fortiori profits) for the market as a whole (i.e., both stations) would increase. 28 Of course, each distributor may have expected the others to act similarly but that was not the result of any prior communication of agreement since each distributor would have rejected The Crest even if it knew that one or more of the others might not. Thus, ‘regardless’ becomes a key word in the defense’s dictionary. 29 The example is badly dated with respect to the price, but it keeps the numbers easier.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 11 /

Date: 7/9

JOBNAME: Duns PAGE: 12 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 67

But the advice comes with some bad news as well. If A increases price to $2, while B keeps its price at $1, B will get all the business, rendering B highly profitable, and driving A into bankruptcy. A, sensing the risk of increasing price in this situation, contemplates calling B and suggesting they agree to raise price together but is cautioned by his antitrust lawyer that this might mean a hefty fine and a long prison sentence. He is about to show the economist the door (without paying her bill) as having provided accurate but useless information. The economist saves the day, however, by asserting that, even though no direct communication between A and B would be allowed, she is quite certain that, if A initiates a price increase, B will follow promptly and they can continue ‘sharing’ the market at a price of $2 and enjoy the supra-normal profits she originally predicted. When asked how she can be so sure, she responds simply: If you increase price to $2 (which can be assumed to be the profit-maximizing market price, the exact price A and B would have agreed to had they met face to face), B will be fully aware of what you have done because your prices are fully visible across the street. So if B wants to follow you, he can. And he will want to. Technically there are three possible outcomes: (a) both A and B charge $2; (b) A charges $2 and B keeps the price at $1; and (c) both A and B charge $1. Now I have conceded that option (b) is the best possible outcome for B30 and the worst possible outcome for A. But B knows that option (b) is not a realistic equilibrium. B knows that if he doesn’t match A, A will know that almost immediately, and will react by reversing the price increase. B won’t get that much business in the time it will take A to react, and the end result, therefore, is that both A and B are stuck at $1 for the foreseeable future. Therefore, of the only two realistic scenarios, B’s matching A’s price increase is clearly preferable for B. The result then is that A increases price to $2, B matches A’s price, and the $2 price remains in effect indefinitely. The $2 price is parallel, interdependent (i.e., A would not charge $2 regardless of what he thought B might do; if he knew that B would not follow, he would keep his price at $1) and, by assumption, not competitive. But it is the result not of any formal agreement, but rather of the oligopolistic structure of the market. For future reference we can call the resulting outcome that of ‘pure oligopoly.’ 30 I have assumed this to be the case. It would not take very unrealistic assumptions about fixed and variable costs to generate that result.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 12 /

Date: 7/9

JOBNAME: Duns PAGE: 13 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

68 Comparative competition law

From the consumer’s perspective, the oligopoly price is just as bad as the cartel price.31 Therefore, given that we have already determined that Section 1 of the Sherman Act does not require a legally binding contract to make up an ‘agreement,’ the natural question is whether the concept of agreement can be stretched to include the pure oligopoly outcome. The language of Interstate Circuit in referring to the letter might be called on here. A’s initial posting of the $2 price would constitute the ‘invitation’ and B’s matching that price would constitute the ‘acceptance.’ However, virtually all courts (and most commentators) that have considered the issue have concluded that ‘pure’ oligopoly pricing does not violate Section 1 even when the result is supra-competitive prices. Typical of courts’ analysis is the following passage by then-Judge Stephen Breyer: Courts have noted that the Sherman Act prohibits agreements, and they have almost uniformly held, at least in the pricing area, that such individual pricing decisions (even when each firm rests its own decision upon its belief that competitors will do the same) do not constitute an unlawful agreement under section 1 of the Sherman Act. That is not because such pricing is desirable (it is not), but because it is close to impossible to devise a judicially enforceable remedy for ‘interdependent’ pricing.32

The European Court of Justice (ECJ) has apparently reached the same conclusion. In 1993 the Court, in commenting on a series of parallel price increases, observed that, although the law prohibits collusion, it does not deprive firms of ‘the right to adapt themselves intelligently to the existing and anticipated conduct of their competitors.’33 And, even 31

To be fair, many standard oligopoly models (such as the Cournot model) compute the oligopoly price as above the competitive level but not quite as high as the pure monopoly (or cartel) price: John E Lopatka and William H Page, ‘Economic Authority and the Limits of Expertise in Antitrust Cases’ (2005) 90 Cornell Law Review 617. For purposes of the discussion, however, all that matters is the conclusion that the oligopoly equilibrium price is above the competitive level. 32 Clamp-All Corp. v Cast Iron Soil Pipe Inst., 851 F 2d 478, 484 (1st Cir 1988) (internal citations omitted). I made the same point some 30 years ago, arguing that there is no ‘culpable’ behavior in oligopoly pricing and that a court could not require a firm to pretend that it was not operating in an interdependent environment: George A Hay, ‘Oligopoly, Shared Monopoly, and Antitrust Law’ (1982) 67 Cornell Law Review 439. 33 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 Ahlstrom Osakeyhtio v Commission (Woodpulp II) [1993] ECR I-1307 para 71.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 13 /

Date: 13/10

JOBNAME: Duns PAGE: 14 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 69

more directly, the Court said that ‘the parallelism of prices and the price trends may be satisfactorily explained by the oligopolistic tendencies of the market …’34 A partial but notable dissent from the above is from Judge Richard Posner, who, early in his career as an academic, advocated applying the Sherman Act to oligopoly pricing, and still believes that the language of the Act is broad enough to encompass such pricing (which he would call a ‘tacit agreement’). But apparently Judge Posner accepts the proposition that courts have not embraced this possibility (and will not).35 More recently, Lewis Kaplow has sought to revive the debate, arguing that oligopoly pricing can and should be covered.36 However, assuming no change in the legal status of oligopoly pricing is imminent, a court presented with evidence of parallel behavior (especially pricing) must consider that there are three possible explanations: (a) the conduct is parallel but independent and competitive – neither harm nor foul; (b) the behavior is parallel, interdependent and noncompetitive (e.g., prices are at supra-competitive levels), but the result solely of oligopolistic interdependence – harm but apparently no foul; (c) the behavior is parallel, interdependent, non-competitive and the result of an agreement – both harm and foul. If the plaintiff has only circumstantial evidence, that evidence must be capable of distinguishing the last explanation (agreement) from either of the first two.

6. DISTINGUISHING AGREEMENT FROM OLIGOPOLISTIC INTERDEPENDENCE While the coincidence/risk analysis identified in Interstate Circuit can still provide the framework on which any pattern of circumstantial 34

Ibid para 126. In re High Fructose Corn Syrup Antitrust Litigation, 295 F 3d 651, 654 (7th Cir. 2002) (hereafter ‘High Fructose Corn Syrup’): ‘[The] statutory language [of section 1 of the Sherman Act] is broad enough … to encompass a purely tacit agreement to fix prices, that is, an agreement made without any actual communication among the parties to the agreement. … Nevertheless, it is generally believed, and the plaintiffs implicitly accept, that an express, manifested agreement, and thus an agreement involving actual, verbalized communication, must be proved in order for a price-fixing conspiracy to be actionable under the Sherman Act.’). 36 Louis Kaplow, ‘An Economic Approach to Price-fixing’ (2011) 77 Antitrust Law Journal 343. 35

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 14 /

Date: 13/10

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

70 Comparative competition law

evidence must be hung, the addition of oligopoly as a complete explanation for supra-competitive pricing complicates the process considerably. Moreover, it is a very abstract and theoretical framework. Subsequent trial and appellate courts have tried to reduce the level of generality one notch by referring to ‘plus factors’ and asserting that, in the absence of direct evidence of an agreement, plaintiffs must, in addition to establishing parallel conduct, identify certain ‘plus factors’ that make the hypothesis of agreement more plausible than any competing hypothesis of either parallel, but competitive conduct, of the kind found in Theatre Enterprises, or parallel non-competitive conduct that is the result solely of oligopolistic interdependence. Unfortunately, although perhaps inevitably, there is no uniformity among courts on what can constitute a plus factor, but there are some commonalities. Perhaps the most common plus factor identified by courts is the structure of the industry. The idea is that certain industry structures are simply not conducive to successful coordination on price. If the industry is unconcentrated, the product is highly differentiated, demand is very elastic, the market is unconcentrated on the buying side, and there are few barriers to entry (other factors could be added as well), then even the most elaborate formal cartel is unlikely to be successful.37 If collusion stands little or no chance of success, why would firms expose themselves to the severe penalties that would flow from detection and prosecution? (If there were an actual cartel agreement, the fact that the cartel was unsuccessful would not immunize it from criminal sanctions although it might diminish or eliminate the damages it would be liable for.) However, while structural factors indicating that the industry is not susceptible to successful collusion may help to eliminate some pricefixing claims, factors suggesting that the industry is susceptible to collusion may also be consistent with the possibility that the parallel prices (even if supra-competitive) could be the result of simple oligopolistic interdependence without any explicit agreement. As Judge Posner asserts in High Fructose, the plaintiffs ‘must present evidence that would enable a reasonable jury to reject the hypothesis that the defendants foreswore price competition without actually agreeing to do so.’38 Therefore we need to look at some of the other factors identified by courts. 37 George A Hay and Daniel Kelley, ‘An Empirical Survey of Price-fixing Conspiracies’ (1974) 17 Journal of Law and Economics 13. See also High Fructose Corn Syrup, 295 F 3d 651. 38 High Fructose Corn Syrup, 295 F 3d 661.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 15 /

Date: 7/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 71

Typical of the approach is the analysis in Blomkest, a case alleging an agreement on price among producers of potash. The Court refers to three plus factors identified by the plaintiff: (1) inter-firm communications between the producers; (2) the producers’ acts against self-interest; and (3) econometric models purporting to prove that prices would have been lower in the absence of collusion.39 (Elsewhere the dissent also refers to the structural conditions of the industry as not being incompatible with collusion.40) Considering the factors in reverse order, the third factor in Blomkest (econometric models), seems to be just another way of saying that prices were above the competitive level. If that is the case, such a plus factor serves to rule out the explanation of parallel, but competitive conduct (Theatre Enterprises) but still leaves either collusion or oligopoly as competing explanations.41 Therefore it falls in the same category as structural conditions conducive to collusion. The second factor (producers’ acts against self-interest) seems almost an oxymoron, since economists assume that firms always act in their own self-interest. What is meant, however, is that the producers’ actions don’t make sense unless we assume that there was an agreement in place. This comes close to simply restating the coincidence/risk framework of Interstate. The behavior is too much of a coincidence or too risky for each firm to undertake without some prior understanding that others would do the same. But this factor falls in the same category as the structural and economic evidence unless the plaintiff can make the case that the necessary ‘understanding’ is not simply a tacit understanding deriving solely from oligopolistic interdependence. This is where Interstate Circuit provides a nice model. In upholding the trial court’s finding of an (explicit) agreement, the Supreme Court determined in effect that the change in contracting was so radical and so risky (and differed from the initial proposal circulated by Interstate via the infamous letter) that 39 See Blomkest Fertilizer, Inc. v Potash Corp of Saskatchewan, 203 F 2d 1028 (8th Cir 2000). 40 Ibid 1044–51 (Gibson J, dissenting). 41 It is hard to say too much more about this factor because the court rejected the expert report that underlay the factor. It is conceivable (although unlikely in this case) that the plaintiff meant that prices were not only above the competitive level, but above the oligopoly level as well, assuming that the underlying oligopoly model predicts a price that is less than the monopoly or cartel price. This distinction will be discussed further when we get to ‘actions against self-interest.’

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 16 /

Date: 7/9

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

72 Comparative competition law

simple oligopolistic interdependence could not explain it. There must have been some explicit communications.42 Turning back to price-fixing cases, we can supplement our discussion of structural factors to make some good use of the concept of actions against self-interest. There may be some oligopoly situations where many of the structural conditions may be susceptible both to explicit collusion and to oligopolistic interdependence leading to supra-competitive prices, such as high concentration, inelastic demand, lack of concentration on the buying side, high barriers to entry, etc., but other structural factors may be more problematic. For example, if prices are not transparent (i.e., each competitor’s prices cannot, in the normal course, be observed by other competitors), this makes both arriving at an initial consensus on price (through price leadership, for example) and monitoring adherence to any supra-competitive price very difficult without formal communication and explicit agreement. Therefore if we observe each firm charging supra-competitive prices, we might conclude that it would not be in any one firm’s interest to have done so without some prior assurance that others would do the same, and that the structural conditions make it difficult/impossible for that assurance to have evolved simply from oligopolistic interdependence. This provides a nice segue to the final plus factor identified in Blomkest (the first one listed by the Court) – inter-firm communications.43 There was evidence that firms were engaged in the practice of price verification. This typically occurs when a customer reports to its usual supplier that a competing supplier has offered an unusually low price (in circumstances where actual sales prices are typically not a matter of public record). The usual supplier then contacts the competitor, reports the customer’s assertion, and asks if the assertion is correct. The custom is for the competitor to be truthful in revealing whether the 42 Of course, as mentioned above, the Court also determined that if the letter were enough to explain the parallel actions of the defendants, the letter (as an invitation) followed by acceptance would constitute an unlawful conspiracy. See Conant (n 21) and accompanying text. See also Interstate Circuit. 43 Some inter-firm communications may be circumstantial evidence in the most fundamental sense; that is while we do not have a video recording of the actual cartel meeting, there is enough evidence of communication (in circumstances where there is no obvious legitimate reason for communications between or among executives) to suggest that the parties were in fact discussing the prices to be charged. This would be similar in kind to evidence that key executives were all staying at the same hotel at the same time without an obvious explanation such as a trade association meeting. The discussion in the text has a different focus.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 17 /

Date: 7/9

JOBNAME: Duns PAGE: 18 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 73

customer’s assertion is accurate. The majority of the appellate panel had difficulty understanding how a discussion of past prices could possibly lead to any agreement as to prices to be charged in the future, in part perhaps because the majority thought that the essential function of any cartel is to achieve some kind of consensus on prices to be charged. If there are differences of opinion as to the correct ‘cartel price,’ these need to be worked out through some kind of direct communication. The dissent, however, identified a second critical function of any cartel, viz., to monitor, detect, and discourage deviations from any consensus price. In that light, the price verification functions as a convenient tool. If a firm knows that its ‘cheating’ will be detected (and presumably matched), there is less incentive to cheat in the first place. The fact that firms were engaged in price verification is taken as evidence that there must have been some underlying agreement on price, and that the verification served to support that agreement. Otherwise why would firms reveal sensitive information about their own transactions? If a firm really wanted to get more business away from its competitors by secretly offering discounts to customers, revealing its actual prices would throw away any advantage those discounts might have helped to secure. Thus, these kinds of inter-firm communications would be against the selfinterest of the producers unless there was an agreement in place that firms would generally try to adhere to a supra-competitive price.

7. FACILITATING PRACTICES Some courts have described practices such as inter-firm communications ‘facilitating practices.’ And in some of these cases, it is merely a label change. An example is then-Judge Sotomayor’s opinion in Todd v Exxon, which also involved an exchange of pricing information. The Court describes information exchange ‘as an example of a facilitating practice that can help support an inference of a price-fixing agreement.’44 The basic claim is that there was an actual agreement (on price). The so-called facilitating practice is relevant as evidence of this underlying agreement in the sense that engaging in the practice would not make sense absent the underlying agreement.45 44

Todd v Exxon Corp., 275 F 3d 191, 198 (2d Cir 2001). I have written extensively about the theory of facilitating practices and efforts to implement it: See for example George A Hay, ‘Facilitating Practices: The Ethyl Case 1984’ in John E Kwoka and Lawrence J White (eds), The Antitrust Revolution: Economics, Competition and Policy 182 (3rd edn, Oxford 45

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 18 /

Date: 7/9

JOBNAME: Duns PAGE: 19 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

74 Comparative competition law

In other cases invoking the facilitating practices label, the approach seems to be somewhat different. Imagine the following scenario somewhat like the potash industry in Blomkest: an industry is oligopolistically structured and almost all the related structural factors are supportive either of a formal cartel or of supra-competitive prices achieved merely through oligopolistic interdependence. But there is one factor that stands in the way of successful coordination merely through oligopolistic interdependence. Suppose for example that one competitor’s prices cannot be directly observed by another, but can be identified only (usually through customer contacts) after the passage of some time. This creates a risk for any firm seeking to implement a price increase. Because the initiator of the price increase knows that its price cannot be directly observed, it has little confidence (even if it was confident that others would want to follow) that rivals will be able follow it at least for some period of time, during which the initiator may stand to lose a lot of business.46 In response to the uncertainty, the firms may adopt practices that reduce it. The most obvious, and legally easiest to deal with, is where the firms agree to exchange certain information, such as the information about recent transactions described above as price verification, that allows competitors to determine whether rivals have been secretly engaged in price discounting. Earlier, the price verification was presented as circumstantial evidence of an underlying agreement. An alternative approach is to attack the agreement to exchange information as an agreement that is unlawful because it leads to a substantial lessening of price competition without a formal agreement once the uncertainty about rivals’ cooperation has been eliminated. Thus, the facilitating practice in effect becomes the violation. But what makes this approach work without clashing with the traditional hands-off approach to oligopolistic conduct is that there was an agreement to exchange the information or otherwise to engage in the ‘facilitating practice’ that can fit comfortably with University Press 1999); George A Hay, ‘Facilitating Practices’ in Wayne D Collins and Joseph Angland (eds), Issues in Competition Law and Policy 1189 (American Bar Association 2008); George A Hay, ‘Horizontal Agreements: Concept and Proof’ (2006) 51 The Antitrust Bulletin 877; George A Hay, ‘The Meaning of “Agreement” Under the Sherman Act: Thoughts from the “Facilitating Practices” Experience’ (2000) 16 Review of Industrial Organization 113 (2000). 46 The lack of price transparency also complicates the problem of cheating and that may have to be dealt with through other facilitating practices, such as most-favored-nations clauses, as discussed below in the text accompanying n 49.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 19 /

Date: 7/9

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 75

traditional interpretations of Section 147 or Article 10148 as outlawing agreements that restrict or distort competition. A more challenging problem is when it cannot be demonstrated that the facilitating practice was adopted via any agreement. So assume, for example, that a firm (and its competitors) without prior agreement to do so, adopt the practice of publicly announcing price increases that will not become effective for, say, 30 days. (‘Effective in 30 days, I plan to increase price by 10% to $300 per unit.’) This substantially eliminates the risk that rivals will be unable to follow if they would otherwise want to. The rivals are made aware of the proposed price increase and have up to 30 days to signal, by a similar announcement (without the lead time) that they plan to follow. Meanwhile, the initiator is not at risk of losing sales since, for the time being, its price is no higher than any of its rivals. If rivals don’t announce their intention to follow within the 30-day window, the initiator simply withdraws the increase. Since the rivals know this will happen, they have every incentive to follow the increase without the need for any direct communication. We do not believe that any direct communication was necessary for the oligopoly to achieve and sustain supra-competitive prices. The facilitating practice (the advance public announcement) was sufficient to do the trick.49 If that is the explanation, the legal analysis changes. Instead of saying that the practice is circumstantial evidence of a traditional agreement, the 47

Todd v Exxon Corp., 275 F.3d 191 (2d Cir. 2001). The opinion, discussed above (n 44), also indicates that a completely different way to package the information exchange is to focus on the agreement to exchange information as the illegal agreement rather than as circumstantial evidence of an agreement to fix prices: ibid 198. An agreement to exchange information will normally be easier to establish but the downside, as Judge Sotomayor points out, is that such agreements are generally treated under the rule of reason, so the plaintiff retains the burden (which is circumvented in per se cases) of showing that the agreement actually had an adverse effect on competition: ibid 198–99. There is a whole line of Supreme Court cases, starting with United States v Container Corp., 393 US 333 (1969), that focus on the agreement to exchange information as the target of inquiry. 48 The agreement to exchange information would be covered by the Council Regulation (EC) Guidelines on the Applicability of Article 101 of the Treaty on the Functioning of the European Union to Horizontal Co-Operation Agreements [2011] OJ C11/1 . 49 As suggested (n 47), I don’t claim that simply the announcement will necessarily bring about a supra-competitive equilibrium without direct communication, but simply that it could, either singly or in conjunction with other facilitating practices.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 20 /

Date: 7/9

JOBNAME: Duns PAGE: 21 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

76 Comparative competition law

argument is that the parallel supra-competitive prices, combined with the use of the facilitating practices, establish an illegal agreement (even if a tacit one). We do not believe (and are not trying to prove) that there was any formal hotel room conspiracy because we believe that the facilitating practices made such a meeting unnecessary. The legal question is whether a court that would reject pure oligopolistic interdependence as constituting an actionable agreement could nevertheless find that oligopolistic interdependence (intentionally) facilitated by certain specific practices forfeits the safe haven for pure oligopolistic interdependence (because now we have intentional, and therefore culpable, conduct).50 Advance public announcements of ‘proposed’ price increases is perhaps the most commonly identified facilitating practice. Another prominent one, also featured in the Ethyl case, was providing customers with a most-favored-nations clause (MFN). This can be useful if one obstacle to achieving a supra-competitive price without any formal communication is each competitor’s fear that another will secretly discount price below the ‘public’ (supra-competitive price). This can trigger pre-emptive price cutting by firms who otherwise might be willing to enjoy the oligopolistic equilibrium. Therefore, each firm (or at least each of the major firms) may wish to find a way to convince its competitors that it will not be initiating a discount. The MFN clause promises each customer that, if another customer gets a discount, that customer will be guaranteed the same discount (perhaps even retroactively).51 This in turn creates a disincentive for the firm to offer a price cut to a single customer to pick up some incremental business because the price cut must be passed on to other customers who might have been prepared to pay the ‘public’ price. If the firm’s rivals are aware that the form has an MFN, this might provide the needed assurance that permits a stable (non-collusive) supra-competitive oligopolistic equilibrium.

50

If the practices were not initiated for the purposes of facilitating oligopolistic interdependence but come to have that effect, the analysis might be different. This was one of the principal problems confronting the FTC in the Ethyl case: Ethyl Corp. v Federal Trade Commission, 729 F 2d 128 (1984); Hay, ‘Facilitating Practices’ (n 45). Other problems with the FTC’s theory in Ethyl are discussed below. 51 There might also have to be some mechanism made available for customers to monitor a firm’s pricing to be sure that no one else has received a better price. Both the MFN and the monitoring provisions were involved in the Department of Justice’s case against General Electric and Westinghouse. See United States v General Electric Co. (1977–1) Trade Cas. (CCH) 712 (1977).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 21 /

Date: 13/10

JOBNAME: Duns PAGE: 22 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

The meaning of ‘agreement’ 77

In all the above examples, it is assumed that there was no agreement to adopt the facilitating practices (even if rivals subsequently but independently behave similarly). Therefore the possibility of focusing on an agreement to do so, either under Section 1 or Article 101, is not available. Hence the claim is that the use (without prior agreement) of a practice that facilitates a supra-competitive oligopolistic equilibrium, taken as a package, constitutes an unlawful agreement on price. Such an approach has neither been endorsed (nor rejected) by the US Supreme Court. Two of the most prominent efforts resulted in settlements prior to any litigated judgment.52 The Federal Trade Commission’s (FTC) effort in Ethyl was rejected by the Court of Appeals for the Second Circuit,53 although it was in part the result of other evidentiary problems with the FTC’s case rather than a complete rejection of the theory.54 While at least one US appellate court seems to have embraced the theory,55 it remains an ‘experimental’ treatment.

8. CONCLUSION The trend toward convergence of substantive antitrust doctrine means that most jurisdictions now condemn agreements among competitors that fix 52

Ibid: see also In re Domestic Air Transportation Antitrust Litigation, DC NGa no. 1:90-CV-2484 MHS & MDL no. 861 (1992). 53 See EI Du Pont De Nemours & Company v Federal Trade Commission, Ethyl Corporation v Federal Trade Commission 729 F2d at 128 (hereafter ‘Ethyl’). Technically Ethyl did not involve an allegation of agreement since it was brought under Section 5 of the FTC Act, but the Second Circuit’s opinion could create problems for the use of a facilitating practices theory under traditional Section 1 allegations. 54 In addition to the possibility that the practices had legitimate business justifications, as discussed at n 52, the Court noted the existence of some non-trivial amount of discounting as well as non-price competition, so that there was not a complete elimination of competition. The Court also had doubts about whether, even if there was an absence of vigorous competition, how much of that could be attributed to the facilitating practices as opposed to the basic oligopolistic structure. 55 See In re Coordinated Pretrial Proceedings in Petroleum Products Litig., 906 F 2d 432, 446 (9th Cir 1990): ‘[a]lthough… mere proof of interdependent pricing, standing alone, may not serve as proof of an antitrust violation, we believe that the evidence concerning the purpose and effect of price announcements, when considered together with the evidence concerning the parallel pattern of price restorations, is sufficient to support a reasonable and permissible inference of an agreement, whether express or tacit, to raise or stabilize prices.’

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 22 /

Date: 13/10

JOBNAME: Duns PAGE: 23 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

78 Comparative competition law

prices. But that same convergence means that those same jurisdictions must wrestle with the problem of how to establish the existence of an agreement, especially in an oligopolistic industry where high prices could, at least in theory, be the result simply of oligopolistic interdependence. Do we condemn such interdependence? Do we ignore it and require an explicit agreement? Or is there some middle ground? This chapter has explored how the US and, to a lesser extent, the EU, have approached the problem of dealing with a cartel when there is no hard evidence of an explicit agreement. The first option is to try to prove the existence of an explicit agreement through circumstantial evidence; a second is to relax somewhat the requirement that there be an explicit agreement. The effort to find the perfect solution continues.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 05_Chapter4

/Pg. Position: 23 /

Date: 7/9

JOBNAME: Duns PAGE: 1 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

5. Anti-competitive agreements: The range of conduct caught John Duns

1. INTRODUCTION The previous chapter dealt with the concept of agreement, with its attendant uncertainties. The current chapter assumes that the requirement of agreement has been met and turns attention to the content of the agreement. What are the key issues confronting competition law and policy when faced with the terms of an agreement and how do jurisdictions seek to resolve these issues? The first difficulty arises from the sheer variety of agreements to be assessed. Probably more than any other area of competition law, the range of conduct calling for analysis in the area of anti-competitive agreements is itself so varied that this creates its own difficulties. Agreements may be entered into in order to conduct business transactions, to avoid competition, to share information, to promote standards, to generate efficiencies and so on. They may range from a hard-core cartel through to a joint venture to engage in research and development (R&D).1 It follows that agreements may be entered into for anticompetitive, pro-competitive or competitively neutral reasons. Indeed, the one agreement may well contain both pro-competitive and anticompetitive (and neutral) provisions. All this might suggest that, as a matter of competition policy, the only way to assess agreements would be to subject them to some sort of lessening-of-competition/rule-of-reason test. That is, given the range of agreements, how else could they be assessed for their anticompetitiveness but on a case-by-case basis? Hard-core cartels provide 1 Agreements may be vertical or horizontal. Vertical agreements, which involve firms at different functional levels, are of course very common in business. However, vertical agreements are the subject of later chapters (Chapters 9 and 10) and so are not considered here. This chapter is primarily concerned with horizontal agreements, that is agreements between competitors or potential competitors. Chapter 12 looks at the particular case of cartel agreements so these are not specifically dealt with in the current chapter.

79

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

80 Comparative competition law

the classic justification for competition law whereas the dynamic efficiency potentially generated by R&D agreements, in contrast, are at the heart of competition law for the opposite reason. However, while a case-by-case analysis might be appropriate from an economic perspective, such an approach would be unacceptable at law. As stated by the United States (US) Supreme Court in Arizona v Maricopa County Medical Society: The elaborate inquiry into the reasonableness of a challenged business practice entails significant costs. Litigation of the effect or purpose of a practice often is extensive and complex. Judges often lack the expert understanding of industrial market structures and behavior to determine with any confidence a practice’s effect on competition.2

Firms need guidance on what agreements are permissible if they are to organize their affairs to avoid anti-competitive agreements. Guidance is also necessary if society is to get the pro-competitive benefits of agreements. As a result of trying to balance the need for business certainty on the one hand with an effective assessment of whether the agreement is anti-competitive on the other, there are two recurring themes that confront competition law and policy in the area of anti-competitive agreements: (a)

Categorization: in order to provide guidance and certainty, guidelines on which agreements are allowed and which are not need to be provided. This involves both a system of categorization of agreements and then a consequent need to characterize a particular agreement to determine which category applies to it. Assessment: to the extent permitted by the categorization process, if agreements are to be assessed for their anti-competitiveness, there must be appropriate criteria developed for this task. One of the underlying issues this assessment process raises is that of the objectives of competition law. Many, but not all, jurisdictions accept that economic efficiency, in its various forms, is the appropriate criterion to assess anti-competitive conduct. The obvious illustration of a departure from the efficiency criterion here is the ‘common market’ objective in European law but there are others who argue for other non-efficiency values to be used to assess

(b)

2 457 US 332 at 343, citing US v Topco Associates, Inc 405 US 596, 609–610 (1972).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 3 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 81

conduct.3 Apart from the special case of the European common market objective, it is beyond the scope of this chapter to revisit this debate. In any event, as will hopefully be shown below, there is to a significant extent a general convergence on this issue, even between the US and Europe,4 and so we will see the promotion of competition (and efficiencies) as at least one of the most significant objectives of competition law. In the light of all this, the aim of this chapter is to draw on approaches taken in different jurisdictions in order to seek answers to the following questions: (a)

(b)

What does competition policy suggest as the most appropriate method of categorizing agreements? As a matter of practice, how successful have the courts and legislatures been in this characterization process? Does the case law effectively differentiate anti-competitive agreements from pro-competitive and competitively neutral agreements?

For the purposes of this analysis, the jurisdictions examined are primarily the US and the European Union (EU), as these are the most significant jurisdictions both in their own right and because of their influence on other jurisdictions. But other jurisdictions will also be considered as comparators. These jurisdictions include Australia, not only because this is the jurisdiction with which I am most familiar but also because it represents, in some respects, a typical response of a smaller jurisdiction that has been influenced by US developments. The structure of the chapter is as follows. Section 2 provides an overview of the key provisions that are relevant to the analysis. Section 3 then takes up an issue of definition. It seeks to clarify what agreements are being examined in this context. Section 4 provides an outline of the ways in which agreements can be anti-competitive. The idea here is to try 3 See for example Giorgio Monti, ‘Article 81 EC and Public Policy’ (2002) 39 Common Market Law Review 1057. 4 See generally Alison Jones, ‘Analysis of Agreements under US and EC Antitrust Law – Convergence or Divergence?’ (2006) 51 The Antitrust Bulletin 691; Daniel J Gifford and Robert T Kudrle, ‘European Union Competition Law and Policy: How Much Latitude for Convergence with the United States?’ (2003) 48 The Antitrust Bulletin 727; and Gunnar Niels and Ariaan Ten Kate, ‘Introduction: Antitrust in the US and the EU – Converging or Diverging Paths?’ (2004) 49 The Antitrust Bulletin 1.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 3 /

Date: 13/10

JOBNAME: Duns PAGE: 4 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

82 Comparative competition law

to discern some criteria that might be useful for analysis. A similar analysis is then undertaken in Section 5 to determine the circumstances in which agreements might be pro-competitive (or competitively neutral). In Section 6, the key issue of categorization is examined. This includes an elaboration of the objectives of the categorizing process and an assessment of how different jurisdictions approach this task, with the aim of drawing some conclusions on how best to categorize horizontal agreements for the purposes of competition law. In the course of the analysis, we will be considering how courts in different jurisdictions assess horizontal agreements for their anti-competitiveness. This will involve an analysis of some of the key types of agreements that have come before the courts and an assessment of how effectively the courts analyse these. Another critical aspect that will be taken up in the course of the analysis is the various guidelines promulgated by regulators. It will be seen that, given the need for guidance, the approach of the regulators to horizontal agreements becomes particularly important. Both the US and EU have developed Guidelines designed to assist firms to predict whether conduct is likely to be challenged by regulators. Conclusions are drawn in the final section.

2. AN OVERVIEW OF THE RELEVANT PROVISIONS 2.1 Section 1 of the Sherman Act (US) The key provision of US law dealing with anti-competitive agreements is s 1 of the Sherman Act 1890. This section provides that ‘every contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal’. It is enforced by civil and criminal sanctions and by private parties and regulators.5 States have their own antitrust

5 Section 5(a)(1) of the Federal Trade Commission Act provides that ‘Unfair methods of competition … are hereby declared unlawful.’ A breach of s 1 of the Sherman Act has been interpreted as an ‘unfair method of competition’ and, consequently, a breach of the Federal Trade Commission Act. Thus the Federal Trade Commission is in this way empowered to enforce s 1 of the Sherman Act. However, only the Department of Justice may seek criminal penalties for a breach of s 1.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 4 /

Date: 7/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 83

laws, which often reflect the terms of the federal legislation.6 The restriction of the prohibition to agreements ‘among the states or with foreign nations’ is the result of federal constitutional limitations. These requirements have been very broadly interpreted and so in practice this restriction has had limited impact on the scope of the prohibition.7 The key terms for present purposes are ‘contract, combination or conspiracy’. ‘Combination’ is the broadest of these and, as discussed in the previous chapter, encapsulates the notion of agreement.8 2.2 Article 101 of the Treaty on the Functioning of the European Union The relevant EU provision dealing with agreements is Article 101 of the Treaty on the Functioning of the European Union (TFEU). As can be seen below, para (1) prohibits anti-competitive ‘agreements’ and ‘concerted practices’. Paragraph (3) provides an exemption from para (1) in specified circumstances. (1)

The following shall be prohibited as incompatible with the common market: all agreements between undertakings, decisions by associations of undertakings and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market, and in particular those which: (a) directly or indirectly fix purchase or selling prices or any other trading conditions; (b) limit or control production, markets, technical development, or investment; (c) share markets or sources of supply;

6

For a comprehensive survey of state antitrust statutes, see American Bar Association Antitrust Section, State Antitrust Practice and Statutes (4th edn, 2009). 7 For a convenient explanation and discussion of relevant cases, see for example Ernest Gellhorn, William Kovacic and Stephen Calkins, Antitrust Law and Economics in a Nutshell (5th edn, Thomson West 2004) Chapter XIII. 8 The reference to ‘in the form or trust or otherwise’ adds nothing although does provide a reminder of the role that trusts originally played in cartels and hence the origins of the term ‘antitrust’. See generally William Letwin, Law and Economic Policy in America: The Evolution of the Sherman Antitrust Act (Random House 1965).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 5 /

Date: 7/9

JOBNAME: Duns PAGE: 6 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

84 Comparative competition law

(d)

apply dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; (e) make the conclusion of contracts subject to the acceptance by other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts. Any agreements or decisions prohibited pursuant to this Article shall be automatically void. The provisions of paragraph 1 may, however, be declared inapplicable in the case of: – Any agreement or category of agreements between undertakings; – Any decision of category of decisions by associations of undertakings; – Any concerted practice or category of concerted practices; which contribute to improving the production or distribution of goods or to promoting technical or economic progress, while allowing customers a fair share of the resulting benefit, and which does not: (a) impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives; (b) afford such undertakings the possibility of eliminating competition in respect of a substantial part of the products in question.

(2) (3)

These provisions came into effect in December 2009 and replaced the previous Article 81 of the EC Treaty. Article 101 is for all relevant purposes in the same terms as the former Article 81. 2.3 Other Jurisdictions Members of the EU may have their own provisions as well as being bound by Article 101 of the TFEU. These may mirror Article 101. This is the case, for example, with the UK Competition Act 1998.9 However, they may also go further. The UK Enterprise Act 2002, for example, contains criminal cartel provisions.10 Other jurisdictions vary considerably in their regulation of anticompetitive agreements and it is not possible to generalize, except perhaps to say that inevitably jurisdictions have been influenced by US or 9 10

Columns Design XML Ltd

/

See s 2(1). See s 188.

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 6 /

Date: 13/10

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 85

EU law. Australia, for example, has a general prohibition of anticompetitive agreements along the lines of s 1 of the Sherman Act but, in addition, has specific criminal cartel provisions.11 The interpretation of the general provision by the Australian courts provides an interesting contrast with the US approach and this is taken up below. The recent Malaysian Competition Act, on the other hand, is directly modelled on Article 101 of the TFEU.12

3. AN ISSUE OF DEFINITION The subject matter of this chapter is essentially agreements between competitors. The relevant US Guidelines13 refer to a ‘competitor collaboration’ as a set of one or more agreements, other than merger agreements, between or among competitors to engage in economic activity, and the economic activity resulting therefrom. … ‘Competitors’ encompasses both actual and potential competitors. Competitor collaborations involve one or more business activities, such as research and development, production, marketing, distribution, sales or purchasing. Information sharing and various trade association activities also may take place through competitor collaborations.14

The scope of what is encompassed by the notion of a competitor agreement raises three issues, the first two of which are explicitly suggested in the above description. One is that an agreement may in fact be sufficiently comprehensive and ‘permanent’ as to be more appropriately treated as a merger and hence outside the scope of anti-competitive agreement provisions. Joint ventures provide probably the most common example of this. The US Guidelines treat a horizontal agreement as a merger, and hence to be dealt with under the Merger Guidelines, if the collaboration ‘involves an efficiency-enhancing integration of economic activity in the relevant market’, it ‘eliminates all competition among the participants’ and ‘does not terminate within a sufficiently limited period’. 11

The general provision referred to here is s 45 of the Competition and Consumer Act 2010 (Cth). The cartel provisions are complex and contained in Div 1 of Part IV of that Act. 12 See s 3 of the Competition Act 2010 (Malaysia). 13 Federal Trade Commission and the US Department of Justice, Antitrust Guidelines for Collaborations Among Competitors, April 2000 (hereafter ‘the US Guidelines’). 14 Ibid para 2.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 7 /

Date: 7/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

86 Comparative competition law

For the purposes of this last requirement, there is a ‘10-year’ rule, though this is a guideline only and will depend on industry-specific circumstances.15 The relevant EU Guidelines are to similar effect.16 The second issue is the requirement that the agreement be between ‘competitors’. However, this is frequently either not a requirement of anti-competitive agreement provisions17 or, if it is, a requirement that is readily met, and so requires little consideration here. As in both EU and US Guidelines, ‘potential competitors’ will be included. A potential competitor is generally treated as one that would be able to enter the market in which the other party to the agreement already operates without substantial new investment.18 In some instances ‘conglomerate’ agreements will be considered, i.e. agreements between parties who operate in different markets. A third issue that jurisdictions have had to deal with in this context, and which requires greater consideration, is what might loosely be called agreements between ‘related companies’. It is generally accepted that the law should be concerned with substance not form in this area and so if an agreement is effectively one made within the one firm then this should be outside the scope of anti-competitive agreement prohibitions. So an agreement between a holding company and its subsidiary would be treated in the same manner, for example, as a decision reached between two divisions within the one firm. The difficulty is in defining the limits to this exemption. In the EU, agreements of this kind are referred to as ‘inter-enterprise agreements’ and the relevant case law as ‘the economic

15 Ibid para 5. See also James L Langenfeld and Louis Silvia, ‘Federal Trade Commission Horizontal Restraint Cases: An Update’ (2004) The Antitrust Bulletin 521, 523–524. 16 European Commission Guidelines on the applicability of Article 101 of the Treaty on the Functioning of the European Union to horizontal co-operation agreements [Official Journal C 11 of 14.1.2011] 1.1 (hereafter ‘the EU Guidelines’). 17 It is a requirement in some cases however. For example, the Australian cartel provisions have a ‘competition condition’ which requires that the parties to the agreement ‘are or are likely to be or but for any [agreement] would be or would likely to be in competition with each other’ in relation to matters covered by the agreement: see Competition and Consumer Act 2010 (Cth) s 44ZZRD(4). And section 4 of the Malaysian Competition Act 2010 requires that the agreement be ‘horizontal’, a term defined in s 2 of the Act as ‘an agreement between enterprises each of which operates at the same level in the production or distribution chain’. 18 Compare EU Guidelines (n 16) para 10.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 8 /

Date: 7/9

JOBNAME: Duns PAGE: 9 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 87

entity doctrine’. In the US it is referred to as the ‘single enterprise’ doctrine. 3.1 The EU Economic Entity Doctrine EU competition law applies to agreements between ‘undertakings’. This is a general term that is not defined in the Treaty. The European Court of Justice has stated that ‘the concept of an undertaking encompasses every entity engaged in an economic activity regardless of the legal status of the entity and the way in which it is financed’.19 The present point is that related companies may be treated as part of the one ‘undertaking’ and, if so, will be outside the reach of Article 101. As stated by the EU Commission in its Guidelines on agreements, ‘Article 101 only applies to agreements between independent undertakings’.20 When will companies be treated as a ‘single economic entity’? In principle, this is meant to be a ‘substance over form’ test. The issue is whether the relevant companies independently compete against each other or whether they are effectively one firm. The courts have emphasized the degree of control exercised by one company over another. As stated in the EU Guidelines: ‘When a company exercises decisive influence over another company they form a single economic entity and, hence, are part of the same undertaking.’21 Such influence is presumed in the case of a fully owned subsidiary.22 The presumption can be rebutted but the burden is on the party alleging independence.23 ‘Siblings’ of the same parent company are also treated as one.24 The doctrine is not

19

Hofner and Elser v Macrotron GmbH Case C 41/90 ECR I-1979, [1993] 4 CMLR 306 para 21, cited in Richard Whish, Competition Law (6th edn, Oxford University Press 2008) 83. ‘Undertaking’ has been interpreted broadly by the courts. It includes all natural and legal persons: see e.g. Case C-41/90 Hofner and Elser v Macroton [1991] ECR I-1979. See generally Whish (ibid) 83–97 for a comprehensive discussion. 20 EU Guidelines (n 16) para 11, citing Case C-73/95, Viho Europe BV v Commission [1996] ECR 1-5457. 21 Ibid para 11. 22 Case 107/82 AEG [1983] ECR-3151, para 50. 23 See e.g. Case T-112/05 Azko Nobel NV v Commission [2007] ECR II-000. 24 EU Guidelines (n 16) para 11. See generally Wouter PJ Wils, ‘The Undertaking as a Subject of EC Competition Law and the Imputation of Infringements to Natural or Legal Persons’ (2000) 25 European Law Review 103.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 9 /

Date: 13/10

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

88 Comparative competition law

confined to related companies. It has been applied, for example, to a principal and agent relationship.25 Inevitably there are grey areas. While the degree of control is used by the courts as a guide, uncertainty increases as the level of control moves away from 100 per cent. For example it may be unclear whether a joint venture company is separate from its owners.26 3.2 The US Single Enterprise Doctrine The US approach is along broadly similar lines to the EU position.27 In Copperweld Corp. v Independence Tube Corp.28 the US Supreme Court held that a parent corporation and its wholly owned subsidiary were effectively the same entity and so an agreement between them was not within s 1 of the Sherman Act. The Court stated that: [T]he coordinated activity of a parent and its wholly owned subsidiary must be viewed as that of a single enterprise … [They] have a complete unity of interest. Their objectives are common, not disparate; their general corporate actions are guided or determined not by two separate corporate consciousnesses, but one. They are not unlike a multiple team of horses drawing a vehicle under the control of a single driver. With or without a formal ‘agreement,’ the subsidiary acts for the benefit of the parent, its sole shareholder. If a parent and a wholly owned subsidiary do ‘agree’ to a course of action, there is no sudden joining of economic resources that had previously served different interests, and there is no justification for § 1 scrutiny.29

More recently, in Texaco, Inc. v Dagher,30 the Supreme Court held that an agreement on pricing by joint venturers Texaco, Inc. and Shell Oil Co 25 Case C-217/05 Confederacion Espanola de Empresarios de Estaciones de Servicio v Compania Espanola de Petroleos [2007] 4 CMLR 8661; Case T-325/01 Daimler Chrysler AG v Commission [2005] ECR II-3319. 26 Compare e.g. Gosme/Martell-DMP OJ [1991] L 185/23 [1992] 5 CMLR 586. Other consequences of the single economic doctrine include extending the liability of one company to other related companies: see generally Whish (n 19) 94. 27 See generally Judd E Stone and Joshua D Wright, ‘Antitrust Formalism is Dead! Long Live Antitrust Formalism! Some Implications of American Needle v. NFL’ [2010] Cato Supreme Court Review 369, 403; Nathaniel Grow, ‘American Needle and the Future of the Single Entity Defense Under Section One of the Sherman Act’ (2011) 48 American Business Law Review 449. 28 467 US 752 (1984). 29 467 US at 771. 30 547 US 1 (2006).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 10 /

Date: 7/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 89

was not within s 1. The Court referred to the joint venture being a ‘single entity’.31 In contrast, in American Needle v National Football League32 the Supreme Court held that the teams in the US National Football League did not constitute a single entity when they combined to form a joint venture to sell various merchandise. The Court held that the issue was whether the relevant firms were ‘separate economic actors’. 3.3 Other Approaches Some jurisdictions have followed this flexible approach. The recent Malaysian Act, for example, which is generally based on EU law, prohibits anti-competitive agreements between ‘enterprises’ and defines enterprises as ‘any entity carrying on commercial activities relating to goods or services, and for the purposes of this Act, a parent and subsidiary company shall be regarded as a single enterprise if, despite their separate legal entity, they form a single economic unit within which the subsidiaries do not enjoy real autonomy in determining their actions on the market’.33 In contrast, other jurisdictions have adopted a more formal approach to this issue. This has the advantages of simplicity and clarity but its drawback is its inflexibility. In the Australian anti-competitive agreement provisions, for example, there is an exemption if the only parties to the agreement are ‘related’ to one another.34 ‘Related’ in this context is defined in terms of a holding-subsidiary or ‘sibling’ relationship. The degree of control required to be a ‘holding company’ is ‘formal’ control, in the sense of the company owning more than 50 per cent of the subsidiary’s shareholding or of its board of directors, as opposed to a flexible text of ‘control’ which would take into account de facto control.35 The Canadian Competition Act is along similar lines.36

31 The Court based its decision on other grounds, however, and so the single entity issue did not ultimately constitute the basis of finding that s 1 did not apply. 32 130 S Ct 2201 (2010). 33 The prohibition is contained in s 3 of the Competition and Consumer Act 2010 (Cth). The definition of ‘enterprise’ is in s 2. 34 Competition and Consumer Act 2010 (Cth) ss 44ZSRN and 45(8). 35 See s 4A for the relevant definitions. 36 See Competition Act 1985 RSC, 1985, C-34, s 45(6) and the relevant definition (of ‘affiliated’) in s 2(2)–(4).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 11 /

Date: 7/9

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

90 Comparative competition law

4. AGREEMENTS AS ANTI-COMPETITIVE The essential concern with agreements is that they may be a means of creating or maintaining market power. As explained in the US Guidelines: ‘[c]ompetitor collaborations may harm competition and consumers by increasing the ability or incentive profitably to raise price above or reduce output, quality, service, or innovation below what likely would prevail in the absence of the relevant agreement.’37 Van den Bergh and Camesasca38 have pointed out that this may occur in any one of three ways. The most common is an agreement specifically directed towards raising prices, restricting output or market sharing. The second category is agreements designed to harm competitors by, for example, raising rivals’ costs through a boycott. The third category is agreements which lessen non-price competition, for example by limiting advertising. Anti-competitive agreements of these kinds create the inefficiencies that competition law seeks to prevent. Allocative inefficiencies occur through the higher prices that result – creating a deadweight loss and encouraging ‘rent seeking’ as parties seek to establish and enforce the agreement, both of which are welfare reducing. Productive and dynamic inefficiencies are also likely to result from the lack of competitive pressures. As a result, competition policy generally seeks to have firms make decisions independently. Despite the well-known fragility of cartels,39 the evidence is overwhelming that they persist and, despite competition law enforcement, clearly remain potentially very profitable. Unconstrained, firms will be tempted to replace market forces with ‘orderly marketing’. The extent to which an agreement will be anti-competitive will largely be determined by the combined market power of those who are party to it. As a matter of practicability, this is more likely to be found where fewer firms are required to meet this threshold requirement.40 Thus, dangerous cartels are most likely to succeed where the market has

37

US Guidelines (n 13) para 2.2. Roger van den Bergh and Peter D Camesasca, European Competition Law and Economics: A Comparative Perspective (2nd edn, Thomson Sweet & Maxwell 2006) 155–156. 39 See e.g. George J Stigler, ‘A Theory of Oligopoly’ (1966) 72 The Journal of Political Economy 44. 40 See generally Van den Bergh and Camesasca (n 38) 156–163. 38

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 12 /

Date: 7/9

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 91

characteristics such as high seller concentration and product homogeneity.41 For the same reason, the height of entry barriers is important here as otherwise fresh entry can undermine the agreement.42 Importantly, it also follows that where the collaborators’ market power is low, the anti-competitive impact of the agreement is correspondingly low.43 It is this fact which has given rise to the notion of creating ‘safe harbour’ thresholds, discussed below.

5. AGREEMENTS AS PRO-COMPETITIVE OR COMPETITIVELY NEUTRAL Despite the concerns that competitor collaboration is likely to create anti-competitive outcomes, it is also the case that firms necessarily deal with one another through agreement and that exchange of information can lead to efficiencies as well as market power. To quote the US Guidelines again: [c]onsumers may benefit from competitor collaborations in a variety of ways. For example, a competitor collaboration may enable participants to offer goods or services that are cheaper, more valuable to consumers, or brought to market faster than would be possible absent the collaboration. A collaboration may allow its participants to better use existing assets, or may provide incentives for them to make output-enhancing investments that would not occur absent the collaboration. The potential efficiencies from competitor collaboration may be achieved through a variety of contractual arrangements including joint ventures, trade or professional associations, licensing arrangements or strategic alliances.44

Cooperation between competitors may lead to economies of scale or scope. One of the most notable examples here is the economies that may be achieved through cooperation in R&D. Innovation (dynamic efficiency) as well as productive efficiencies may only be available through firms cooperating with each other – for example, to spread the costs and risks of R&D, to engage in technology transfer and to avoid free riding on individual innovation.45 Transaction cost economics are important 41

See generally Dennis W Carlton and Jeffrey M Perloff, Modern Industrial Organisation (4th edn, Pearson/Addison Wesley 2005). 42 See US Guidelines (n 13) para 3.35. 43 See EU Guidelines (n 16) paras 39–47. 44 US Guidelines (n 13) para 2.1. 45 See generally Van den Bergh and Camesasca (n 38) 169–177.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 13 /

Date: 7/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

92 Comparative competition law

here. It may be that agreements between firms are simply a means of reducing transaction costs and thus of leading to greater efficiency.46

6. CATEGORIES AND CHARACTERIZATION As noted above, categorizing conduct allows for greater certainty and this is important in commercial transactions. As explained by Lemley and Leslie, ‘the tradeoff is generally one between certainty and accuracy. If neither costs nor uncertainty were a problem, standards would always be preferable to rules in antitrust cases because they are more likely to achieve the right result. That is, everything would be subject to an overarching rule of reason.’47 If it is accepted that a case-by-case analysis of whether agreements are anti-competitive is too cumbersome and creates too much uncertainty, then agreements must be categorized in some manner. This categorization process involves questions of (i) choosing the appropriate categories and (ii) characterizing agreements, i.e. determining the category into which a particular agreement falls. An inappropriate use of categories may do more harm than good.48 6.1 Choosing the Appropriate Categories The simplest categorization is a dichotomy. Most jurisdictions adopt this approach and treat some types of agreements as inherently (per se) anti-competitive, with the remainder requiring a full assessment of their anti-competitiveness. This is the approach taken, for example, in the EU and Australia, as will be seen below. Even with this approach, it will be seen that decisions still have to be made about characterizing particular agreements. Nevertheless, the simplicity of this categorization approach has made it popular. More recently the US appears to have moved to a more graduated ‘series’ of categorizations, even to the point, it has been suggested in some decisions, of replacing distinct categories with a ‘continuum’. This avoids the arbitrariness of the simple dichotomous 46

See Thomas M Jorde and David J Teece, ‘Rule of Reason Analysis of Horizontal Arrangements: Agreements Designed to Advance Innovation and Commercialise Technology’ (1993) 61 Antitrust Law Journal 579. It is this feature which leads to the argument that markets and firms are part of the same continuum. 47 Mark A Lemley and Christopher R Leslie, ‘Categorical Analysis in Antitrust Jurisprudence’ (2008) 93 Iowa Law Review 1207, 1256. 48 See generally ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 14 /

Date: 7/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 93

approach but, as will be seen, does so at the risk of creating uncertainty and confusion. 6.2 Categorizing Agreements as Per Se Anti-competitive Categorizing certain agreements as prohibited per se is both consistent with economic theory49 and brings with it the advantages of a bright line rule.50 Firms can manage their conduct around the rule and in this way the prohibition discourages anti-competitive conduct and encourages efficient conduct. A bright line rule is particularly appropriate if criminal sanctions are attached to the conduct. The other primary advantage of a per se categorization is that it avoids or at least minimizes the necessity and scope of legal proceedings. As stated by the US Supreme Court in Northern Pacific Railway v US: This principle of per se unreasonableness not only makes the type of restraints which are proscribed by the Sherman Act more certain to the benefit of everyone concerned, but it also avoids the necessity for an incredibly complicated and prolonged economic investigation into the entire history of the industry involved, as well as related industries, in an effort to determine at large whether a particular restraint has been unreasonable – an inquiry so often wholly fruitless when undertaken.51

Or in the more colourful words of the US Supreme Court in FTC v Superior Court Trial Lawyers Association: The per se rules in antitrust law serve purposes analogous to per se restrictions upon, for example, stunt flying in congested areas or speeding. Laws prohibiting stunt flying or setting speed limits are justified by the State’s interest in protecting human life and property. Perhaps most violations of such rules actually cause no harm. No doubt many experienced drivers and pilots can operate much more safely, even at prohibited speeds, than the average citizen. … In part, the justification for these per se rules is rooted in administrative convenience. They are also supported, however, by the observation that every speeder and every stunt pilot poses some threat to the community. An unpredictable event may overwhelm the skills of the best driver or pilot, even if the proposed course of action was entirely prudent when initiated. A bad driver going slowly may be more dangerous than a good driver going quickly, but a good driver who obeys the law is safer still. 49 50 51

Columns Design XML Ltd

Van den Bergh and Camesasca (n 38) 178–185. See generally Lemley and Leslie (n 47) 1258–1259. 356 US 1 (1958) 5.

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 15 /

Date: 7/9

JOBNAME: Duns PAGE: 16 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

94 Comparative competition law So it is with boycotts and price fixing. Every such horizontal arrangement among competitors poses some threat to the free market.52

Thus per se categorization is designed to minimize both type I and type II errors. That is, a full, case-by-case assessment of agreements would create uncertainty. To avoid the costs associated with this, firms will be tempted to avoid them and thus there is a risk that pro-competitive agreements will be discouraged (type I error). Equally, the rigours of a full assessment are a burden on regulators and potential plaintiffs. The risk here is that there will be under-enforcement of anti-competitive agreements (type II error).53 The strengths of this per se categorization also constitute its potential weaknesses. This was well explained by Justice Marshall in the Container Corp case: Per se rules always contain a degree of arbitrariness. They are justified on the assumption that the gains from imposition of the rule far outweigh the losses and that significant administrative advantages will result. In other words, the potential competitive harm plus the administrative costs of determining in what particular situations the practice may be harmful must far outweigh the benefits that may result. If the potential benefits in the aggregate are outweighed to this degree, then they are simply not worth identifying in individual cases.54

Set out below is how this issue of per se characterization has been approached in different jurisdictions. 6.2.1 USA At least initially, the US had two categories for agreements to be assessed under s 1 of the Sherman Act. Either the agreement fell into a category that was prohibited per se or, if not, it was to be assessed on a case-by-case basis to determine whether or not it was anti-competitive (‘in restraint of trade’). This was explained by the Supreme Court in National Society of Professional Engineers v US: 52

493 US 411, 433–434 (1990). Matthew Bennett and Phillip Collins, ‘The Law and Economics of Information Sharing: the Good, the Bad and the Ugly’ (2010) 6 European Competition Law Journal 311. 54 United States v Container Corp., 393 US 333, 341 (1969) (Marshall, J, dissenting). Although Justice Marshall was in dissent in relation to the outcome in this case he was not dissenting on this point. See generally Paul Dennis, ‘Focusing on the Characterization of Per Se Unlawful Horizontal Restraints’ (1991) 36 Antitrust Bulletin 641; and Lemlie and Leslie (n 47). 53

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 16 /

Date: 13/10

JOBNAME: Duns PAGE: 17 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 95 There are, thus, two complementary categories of antitrust analysis. In the first category are agreements whose nature and necessary effect are so plainly anti-competitive that no elaborate study of the industry is needed to establish their illegality – they are ‘illegal per se.’ In the second category are agreements whose competitive effect can only be evaluated by analyzing the facts peculiar to the business, the history of the restraint, and the reasons why it was imposed. In either event, the purpose of the analysis is to form a judgment about the competitive significance of the restraint; it is not to decide whether a policy favoring competition is in the public interest, or in the interest of the members of an industry. Subject to exceptions defined by statute, that policy decision has been made by the Congress.55

This simple bifurcation of agreements had few refinements. Even the issue of the parties’ market power was irrelevant to this categorization process.56 The question was essentially whether the agreement was of a kind that was ‘plainly anti-competitive’. Although the courts claimed some discretion over the kinds of agreements that fell within this per se category,57 such agreements were more or less settled at an early stage.58 These were agreements to fix prices or reduce output,59 to divide up the market (e.g. territories or customers)60 or rig bids. One qualification was that only ‘naked’ agreements of these types were prohibited per se. That is, if the price-fixing etc provision were ‘ancillary’ to a lawful object, then the per se label would not be applied.61 55

435 US 679 at 692 (1978) (hereafter ‘Professional Engineers’). Federal Trade Commission v Superior Court Trial Lawyers 403 US 411 (1990) 435–436; National Collegiate Athletic Association v Board of Regents of the University of Oklahoma 468 US 85 (184) 109 (‘As a matter of law, the absence of proof of market power does not justify a naked restriction on price or output’.). 57 For example, the Court in US v Microsoft Corp 253 F 3d 34 (2001) refused to treat a tie as per se anti-competitive in the computer software context on the basis that the Court had insufficient experience of such conduct in ‘new economy’ products. 58 See generally US Guidelines (n 13) para 1.2 where per se agreements are summarized. 59 US v Trenton Pottery 273 US 392 (1927); US v Socony-Vacuum Oil Co 310 US 150 (1940); FTC v Superior Court Trial Lawyers Association 493 US 411, 432–436 (1990). 60 US v Topco Associates, Inc 405 US 596; Palmer v BRG of Georgia Inc 498 US 46 (1990). 61 Broadcast Music, Inc v Columbia Broadcasting Sys 441 US 1, 19–20 (1979); see also National Bankcard Corp v Visa USA Inc 779 F 2d 958 and National Collegiate Athletic Association v Board of Regents of the University of Oklahoma 468 US 85. 56

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 17 /

Date: 13/10

JOBNAME: Duns PAGE: 18 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

96 Comparative competition law

Another qualification more troubling to the categorization process was that, in the case of boycotts, the Court required a number of requirements to be met before the per se label was applied.62 Significantly, one requirement was that the parties to the agreement have market power. Another was that the boycott be ultimately targeted at a competitor of the parties to the boycotting agreement.63 This issue of ‘modifying’ the per se category is considered further below. 6.2.2 EU Article 101(1) of the TFEU prohibits agreements and concerted practices ‘which have as their object or effect the prevention, restriction or distortion of competition’. The courts, in their interpretation of ‘object or effect’, have made it clear that these are alternatives and are not to be read cumulatively.64 Thus they are directed at two different categories of agreement or concerted practice. Although not necessarily obvious from the wording itself, agreements prohibited by ‘object’ are those prohibited per se. As explained by the European Court of Justice: ‘[t]he distinction between “infringements by object” and “infringements by effect” arises from the fact that certain forms of collusion between undertakings can be regarded, by their very nature, as being injurious to the proper functioning of normal competition.’65 The EU Guidelines are to similar effect: ‘[r]estrictions of competition by object are those that by their very nature have the potential to restrict competition within the meaning of Article 101(1). It is not necessary to examine the actual or potential effects of an agreement on the market once its anti-competitive object has been established.’66 This categorization into ‘object’ and ‘effect’ is strongly reminiscent of the US per se/rule-of-reason dichotomy. It is true that, at least nominally, there are some differences between the US and EU positions. The first is that under the EU provisions there is the possibility of exemption from 62 A similar approach has been taken to agreements with ‘ties’: see Jefferson Parish Hospital v Hyde 446 US 2 (1984); US v Microsoft Corp 253 F 3d 34 (2001). 63 See for example FTC v Indiana Federation of Dentists 476 US 462 (1986). 64 Case 8/08 [2009] T Mobile Netherlands BV v Raad van Bestuur van de Nederlandse Mededingingsautoritleil, para 28. 65 Ibid para 29, referring to Case C-209/07 Competition Authority v Beef Industry Development Society [2008] ECR I-8637. See also GlaxoSmithKline Services Unlimited v Commission of the European Communities [2010] 4 CMLR 2, [2009] ECR I-929, para 55. 66 EU Guidelines (n 16) para 24, footnotes omitted.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 18 /

Date: 7/9

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 97

Art 101(1) by virtue of Art 101(3). There is no equivalent exemption option in US antitrust law. The scope of this exemption is considered below. Secondly, EU case law has required that all agreements, to come within Art 101(1), must have an ‘appreciable’ impact on competition.67 Thus even agreements that have an anti-competitive ‘object’ will only be within Art 101(1) if some impact on competition can be established. As seen above, this is generally not required in US antitrust law. But these distinctions are more theoretical than practical. As a matter of practice, once courts and regulators have found that an agreement is one that falls within the ‘object’ category, it will automatically be in breach of Art 101(1) and the two qualifications referred to will not save it. In this sense, EU law has much in common with US antitrust law in its categorization process.68 This similarity extends to an ‘ancillary’ doctrine. That is, if the relevant restriction is ‘ancillary’ to a lawful objective, it will not be in breach of Art 101(1). A typical example might be a provision in a joint venture agreement in which the parties agree not to compete. Agreements struck within trade associations may also be saved under this doctrine.69 An ancillary restriction is protected only so long as it is necessary to achieve the lawful objective.70 In terms of the types of cases that have been treated as prohibited ‘by object’, these also bear close similarity with the US case law. Price fixing and market sharing71 have been held to be agreements with an anticompetitive ‘object’. So has bid rigging.72 The list may be somewhat

67 Case 22/71 Beguelin Import Co v GL Import-Export SA [1971] ECR 949, [1972] CMLR 81. 68 Jones (n 4). 69 See for example Case C-250/92 Gottrup Klim v DLG [1994] ECR I-5641, [1996] 4 CMLR 191. 70 See for example Reuters v BASF [1976] OJ L254/40, [1976] 2 CMLR D44; Case 42/84 Remia BV and NV Verenigde Bedrijven Nutricia v Commission [1985] ECR 2545, [1987] 1 CMLR 1; Case C-250/92 Gottrup-Klim v Dansk Landbrugs Govvareselskab AmbA [1994] ECR I-5641, [1996] 4 CMLR 191; Case T-112/99 Metropole Television (M6) v Commission [2001] ECR II-2459, [2001] 5 CMLR 33. 71 European Night Services Ltd v Commission [1998] ECR II-3141, [1998] 5 CMLR 718. For market sharing (parallel import bans) see also Joined Cases 56/64 and 58/64 Consten and Grundig v Commission [1966] ECR 299 (compare Case T-168/01 GlaxoSmithKline v Commission [2006] ECR II-2969, [2006] 5 CMLR 1589). 72 Case T-29/92 Vereniging van samenwerkende prijsregelended oranisaties in de bouwnijverheid (SPO I) v Commission [1995] ECR II-289.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 19 /

Date: 7/9

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

98 Comparative competition law

longer as it is likely that the agreements listed in Art 101(1) are to be treated as anti-competitive by object,73 namely: (a)

directly or indirectly fix purchase or selling prices or any other trading conditions; limit or control production, markets, technical development, or investment; share markets or sources of supply; apply dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage; and make the conclusion of contracts subject to the acceptance by other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject of such contracts.

(b) (c) (d)

(e)

Whish suggests that output restrictions, exchange of future pricing information and ‘collective exclusionary dealings’ should also be included in this list.74 6.2.3 Other jurisdictions Other jurisdictions adopt a variety of approaches but generally follow a similar path. The Malaysian Competition Act, for example, directly adopts the EU ‘object or effect’ categorization and then explicitly deems price fixing, market sharing, output restrictions and bid rigging to be anti-competitive by object.75 The Canadian Competition Act explicitly prohibits agreements which contain price fixing, output restrictions or market sharing.76 The Australian Act has a similar list but then, oddly, adds a broad definition of collective boycotts.77 There are inevitably more subtle differences between jurisdictions however, as each reflects its own legal culture and infrastructure. These 73

Joanna Goyder, Albertina Albors-Llorens, Goyder’s EC Competition Law (5th edn, Oxford University Press 2009) 118. However, Whish suggests that the courts will retain some discretion to refine or reject provisions listed in Art 101(1): Whish (n 19) 120–122. 74 Citing Nederlandse Federative Vereniging voor de Grootlandel op Elektrotechnisch Gebied and Technische Unie (FEG and TU) [2000] OJ L 39/1, [2000] 4 CMLR 1208, para 105; Whish (n 19) 119. 75 Competition Act 2010 (Malaysia) s 4. 76 Competition Act 1985 (Canada) s 45. 77 Competition and Consumer Act 2010 (Cth) ss 45, 44ZZRF, 44ZZRG, 44ZZRJ, 44ZZRK. This is considered further below.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 20 /

Date: 7/9

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 99

factors make it difficult to generalize. In some cases this is simply represented by the source of the relevant law. The US has developed its categorization process through case law whereas other jurisdictions, such as Australia, have been far more reliant on legislative prescription.78 6.3 Full Assessment of Anti-competitiveness Under the dichotomous categorization process, if the relevant provision of the agreement does not fall within the per se category, it will thereby be subject to a full assessment to determine whether it should be prohibited. An earlier period of competition law saw considerable debate over what criteria should be employed in undertaking this task.79 It is now more readily accepted that the objective in prohibiting anticompetitive conduct is to promote competition, in order to reap the efficiencies that competition brings. Thus efficiency is seen as the appropriate criterion for assessing conduct. There is still room for non-efficiency objectives to play a role in the assessment of conduct and this is one of the issues considered in the discussion below on how jurisdictions approach this question in the context of analysing agreements. 6.3.1 USA It was noted above that the Professional Engineers case suggested a dichotomy between per se and ‘rule of reason’. That is, if the conduct does not fit within the per se category then it automatically falls into the rule-of-reason basket. The rule of reason is expressed in very broad terms but is essentially concerned to determine whether the agreement is anti-competitive. In the well-worn words of Brandeis J in Board of Trade of City of Chicago v US, ‘the true test of legality is whether the restraint imposed is such as merely regulates and perhaps thereby promotes competition or whether it is such as may suppress or even destroy 78

For example, the ancillary restraints doctrine referred to above has been developed through case law in the US whereas Australian courts have apparently not felt sufficiently bold to adopt it there. This has created some problems. See for example Radio 2UE Sydney Pty Ltd v Stereo FM Pty Ltd (1982) 62 FLR 437 and, on appeal (1983) 68 FLR 70. In Canada, for example, the ancillary restraints doctrine is expressed in statutory form: Competition Act 1985, s 45(4). 79 See, for example, Kenneth G Elzinga, ‘The Goals of Antitrust: Other than Competition and Efficiency, What Else Counts?’ (1977) 125 University of Pennsylvania Law Review 1191; Lawrence Sullivan, ‘Economics and More Humanistic Disciplines: What are the Sources of Wisdom for Antitrust?’ (1977) 127 University of Pennsylvania Law Review 1.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 21 /

Date: 7/9

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

100 Comparative competition law

competition’.80 Similarly, in Professional Engineers the Court stated that the rule of reason: has been used to give the Act both flexibility and definition, and its central principle of antitrust analysis has remained constant. Contrary to its name, the Rule does not open the field of antitrust inquiry to any argument in favor of a challenged restraint that may fall within the realm of reason. Instead it focuses directly on the challenged restraint’s impact on competitive conditions.81

Accepting then that the rule of reason is essentially concerned with whether the agreement is anti-competitive, this essentially becomes a quest to determine whether the agreement creates market power in the hands of the parties to the agreement. This is recognized in the US Guidelines: ‘[t]he central question is whether the relevant agreement likely harms competition by increasing the ability or incentive profitably to raise price above output, quality, service, or innovation below what likely would prevail in the absence of the relevant agreement’.82 The analysis requires an assessment of the future relevant market with the agreement in place in comparison with a market without it. The assessment of the market is to determine its competitiveness and requires the court or regulator to define the relevant market and then assess the state of competition in that market. The ‘structure’ of the market, particularly the market share of the parties to the agreement and the height of any barriers to entry to that market, is, and despite criticism of that concept, still heavily influential in this analysis.83 The US Guidelines create ‘safety zones’ for competitor collaboration.84 These are based on the collaborators’ market share. The Guidelines state that ‘[a]bsent extraordinary circumstances, the Agencies do not challenge a competitor collaboration when the market share of the collaboration and its participants collectively account for no more than twenty percent of each relevant market’.85 Per se, ‘quick look’ rule-of-reason cases and collaborations within the Merger Guidelines are excluded from this protection. In the case of ‘innovation markets’, the safety zone applies 80

246 US 231 (1918), 238. 435 US 679 (1978), 688. 82 US Guidelines (n 13) para 3.3. See also Schering-Plough Corp v FTC 402 F 3d 1056 (11th Cir 2005) and Tunis Bros Co v Ford Motor Co 952 F 2d 715 at 727 (3rd Cir 1991). 83 US Guidelines (n 13) paras 1.2 and 3.3. 84 Ibid para 4. 85 Ibid para 4.2. 81

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 22 /

Date: 7/9

JOBNAME: Duns PAGE: 23 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 101

‘where three or more independently controlled research efforts in addition to those of the collaboration possess the required specialized assets or characteristics and the incentive to engage in R&D that is a close substitute for the R&D activity of the collaboration’.86 6.3.2 EU Article 101(1) of the TFEU prohibits agreements that have as their ‘effect’ ‘the prevention, restriction, or distortion of competition within the common market and which may affect trade between Member States’. In its Guidelines the EU Commission states that ‘effect’ includes ‘likely’ effect.87 In determining whether an agreement has this anti-competitive effect, the Commission makes an assessment of whether the agreement has an appreciable adverse impact on at least one of the parameters of competition on the market, such as price, output, product quality, product variety or innovation … This means that the agreement must reduce the parties’ decision-making independence, either due to obligations contained in the agreement which regulate the market conduct of at least one of the parties or by influencing the market conduct of at least one of the parties by causing a change in its incentives.88

As noted above, an agreement is prohibited only where it has an ‘appreciable’ impact on competition. The Commission has published a Notice on what is ‘appreciable’ in this context.89 This effectively establishes a ‘safe harbour’ for certain types of agreements. In this Notice, the Commission states90 that agreements between competitors91 do not appreciably restrict competition:

86 Ibid para 4.3. ‘Innovation markets’ are defined, in para 3.32(c) of the Guidelines, as consisting of ‘the research and development directed to particular new or improved goods or processes and the close substitutes for that research and development’. 87 EU Guidelines (n 16) para 26. 88 Ibid para 27. 89 Commission Notice on agreements of minor importance which do not appreciably restrict competition under Article 81(1) of the Treaty establishing the European Treaty [now Article 101(1) of the Treaty on the Functioning of the European Union] (de minimis). 90 Ibid para 7. 91 Vertical agreements that fall within this Notice are considered in Chapter 9.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 23 /

Date: 13/10

JOBNAME: Duns PAGE: 24 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

102 Comparative competition law

(a)

if the aggregate market share held by the parties to the agreement does not exceed 10% on any of the relevant markets affected by the agreement … if the market share held by each of the parties to the agreement does not exceed 15% on any of the relevant markets affected by the agreement …

(b)

This safe harbour does not apply, however, to agreements that have the ‘object’ of fixing prices, limiting output or sales or allocating markets or customers.92 An issue that has dogged EU law in this area has been the relationship between paragraphs (1) and (3) of Art 101. Paragraph (3) (set out above in full) provides that para (a) ‘may be declared inapplicable’ if the relevant provisions in the agreement ‘contribute to improving the production or distribution of goods or to promoting technical or economic progress, while allowing customers a fair share of the resulting benefit’ (that is, provided the relevant provision is necessary and does not otherwise substantially eliminate competition). The role to be played by this paragraph depends to some extent on the role of para (1). A broad view of para (1) is that any restriction on competition will be sufficient to bring an agreement within its terms. On this view the court is not required to take into account any pro-competitive or efficiency issues. These become a matter for para (3).93 This broad view provides a prominent role for para (3). A narrow view of para (1) is that a provision of an agreement will fall within it only if the provision is anticompetitive in a ‘net’ sense, i.e. taking into account any pro-competitive or efficiency justifications. Thus a narrow view of para (1) will operate as a kind of ‘filter’ to cases that would otherwise be considered under para (3). This issue was particularly significant when the Commission was the arbiter of whether conduct fell within the previous Art 81(3) of the EC Treaty. In its Guidelines the Commission considered that efficiencies were appropriately considered by it under para (3) and not by the courts 92

The Commission notice (n 89) para 11. This was the view taken in Case T-528/93 Metropole Television SA v Commission [1996] ECR II-649, 5 CMLR 386 at para 72, where the court considered that pro-competitive benefits are to be examined under the predecessors to Art 101(3) rather than 101(1). However, compare Case C-309/99 Wouters v Algemene Rand van de Nederlandse Order van Advocaten [2002] ECR I-1577, 4 CMLR 913 (2002) and Meca-Medina and Majcen v Commission [2004] ECR II-3291. 93

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 24 /

Date: 7/9

JOBNAME: Duns PAGE: 25 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 103

under para (1).94 However the situation has recently changed. National courts (and regulators) now also have jurisdiction to consider whether Art 101(3) applies95 and as a result the issue has lost much of its sting. In addition to its exemption provision in Art 101(3), another distinctive feature of EU competition law is that it has traditionally sought goals other than efficiency, the most prominent of these being market integration of the various Member States. This sharply distinguishes it, at least potentially, from US antitrust law. And there has been the further issue of whether broader public interest arguments are relevant to the analysis under Art 10196 and even whether an appropriate objective of competition law is to protect competitors as well as competition.97 It is generally accepted, however, that at least the Commission has moved towards a more economics-based approach to assessing Art 101. This has raised the question of whether the EU can now be said to have a ‘rule of reason’, along the lines of the US. A number of EU commentators have spoken out against such a view.98 While it is understandable that the EU may be reluctant to incorporate wholesale a US development which is so clearly tied to US antitrust law and development, and while it may be strictly true that differences remain between the two bodies of law in this respect, it nevertheless remains true that the process of analysis under Art 101 is broadly similar to the US rule of reason.99 6.4 Variations: Modified Rules and a Continuum In the National Society of Professional Engineers case the US Supreme Court, in the statement quoted at 6.2.1 above, referred to a two-category dichotomy – agreements fall into either the per se or the rule-of-reason category. This dichotomy has been adopted in many jurisdictions. The EU categorization is one, as shown above. Australia has adopted a similar approach. The advantage this brings is simplicity, but this is also its danger. If a type of agreement is incorrectly categorized, there is a risk of 94

EU Guidelines (n 16) para 29. Council Regulation (EC) 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 L001/0001, Art 1, which came into effect in May 2004, provides for this. 96 See for example Monti (n 3). 97 This derives from the so-called ‘ordoliberal’ school. See for example Jones (n 4) 742. 98 See e.g. Whish (n 19) 131–133; Goyder (n 73) 111–112. 99 See Van den Bergh and Camesasca (n 38) 193–194. 95

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 25 /

Date: 13/10

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

104 Comparative competition law

overreach (if categorized as per se) or under-reach (if categorized as rule of reason). An example is the Australian per se prohibition of boycotts. This prohibition is in the broadest terms, and unlike the US approach requires neither market power in the collaborators nor that the target of the boycott be a competitor of the collaborators.100 It has been forcefully argued that in failing to incorporate these US qualifications, the Australian provisions are a ‘mistranslation’ of US boycott law and have created serious problems of overreach.101 However, more recent US case law suggests a move away from a dichotomous approach to a more nuanced approach. It was noted above that even the simple dichotomy admitted of some refinement in the case of boycotts. A broader attack came from the Supreme Court in National Collegiate Athletic Association v Board of Regents of the University of Oklahoma.102 This decision established what came to be known as the ‘quick look’ rule of reason. In its decision, the Court suggested that in some cases there is no need for a detailed analysis when applying the rule of reason. It stated that ‘the essential point is that the rule of reason can sometimes be applied in the twinkling of an eye’, an expression used by Philip Areeda.103 This approach was later consolidated in the Supreme Court’s decision in Californian Dentists Association.104 This case concerned the Code of Ethics of a non-profit association of dentists, representing 75 per cent of California’s dentists. This Code required, inter alia, that members would not engage in ‘false or misleading’ advertising. The association also issued ‘advisory opinions’ about the meaning of various provisions of the Code. The FTC argued that one of the advisory opinions in relation to the 100

The relevant provisions are referred to as ‘exclusionary provisions’, a term defined in s 4D of the Competition and Consumer Act 2010 (Cth). 101 Warren Pengilley, ‘Trade Associations and Collective Boycotts in Australia and New Zealand: A Mistranslation of the Sherman Act Down Under’ (1987) 32 The Antitrust Bulletin 1019. Note that since this article was written, New Zealand has amended its provisions to require that the target of the boycott be a competitor and has introduced a defence that the boycott does not have the purpose, effect or likely effect of substantially lessening competition: see s 29 Commerce Act 1986 (NZ). 102 468 US 85 (1984). 103 Perhaps surprisingly, this comment was contained in a footnote (footnote 39). For comment, see generally Stephen Calkins, ‘California Dental Association: Not a Quick Look but not the Full Monty’ (2000) 67 Antitrust Law Journal 495; and Terry Calvani, ‘Some Thoughts on the Rule of Reason’ [2001] European Competition Law Review 201. 104 526 US 756 (1999).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 26 /

Date: 7/9

JOBNAME: Duns PAGE: 27 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 105

meaning of ‘false or misleading advertising’ had effectively restricted advertising of prices and quality by members. The Supreme Court referred to its previous decisions in National Collegiate Athletic Association, Professional Engineers and Indiana Federation of Dentists as forming ‘the basis for what has come to be called abbreviated or “quick-look” analysis under the rule of reason’. The Court stated that a ‘quick look’ is appropriate where ‘an observer with even a rudimentary understanding of economics could conclude that the arrangements in question would have an anti-competitive effect on customers and markets’.105 And that ‘a quick-look analysis carries the day when the great likelihood of anti-competitive effects can easily be ascertained’.106 Finally, the Court concluded by doubting the utility of the traditional dichotomy: The truth is that our categories of analysis of anti-competitive effect are less fixed than terms like ‘per se’, ‘quick look’, and ‘rule of reason’ tend to make them appear. We have recognized for example, that ‘there is often no bright line separating per se from Rule of Reason analysis,’ since ‘considerable inquiry into market conditions’ may be required before the application of any so-called ‘per se’ condemnation is justified.107

The essence of the ‘quick look’ approach is that the plaintiff is not required to provide complete proof of the defendant’s market power. Having said that, it is nevertheless generally accepted that Californian Dentists Association requires more than a ‘quick look’.108 More recently the courts, or at least the lower courts, have formulated their approach in terms of a shifting burden of proof and this has given rise to the view that per se and rule of reason have now become part of a ‘continuum’ and that this has displaced the older notion of categorization. In Polygram Holding Inc v FTC,109 for example, the Court of Appeals agreed with the Federal Trade Commission’s suggestion that if a restraint is ‘inherently suspect’ then the burden shifts to the defendant to provide

105

Ibid 770. Ibid. 107 Ibid 779, quoting National Collegiate Athletic Association v Board of Regents of the University of Oklahoma 468 US 85 (1984). 108 Calkins (n 103). 109 416 F 3d 29 (DC Circuit 2005). 106

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 27 /

Date: 7/9

JOBNAME: Duns PAGE: 28 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

106 Comparative competition law

an efficiency defence. If the defendant discharges this burden, then the Commission would be required to establish anti-competitive effects.110 6.5 Characterization The more rigid the categories adopted to assess agreements the more important the process becomes of determining within which category a particular case falls. This characterization process will also include determining whether an agreement is naked or ancillary.111 As already noted, a party who has the burden of proving a breach of the competition provision has a significant incentive to argue that the agreement is prohibited per se and there is no doubt that this characterization process that the courts must undertake will generally be critical to the outcome of the case. The characterization process was explained in the context of price fixing in the US decision of Broadcast Music, Inc v Columbia Broadcasting System, Inc: [T]his is not a question simply of determining whether two or more potential competitors have literally ‘fixed’ a ‘price’. As generally used in the antitrust field, ‘price fixing’ is a shorthand way of describing certain categories of business behavior to which the per se rule has been held applicable. … Literalness is overly simplistic and often overbroad … it is necessary to characterize the challenged conduct as falling within or without that category of behavior to which we apply the label ‘per se price fixing’. That will often, but not always, be a simple matter.112

110

See also Broadcast Music Inc v Columbia Broadcasting System, Inc 441 US 1 (1978); and generally Langenfeld and Silvia (n 15); Femi Alese, Federal Antitrust and EC Competition Law Analysis (Ashgate 2008) 113. 111 Australia has not adopted an ancillary doctrine and for an illustration of the difficulties this can create, see Radio 2UE Sydney Pty Ltd v Stereo FM Pty Ltd (1982) 44 ALR 557 and, on appeal, (1983) 48 ALR 361 and generally Jacqueline Bos, ‘Antitrust Treatment of Cartels: A Comparative Survey of Competition Law Exemptions in the United States, the European Union, Australia and Japan’ (2002) 1 Washington University Global Studies Law Review 415. 112 441 US 1, 19–20 (1979) 9; See generally Denis (n 54). For a general treatment of the economics of legal rules in the antitrust context see William Landes, ‘Optimal Sanctions for Antitrust Violations’ (1983) 50 University of Chicago Law Review 652. See also Douglas Ginsburg, ‘The Appropriate Role of the Antitrust Enforcement Agencies’ (1988) 9 Cardozo Law Review 1277, 1278–1281.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 28 /

Date: 13/10

JOBNAME: Duns PAGE: 29 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

The range of conduct caught 107

How do the courts and regulators engage in this characterization? The US Guidelines suggest a combination of the purpose and effect of the restraint must be considered. The Guidelines state that: ‘[i]n examining the nature of the relevant agreement, the Agencies take into account inferences about business purposes for the agreement that can be drawn from objective facts. The Agencies also consider evidence of the subjective intent of the participants to the extent that it sheds light on competitive effects.’113 The approach in the EU is similar. In T Mobile Netherlands BV v Raad van Bestuur van de Nederlandse Mededingingsautoritleil the European Court of Justice stated that: With regard to the assessment as to whether a concerted practice is anticompetitive, close regard must be paid in particular to the objectives which it is intended to attain and to its economic and legal context … Moreover, while the intention of the parties is not an essential factor in determining whether a concerted practice is restrictive, there is nothing to prevent the Commission of the European Communities or the competent Community judicature from taking it into account.114

In contrast, if a jurisdiction has less distinct categories, more along the lines of a ‘continuum’, then conduct characterization becomes less significant. Indeed, it has been suggested by Alese that in such a case characterization is ‘a thing of the past’.115 Under this approach, the process is more a question where the burden of proof lies and in what circumstances this burden should shift between parties.

7. CONCLUSION At the beginning of this chapter it was suggested that, as a matter of competition policy, agreements pose a challenge because of their variety. Agreements range from the most anti-competitive conduct (price-fixing cartels) through to the positively pro-competitive (such as some R&D joint ventures). And the one agreement may contain provisions that are anti-competitive, pro-competitive and competitively neutral. This breadth creates challenges for competition law, which needs to balance the need for certainty in the law with an ability to have sufficient flexibility to determine whether an agreement is indeed anti-competitive. The techniques adopted to engage with this process in different jurisdictions have 113 114 115

Columns Design XML Ltd

US Guidelines (n 13) para 12. Case C-8/08, para 27. Alese (n 110) 111.

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 29 /

Date: 7/9

JOBNAME: Duns PAGE: 30 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

108 Comparative competition law

been examined in this chapter. Flexible categories, along the lines of the US approach, are attractive in their ability to more effectively analyse whether a provision is anti-competitive, but the boundaries of these categories can be difficult to maintain.116 Most jurisdictions have adopted a simpler dichotomous categorization process, involving either a per se prohibition or a full competition analysis. Whichever approach is taken, there is much to be said for regulators developing ‘safe harbours’ to guide parties facing a full competitive analysis of their agreements.

116

Columns Design XML Ltd

See generally Calkins (n 103).

/

Job: Duns-Comparative_competition_law

/

Division: 06_Chapter5

/Pg. Position: 30 /

Date: 7/9

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

6. Understanding market power Alexandra Merrett

This chapter provides an outline of the reasons for which we regulate market power and then considers how market power can be established. It contrasts ‘classical’ or structural approaches to market power with more modern strategic considerations, before examining several cases in which market power was alleged to have been misused.

1. WHY IS MARKET POWER A PROBLEM? A useful starting point for analysing market power is to approach it backwards. If we can understand what problem we are trying to fix when regulating or eliminating market power, we can better understand what market power is. This purposive approach to defining market power also helps to resolve the contest between classical approaches to defining market power, as against more strategic (or Bainian) approaches, as discussed in the second part of this chapter. Concerns with market power essentially relate to its adverse impact on ‘welfare’. In economic terms, the arguments against market power are represented by the diagrams in Figure 6.1 (the second being a more simplified version of the first). In accordance with this standard analysis, market power gives rise to a power over price that is exercised by a profit-maximising firm. The profit-maximising price is calculated by reference to the marginal revenue of the firm with market power. Unlike a firm acting in a competitive market, a firm with market power is able to restrict supply and charge a higher price. This price (which exceeds marginal cost) in turn means a reduction in total output. Consequently, a deadweight loss occurs (or DWL, shown as the dark triangle in Figure 6.1). This represents the demand that would have been satisfied if the product had been sold at marginal cost. In economists’ language, this means allocative efficiency has not been maximised. Meanwhile, some of the ‘consumer surplus’ – being the savings of consumers who would have paid more if they had had to – passes back to the producer (shown as the light triangle in Figure 6.1). This is known as a transfer. 109

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 1 /

Date: 7/9

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

110 Comparative competition law P

P D

D

Transfer from consumers to producers DWL

p1

MC

pc

Transfer from consumers to producers DWL

p1 pc

MR q1

qc

AC = MC

D

D Q

q1

Detailed illustration

qc

Q

Simplified illustration

Figure 6.1 The impact of market power Finally there are concerns which are not represented in Figure 6.1. These relate to the adverse impact of market power upon productive and dynamic efficiency. Productive inefficiency arises on account of the ‘quiet life’ experienced by a firm with market power – outside the rat race of vigorous competition, such a firm does not strive to contain costs as it would otherwise have to. Accordingly, the marginal cost curve shown in Figure 6.1 effectively moves up. Dynamic inefficiency arises for the same reason, and (as well as affecting the marginal cost curve) means that the demand curve remains more static than it otherwise would; this reflects the firm’s lack of imperative to improve its product through innovation in order to ‘stay ahead’ or to beat competition. While an economist will reel these answers off at speed, there is debate as to which (or what combination) of the above factors explain the genesis of our competition laws and to what extent they should impact upon the law’s current interpretation. In the case of the United States – which to a large degree provides the model for the laws in other jurisdictions such as Australia’s Competition and Consumer Act 2010 (Cth) (CCA)1 – it has been noted that, ‘[c]onfusion exists over the theoretical bases of antitrust law, confusion which stems directly from the

1

As alluded to in Commonwealth, Parliamentary Debates, Senate, 30 July 1974, 42–43 (Senator Lionel Murphy, Attorney General). It is noted that the CCA was formerly known as the Trade Practices Act 1974 (Cth).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 2 /

Date: 7/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 111

fact that no one can tell from the plain language of the predominant antitrust statutes what interests they are designed to protect’.2 1.1 Efficiencies As noted by Lande, the ‘prevailing view’, premised on the notion of Pareto optimality,3 is that the antitrust laws in the United States today function to promote economic efficiencies (i.e. allocative, productive and dynamic).4 Of these, the primary focus is upon allocative efficiency and the extent of any deadweight loss. As such, Bork states that ‘[t]he whole task of antitrust can be summed up as the effort to improve allocative efficiency without impairing productive efficiency so greatly as to produce either no gain or a net loss in consumer welfare’.5 Similarly, Gans et al baldly state that ‘[t]he problem in a monopolised market arises because the firm produces and sells a quantity of output below the level that maximises total surplus’, giving rise to a deadweight loss.6 In the case of Australia, the size of monopoly deadweight loss is estimated to be about 1 per cent of total production.7 1.2 Transfers Often those who consider the improvement of allocative efficiency to be the primary goal of competition law regard any concern with transfers as inappropriately elevating the welfare of consumers above that of other members of the economy (namely, producers and their shareholders). Such views note that the transfer which occurs when a consumer pays more to a producer than would be the case in a competitive market does not affect the total production in the economy. As such, it is a matter of distribution (who gets what), which competition laws are ill-equipped to handle. Nonetheless, as illustrated by Figure 6.1, ‘the redistributive effects of market power generally exceed the allocative inefficiency 2

Thomas G Krattenmaker, Robert H Lande and Steven C Salop, ‘Monopoly Power and Market Power in Antitrust Law’ (1987) 76 Georgetown Law Journal 241, 243. 3 See further Robert H Lande, ‘Wealth Transfers as the Original and Primary Concern of Antitrust: The Efficiency Interpretation Challenged’ (1999) 50 Hastings Law Journal 871, 879. 4 Ibid 873. 5 Robert H Bork, The Antitrust Paradox: A Policy at War with Itself (1978) 191. 6 Joshua S Gans et al, Principles of Economics (2nd edition, 2003) 320. 7 See research of Dixon and Creedy as cited in ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 3 /

Date: 7/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

112 Comparative competition law

effects by a substantial amount’.8 Indeed, Lande claims that ‘[u]nder market conditions most likely to be encountered’, the transfer is likely to be between two and forty times greater than the deadweight loss.9 Unsurprisingly, therefore, issues of transfers tend to attract a degree of political attention. Having analysed the Congressional debates which preceded the principal antitrust statute in the United States, the Sherman Act, Lande identifies that transfers were the primary concern of the legislature;10 indeed, he notes that the legislature at the time would have had very little, if any, understanding of allocative efficiency.11 1.3 Other Concerns There are other issues with market power which are raised from time to time. For example, economists condemn the inefficiencies which arise due to firms seeking to become monopolists – this is generally known as rent-seeking behaviour.12 From a political perspective, the interests of small business are often given as a reason against the accumulation of market power. Again, competition law is considered by economists to be an inappropriate instrument by which to protect the interests of small business. Nonetheless, this was indeed a concern expressed by Congress

8

Lande (n 3) 880. Ibid. 10 ‘The antitrust laws were passed primarily to further what may be called a distributive goal, the goal of preventing unfair acquisition of consumers’ wealth by firms with market power. It should be stressed, however, that Congress did not pass the antitrust laws to secure the ‘fair’ overall distribution of wealth in our economy or even to help the poor. Congress merely wanted to prevent one transfer of wealth that it considered inequitable, and to promote the distribution of wealth that competitive markets would bring. In other words, Congress implicitly declared that ‘consumers’ surplus’ was the rightful entitlement of consumers; consumers were given the right to purchase competitively priced goods. Firms with market power were condemned because they acquired this property right without compensation to consumers.’: ibid 875. 11 ‘It is extremely unlikely that [when the Sherman Act was passed in 1890] the legislators’ distaste for monopoly pricing could have been based upon its impact on allocative efficiency: the concept of allocative efficiency was, at best, on the verge of discovery by leading economic theorists when the Sherman Act was passed’: ibid 960. ‘More importantly, leading economists of the day had very little influence on the passage of the Act. It is unlikely, then, that the legislators who passed the early antitrust laws were aware that monopoly pricing led to allocative inefficiency’: ibid 894. 12 See for example Gans et al (n 6) 320. 9

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 4 /

Date: 7/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 113

prior to the passing of the Sherman Act,13 and it is still a notion with political traction as demonstrated by the 2004 Senate inquiry into the protection provided to small business by Australia’s competition laws.14 1.4 Determining the Priority of Concerns These various concerns will at times conflict; consequently, their resolution requires prioritisation. Krattenmaker et al argue that a broad conception of consumer welfare is the appropriate measure: ‘[u]nder this interpretation, a practice restrains trade, monopolizes, is unfair, or tends to lessen competition if it harms consumers by reducing the value or welfare they would have obtained from the market-place absent the practice.’15 As such, transfers will be important. Efficiencies also remain significant. Indeed, Krattenmaker et al note that the exercise of market power, ‘reduces allocative efficiency and transfers wealth from consumers to the owners of the firms exercising monopoly power. In addition, for Bainian market power [discussed below], production efficiency also is reduced’.16 Furthermore, in assessing whether uses of market power warrant regulatory intervention, the consumer welfare approach considers the efficiency gains that the conduct in question may generate. These benefits are to be weighed against any costs of the conduct.17 There is limited guidance as to how the regulation of market power in Australia seeks to determine its hierarchy of priorities. Somewhat in a similar manner to the history of the Sherman Act, the Second Reading Speech which preceded the passing of the CCA (then the Trade Practices Act) does not suggest that minimising deadweight loss was a priority. Indeed the focus appears to fall mainly on a highly charged issue of the early to mid-1970s, inflation.18 The only other guidance is a general statement noting that restrictive trade practices

13

See for example Lande (n 3) 907. See Senate Economics References Committee, The Effectiveness of the Trade Practices Act 1974 in Protecting Small Business (2004). 15 Krattenmaker, Lande and Salop (n 2) 244. 16 Ibid 250. 17 See for example, Steven C Salop and R Craig Romaine, ‘Preserving Monopoly: Economic Analysis, Legal Standards, and Microsoft’ (1999) 7 George Mason Law Review 617, 642. 18 See Commonwealth, Parliamentary Debates, Senate, 30 July 1974, 540–541 (Senator Lionel Murphy, Attorney General). 14

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 5 /

Date: 7/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

114 Comparative competition law cause prices to be maintained at artificially high levels. They enable particular enterprises or groups of enterprises to attain positions of economic dominance which are then susceptible to abuse; they interfere with the interplay of competitive forces which are the foundation of any market economy; they allow discriminatory action against small businesses, exploitation of consumers and feather-bedding of industries …19

As such, transfers again appear to be of concern. In large part, it remains for the courts to determine what the focus should be. In the absence of clear direction, however, there is considerable scope for lower courts and regulators of market power to choose their own priorities.

2. APPROACHES TO DEFINING MARKET POWER While economists and lawyers can agree on a simplified definition of market power – the ability to raise prices above their competitive level (the price elevation test) – there are substantial differences of opinion concerning how market power may be identified and exercised. As such, it is generally agreed that market power is a discretion, enabling a firm to control price profitably by directly restraining its own output20 (or, put more simply, by ‘giv[ing] less and charg[ing] more’).21 When examined closely, however, the agreement underlying these common descriptions is somewhat superficial. Indeed, the differences of opinion as to how market power may be identified and exercised reflect the subtleties that underpin the price elevation test – subtleties that provide significant scope for debate. Accordingly, while there is prima facie agreement about how market power is to be defined, once the price elevation test is explored in greater detail, significant disagreement emerges. Before examining the price elevation test in further detail, however, it is important to clarify terminology. The terms ‘market power’ and ‘monopoly power’ can, on occasion, create some confusion as to whether they imply a difference in meaning. As noted by Krattenmaker et al, 19

Ibid 540. J Neil Lombardo, ‘Resuscitating Monopoly Leveraging: Strategic Business Behavior and its Implications for the Proper Treatment of Unilateral Anti-competitive Conduct under Federal Antitrust Laws’ (1996) 41 St Louis University Law Journal 387, 391. 21 Maureen Brunt, ‘“Market definition” Issues in Australian and New Zealand Trade Practices Litigation’ (1990) 18 Australian Business Law Review 86, 93. 20

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 6 /

Date: 7/9

JOBNAME: Duns PAGE: 7 SESS: 6 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 115

‘courts define “market power” and “monopoly power” in ways that are both vague and inconsistent”.22 While this statement was made in the context of United States decisions, it is equally applicable to any examination of Australian jurisprudence. In like manner to Krattenmaker et al, this chapter proceeds on the basis that there is no qualitative difference between the two, both being the antithesis to competition. Accordingly, while the term ‘market power’ is preferred throughout, where ‘monopoly power’ is used (for example, in a cited passage), no difference in meaning is intended unless specifically noted. Similarly, unless otherwise stated, references to ‘monopolists’ should simply be taken to mean a firm with market (or monopoly) power. 2.1 Classical Approaches to Market Power: Power over Price Classical formulations of market power, often attributed to Stigler or more generally the Chicago School, centre on concepts of price increase23 and, its converse, output restriction.24 This approach is sometimes dubbed ‘output theory’. While market power is regarded as a continuum, with all competitors in a given market likely to have some degree of power,25 market power, according to such an approach, should only be of concern when it permits a given competitor to (profitably) raise prices above competitive levels26 (generally assessed in terms of the marginal cost of production and distribution, both direct and indirect). 22 Krattenmaker, Lande and Salop (n 2) 241. They further note (at 246): ‘Supreme Court opinions demonstrate a marked inconsistency as to whether market power and monopoly power are similar or distinct concepts. We can find no Supreme Court opinion that contrasts the terms “market power” and “monopoly power” deliberately and explicitly, i.e. that finds the existence of one but not the other … Despite these references, however, the Supreme Court, in other cases, seems to have articulated standards for “monopoly power” and “market power” that, at least linguistically, are incompatible …’ 23 Assessed by means of the ‘SSNIP’ test – or the ability of a firm to initiate (and maintain) a small but significant non-transitory increase in price. 24 See, for example, Eleanor M Fox, ‘What is Harm to Competition? Antitrust, Exclusionary Practices, and Anti-competitive Effect’ (2002) 70 Antitrust Law Journal 371, 383. 25 ‘[M]arket power is literally everywhere: almost every firm has some (usually minor) degree of power to set its price’: Thomas C Arthur, ‘The Costly Quest for Perfect Competition: Kodak and Non-structural Market Power’ (1994) 69 New York University Law Review 1, 24. See also Explanatory Memorandum, Trade Practices Revision Bill 1986 (Cth), s 40. 26 George A Hay, ‘Market Power in Australasian Antitrust: An American Perspective’ (1994) 1 Competition and Consumer Law Journal 215, 222.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 7 /

Date: 14/10

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

116 Comparative competition law

Before market power should prompt concern, however, the ability to raise prices must be both significant and durable.27 As such, it has been stated that assessing whether a firm has market power requires consideration of three important questions: (a) (b) (c)

Could a firm increase prices by restricting its output? Would increasing prices be profitable for that firm? Could the prices be maintained above competitive levels for a significant period of time?28

This in turn requires analysis of buyers’ responsiveness to price change and the availability of suitable substitutes. Both these issues must be measured over an appropriate time frame; as such, a sufficient period of time should be allowed to enable a new entrant to become established and to force greater competition, as it is only in the absence of such prospective entry that market power is likely to be of sufficient duration to be of concern. ‘If entry is easy, then market power will be eliminated or minimized depending on economies of scale.’29 The relationship between price and output is shown in the (simplified) diagram in Figure 6.2. Any movement in price (shown as p in Figure 6.2) is inextricably linked with a corresponding (and inverse) movement in output (q), the extent of which is dependent upon the slope of the demand curve (reflecting elasticities).30 While, in a competitive market, each individual firm ‘takes the price of its output as given and then chooses the quantity supplied so that price equals [long-run] marginal cost’ (shown as MCi in Figure 6.3),31 the output (and hence the price to be charged) of a profit-maximising firm with market power is determined by the intersection of the marginal revenue curve (shown as MRm) and the marginal cost curve.32 These two scenarios may be compared as shown in Figure 6.3. 27 Jeffrey Church and Roger Ware, Industrial Organization: A Strategic Approach (2000) 603. 28 Michael S McFalls, ‘The Role and Assessment of Classical Market Power in Joint Venture Analysis’ (1998) 66 Antitrust Law Journal 651, 654. 29 Church and Ware (n 27) 603. 30 As noted by Stoft, however, elasticity is not constant along a linear demand curve (Steven Stoft, Power System Economics 340). 31 Gans et al (n 6) 305. Di represents that firm’s individual demand curve. 32 Ibid, 314.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 8 /

Date: 7/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 117 P D

p1 p2

D Q q1

q2

Figure 6.2 The relationship between price and output P

P D

D

pm

MCi

MCm pc

Di D qc

MRm Q

D

qm

Q

Figure 6.3 Setting price and output: A competitive firm as compared with a firm with market power As such, under classical approaches to assessing market power, the key issue is the ability (or otherwise) to ‘control price by restraining one’s own output’.33 2.1.1 Underlying assumptions Of course, a monopolist may elect not to restrain output and thus increase price above the competitive level; famously, Sir John Hicks noted that

33

Columns Design XML Ltd

Krattenmaker, Lande and Salop (n 2) 249.

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 9 /

Date: 7/9

JOBNAME: Duns PAGE: 10 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

118 Comparative competition law

‘the best of all monopoly profits is a quiet life’.34 As such, ‘the essence of market power is “discretionary power”’.35 Nonetheless, a key premise of economic theory is that of the profit-maximising firm: accordingly, in normal circumstances, one would expect a firm with the ability to control price by restraining its own output to do just that and to price above the competitive level. Another assumption which is important to articulate concerns the emphasis on marginal cost, being the cost of the last unit supplied or, alternatively, the savings that would be involved if that unit had not been produced. As noted by Carlton and Perloff, the ‘ability to set price above marginal cost implicitly uses the model of perfect competition as a benchmark against which to measure the behavior of firms’.36 For some industries, however, departures from perfect competition may mean marginal cost is a less than satisfactory guide. This is particularly true for industries characterised by high fixed (as against variable) costs or significant sunk costs. 2.1.2 Exercising classical market power In light of the classical approach to defining market power, it is unsurprising that the most obvious means by which such power is likely to be exercised is the charging of supra-competitive or ‘monopoly’ prices. Such behaviour, however, is not generally the focus of anti-competitive prohibitions (although in the case of merger regulation, the objective is to prevent firms from attaining a position whereby they are able to engage in supra-competitive pricing). More often concerns with pricing behaviour focus upon discriminatory pricing, whereby a firm with market power is able to discriminate between its customers, charging a variety of prices with the purpose of extracting to the greatest degree possible the ‘consumer surplus’. Another concern is predatory pricing, whereby a firm prices below avoidable cost with the objective of deterring competitive behaviour from actual or potential competitors. Classical theorists, however, tend to reject claims that firms with market power engage in predatory behaviour, dismissing such behaviour as ‘foolish and self-defeating’.37 Predation is said to cause more harm to the party engaging in the behaviour (particularly the greater their market share). In the absence of a clear 34

John Hicks, ‘Annual Survey of Economic Theory: The Theory of Monopoly’ (1935) 3 Econometrica 1, 8. 35 Brunt (n 21) 93. 36 Dennis W Carlton and Jeffrey M Perloff, Modern Industrial Organization (3rd edition, 1999) 610. 37 Bork (n 5) 309.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 10 /

Date: 13/10

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 119

ability to recoup any lost profits, it is asserted that a profit-maximising firm with market power would not engage in such conduct.38 Similarly, classical theorists generally deny the likelihood of anticompetitive leveraging by a party with market power. Leveraging occurs when a party with power in one market uses that power in another (related) market to distort market outcomes (see, for example, the Baxter case, discussed in the last section of this chapter). As argued by the Chicago School, however, a firm with market power is able to earn only a single monopoly rent; a profit-maximising firm will extract that rent in its entirety from the product in relation to which it has market power.39 There is no need (and rarely an incentive) to engage in conduct such as tying to increase the profits it can already obtain from the monopolised product. On the basis of this ‘fixed sum’ or ‘single monopoly profit’ argument, leveraging – as viewed by proponents of classical market power – does not tend to raise competition concerns. Finally, ‘exclusionary conduct’ – whereby a firm with market power engages in conduct which does not directly impact upon the price/output relationship, but instead targets the firm’s competitors in order to effect long-term changes to the market – is likely to occur only rarely in a classical market power scenario. As with the leveraging argument, where a firm with market power has direct power over price and output, classical theorists consider that it is unlikely to resort to indirect means to maximise its profits (except, perhaps, in order to escape regulatory scrutiny). In any case, exclusionary conduct, according to this approach, cannot occur without the willing co-operation of the firm’s customers (and consumer sovereignty should be the ultimate guide to whether conduct is problematic).40 As such, Bork claims: [T]here has never been a case in which exclusive dealing or requirements contracts were shown to injure competition. A seller who wants exclusivity must give the buyer something for it. If he gives a lower price, the reason 38

See, for example, the discussion in Thomas J Campbell, ‘Predation and Competition in Antitrust: The Case of Nonfungible Goods’ (1987) 87 Columbia Law Review 1625, 1655. 39 See, for example, the discussion in Warren S Grimes, ‘Antitrust and the Systemic Bias against Small Business: Kodak, Strategic Conduct, and Leverage Theory’ (2001) 52 Case Western Reserve Law Review 231, 253 and Lombardo (n 20) 418. 40 See Bork (n 5) 304–305, where he argues that, when faced with a firm (‘Alpha’) offering exclusive contracts to customers, either that firm’s rivals will meet competition, or ‘the market is destined for monopoly anyway, because of Alpha’s superior efficiency’.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 11 /

Date: 7/9

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

120 Comparative competition law must be that the seller expects the arrangement to create efficiencies that justify the lower price …41

As we will see, when strategic market power is discussed, not all agree with Bork’s assertion. 2.1.3 Identifying classical market power While it is generally agreed that a price rise over a relevant period of time evidencing market power should be in the range of 5–10 per cent,42 there are various means by which such an increase may be predicted. Generally speaking, under the classical approach, findings of market power rely upon scrutinising market structure. Brunt, for example, states that assessing market power requires an examination of the number of firms constraining the conduct in question, patterns of substitution within the industry and the extent of prevailing barriers to entry.43 Hay proposes a similar methodology, stating that determining the degree to which prices can be raised above the competitive level ultimately depends upon three factors: The first is the amount by which a ‘hypothetical monopolist’ could profitably raise prices above the competitive level. This will depend on the elasticity of demand for the product. The second is the firm’s current market share (i.e. how close is the firm to being a ‘true’ monopolist). The third is the elasticity of supply of the remaining producers of the products (i.e. the extent to which other producers will expand output and take away business from the firm that attempts to raise prices).44

Hay points out, however, that the data necessary to measure elasticity of demand and supply are rarely available, hence there is a need for a ‘somewhat more pragmatic approach … [which] relies on market share as the primary determinant of market power’.45 On this view, therefore, classical market power is likely to arise in a situation where there is a protected market position – perhaps due to key intellectual property rights or special government status – giving rise to significant market share entrenched by barriers to entry. As such, ‘[t]he 41

Ibid 309. Hay (n 26) 218. 43 Brunt (n 21) 93. 44 Hay (n 26) 222. See also Lauren Johnston Stiroh and Richard T Rapp, ‘Market Power in Technology Markets’ (1999) SD72 ALI-ABA Course of Study: Antitrust/Intellectual Property Claims in High Technology Markets 61, 67. 45 Ibid. 42

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 12 /

Date: 7/9

JOBNAME: Duns PAGE: 13 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 121

ability to exercise market power depends upon the possession in some degree of what any monopolist46 has in full measure: a desirable good or service for sale, fewness of competing sellers and a means for forestalling entry in response to high prices and profits’.47 These are clearly structural factors, which are able to be examined absent any alleged use of market power. 2.1.4 Measuring classical market power With its emphasis on marginal cost, classical market power appears to lend itself readily to precise measurement. It is notable that while market power is said to be a discretion, attempts to measure it assume it has been exercised at least in so far as pricing is concerned (reflecting the underlying assumption of profit maximisation). Indeed, the Lerner Index purports to measure directly a firm’s departure from marginal cost ((P – MC)/P), and hence its degree of market power. Nonetheless, the marginal cost in question should not be that of the individual producer, but rather that of the least efficient producer in the market (the marginal supplier).48 Marginal cost is notoriously difficult to measure, so a number of proxies have been developed in its place. The closest such proxy is average variable cost – the sum of all variable costs, divided by the units of output. Nonetheless, the distinction between variable and fixed costs itself involves a degree of discretion and judgement, and can make this approach difficult to apply. Further, in cases of high fixed or sunk costs, pricing at marginal cost or average variable cost may not in fact provide a profitable return to a producer. Other approaches, involving assessments of profitability or valuation of assets, are similarly fraught, being vulnerable to economic and accounting differences.49 Traditionally, therefore, classical approaches to market power have defaulted to a structureconduct-performance model, with emphasis upon concentration and 46

‘Monopolist’ in this sense should be taken to mean a single producer. Stiroh and Rapp (n 44) 64. 48 As such, allowance for Ricardian rents are made, reflecting superior factors of production: see further Church and Ware (n 27) 120. 49 Indeed, Baker and Bresnahan state that this methodology is ‘far from ideal: For example, high profits or margins might reflect efficiencies, such as low costs or superior product design, rather than market power. In addition, the way accountants spread costs over time and adjust asset values for depreciation frequently causes accounting measures of profit to bear little relation to those underlying economic concepts that might in principle be related to market power. These problems loom so large that antitrust today does not rely heavily on 47

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 13 /

Date: 10/9

JOBNAME: Duns PAGE: 14 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

122 Comparative competition law

market share, as market share is said to be ‘indirectly related to the firm’s ability to set price above marginal cost’.50 As such, in the American context, it has been famously stated that a 90 per cent market share would be enough to constitute monopoly, 60–64 per cent would be doubtful and 33 per cent insufficient51 – a breakdown which Jebsen and Stevens describe as still ‘roughly true’.52 American jurists are generally suspicious of claims of market power where the market share in question is less than 50 per cent53 and, while the United States Supreme Court has not stipulated a minimum market share, Hay infers from lower court decisions that (in the absence of special circumstances) less than 50 per cent is too little.54 Even where special circumstances prevail, Hausman is not prepared to concede that a firm with less than 30 per cent market share can have market power.55 Nonetheless, it has been argued that, in smaller economies, a given market share should be more significant than would otherwise be the case in a larger economy.56 profitability measures in making inferences about market power …’: Jonathan B Baker and Timothy F Bresnahan, ‘Empirical Methods of Identifying and Measuring Market Power’ (1992) 61 Antitrust Law Journal 3, 5. 50 McFalls (n 28) 661. 51 Learned Hand J in United States v Aluminum Co of America 148 F2d 416 (2d Cir, 1945). See also Brian A Facey and Dany H Assaf, ‘Monopolization and Abuse of Dominance in Canada, the United States, and the European Union: A Survey’ (2002) 70 Antitrust Law Journal 513 for a discussion of the market share thresholds as accepted in those jurisdictions. 52 Per Jebsen and Robert Stevens, ‘Assumptions, Goals and Dominant Undertakings: The Regulation of Competition under Article 86 of the European Union’ (1996) 64 Antitrust Law Journal 443, 447: although they note that 60–64% is ‘perhaps more dangerous now’. 53 Lombardo (n 20) 446: ‘When addressing market power, most courts presume that there is only one type, the Stiglerian variety [and] Stiglerian market power can only be exercised when a firm attains at least a fifty per cent market share’. Cf Thomas G Krattenmaker and Steven C Salop, ‘Anti-competitive Exclusion: Raising Rivals’ Cost to Achieve Power over Price’ (1986) 96 Yale Law Journal 209. 54 Hay (n 26) 226. 55 Professor Jerry Hausman: see Australian Competition & Consumer Commission v Universal Music (2001) 115 FCR 442, [377]. 56 ‘In small economies the typical market share threshold that will signify market dominance should be lower than in larger ones because elasticity of supply will usually be lower, given the prevalence of scale economies and oligopolistic interdependence’: Michal S Gal, ‘Size Does Matter: The Effects of Market Size on Optimal Competition Policy’ (2001) 74 Southern California Law

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 14 /

Date: 13/10

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 123

Analysing market power on such a basis assumes both a market in equilibrium57 and the correct market definition. If the market is incorrectly defined, then assessing its concentration, perhaps by way of the Herfindahl-Hirschman index (HHI)58 or the more basic application of a concentration ratio such as the CR4 or CR8, is meaningless.59 Even aside from this issue, conclusions based upon market share (whether in combination with concentration ratios or otherwise), are said to be an unreliable guide. Baker and Bresnahan cite three reasons to doubt inferences drawn on the basis of market share alone: First, if entry into a market is easy, no firm can exercise market power, no matter how large its market share. Second, a firm could have a large market share and the market could appear concentrated, not because the firm has market power but because it has low costs or sells superior products. Finally, market definition treats each substitute product as either inside the market or outside the market. This approach does not recognize the competitive discipline exerted by those products just outside the market on the products within, and it does not recognize differences in degree to which firms selling within the market constrain each other.60

Further, as noted by Stiroh and Rapp, ‘[t]he higher the elasticity of supply of other firms, the lower the index of market power since any

Review 1437, 1463. See also Campbell (n 38); Jonathan B Baker, ‘Unilateral Competitive Effects Theories in Merger Analysis’ (Spring 1997) 11 Antitrust 21: as discussed at n 68 below. 57 This assumption may be particularly fraught in the case of technology markets. See for example Adolph C Iannaccone and Linda Volonino, ‘Marginal Cost and Relevant Market Determination in Information Technology Antitrust Cases’ (1992) 18 Rutgers Computer and Technology Law Journal 681, 697: ‘Markets are only coincidentally in equilibrium. Most often they are in flux due to innovation’. See also ibid 695. 58 H = Σ si2 – being the sum of each individual firm’s market share, squared. 59 The concentration ratio aggregates the market share of the four (or eight) largest competitors in a market. Note that, in contrast to the HHI, concentration ratios do not adjust for variation in firm size. For instance, the CR4 for an industry when there are five firms of equal size is 0.80. A market with six sellers where the distribution of market shares (in percentages) is 50, 10, 10, 10, 10, 10 would also have a CR4 of 0.80. However, the HHI for the first industry is 0.20 (where the shares have not been squared), while in the second industry it is 0.30, reflecting the larger variation in relative firm size even though the number of firms is greater: Church and Ware (n 27) 429. 60 Baker and Bresnahan (n 49) 4.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 15 /

Date: 7/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

124 Comparative competition law

firm’s attempt to raise price will encourage a substantial increase in the production of all other firms, tending to defeat the price increase’.61 Once a proxy is used to measure departures from marginal cost (even assuming marginal cost is a valid measure), it is necessary to employ a broader analysis than one that simply relies on concentration and market share. This effectively forms a cross-check on any conclusions based upon market share, recognising that such conclusions may not be robust. While econometric techniques have developed recently, allowing for the calculation of a precise answer to the market power question on the basis of elasticity,62 it is generally the case that the qualitative factors demonstrated by the relevant market are scrutinised. Such features may include the extent of unionisation in the industry or the countervailing power of buyers. Fundamentally, however, barriers to entry become the critical point of inquiry. This is because, if barriers are low and entry is possible and likely, this will restrain the market power of those already operating in the market. The Chicago School approach to barriers to entry is quite confined. As an example, in Bork’s ground-breaking text, The Antitrust Paradox: A Policy at War with Itself, the chapter dedicated to discussing barriers to entry is largely concerned with identifying what does not constitute a barrier.63 He concludes, however, that the focus of enquiry must be upon artificial entry barriers: ‘These must be barriers that are not forms of superior efficiency and which yet prevent the forces of the market – entry or the growth of smaller firms already within the industry – from operating to erode market positions not based on efficiency.’64 Such barriers must be distinguished from natural barriers, which are not the appropriate focus of regulatory concern.65

61

Stiroh and Rapp (n 44) 67. Other techniques have been developed to analyse responses to variations in cost and methods based upon detecting multiple pricing schemes (discriminatory pricing often being considered a sign of market power). See further Baker and Bresnahan (n 49). 63 Bork (n 5) Chapter 16. 64 Ibid 311. 65 ‘Where the product and its service are complex and expensive, it is natural that the entrant will have to do many complex and expensive things, and do them well, in order to succeed. There are natural barriers or costs of entry. To identify them is merely to make a descriptive statement, one that does not imply the propriety of invoking law to alter the size or behavior of firms already in the market’: ibid 329. 62

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 16 /

Date: 7/9

JOBNAME: Duns PAGE: 17 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 125

2.2 Strategic Market Power: Raising Rivals’ Costs The market power which tends to be placed in opposition to classical market power is strategic market power (also frequently known as Bainian or exclusionary market power). It is important to understand, however, that its proponents put it forward to complement, not to supplant, its more traditional counterpart. In contrast to classical market power – which focuses on unilateral power over price – strategic market power closely examines the impact of the conduct in question on the impugned firm’s competitors: is the conduct designed to (and likely to) raise rivals’ costs and/or to reduce the demand for their product? It is important to note that such an examination, by its nature, tends to be retrospective. 2.2.1 Contextualising market power: A focus upon incentives and effects Leading proponents of strategic market power, including Krattenmaker and Salop, strongly rebut the notion that the process of establishing the existence or otherwise of market power can be done without considering the conduct in question: ‘Bainian power cannot be evaluated in a vacuum, independent of and prior to analysis of the allegedly exclusionary conduct. It is the exclusionary conduct that creates the market power being evaluated, not the other way around.’66 As such, the examination does not commence with a review of market structure. Rather, the incentives and likely strategies of various participants in the market are closely considered, in order to understand whether conduct has been engaged in for the purpose of ‘harm[ing] competition by excluding or foreclosing competitors from a given market’.67 Strategic market power can therefore be found to exist in circumstances which would not give rise to market power in a more classical sense.68 Nonetheless, a number of authors have sought to demonstrate 66

Krattenmaker, Lande and Salop (n 2) 255. Lombardo (n 20) 391. 68 Speaking in the context of mergers, for example, Baker posits an auction model, whereby there are 10 producers in a market with equal market shares (but variant production costs). Working through a scenario from the customer’s perspective, he demonstrates that a merger between two producers in such a market may create a discretion over pricing, even though under standard merger guidelines, such a merger would not raise antitrust concerns. He concludes, ‘market shares may not capture all the factors relevant to assessing unilateral incentives to raise price following merger …’: Baker (n 56) 25. 67

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 17 /

Date: 13/10

JOBNAME: Duns PAGE: 18 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

126 Comparative competition law

that strategic market power leads to the same sort of harm to consumer welfare as classical market power, and Lombardo goes so far as to state that ‘the exercise of Bainian market power may lead to greater welfare losses for society than the exercise of Stiglerian market power’.69 Further, the long-term effects of exercising strategic market power can give rise to structural changes to the market, thereby creating classical market power.70 Given these factors, Salop argues forcefully that market power cannot be treated as a threshold issue to be considered and resolved prior to a court (or enforcer) examining the conduct of concern. He states, ‘[i]t is impossible to evaluate market power accurately without understanding the conduct and effect claims at issue and analysing market power in the context of those claims’.71 Indeed, in examining the manner in which strategic market power may be exercised, it is clear that the ability and the act become conflated, effectively requiring an iterative analysis. 2.2.2 Exercising strategic market power Examining strategic market power from the perspective of exclusion, Krattenmaker and Salop identify a number of distinct methods whereby such market power may be exercised. These methods are designed to increase the production costs incurred by the offending party’s rivals (or, conversely, to diminish demand for their output).72 Briefly described, these methods fall into two categories, the first being methods of direct foreclosure, whereby rivals’ costs are increased following some form of interference by the offending party which affects the rivals’ access to inputs. The second category is more oblique, where the conduct induces suppliers of the input (to which access has been restricted as a consequence of the exclusionary conduct) ‘to restrict output in response to incentives created by the exclusionary rights agreement’.73 As such, this category involves conduct which facilitates tacit or express collusion, consequently restricting output in the input market, and thereby raising the costs to competitors in the primary market. The direct foreclosure scenarios can be summarised as follows: 69

Lombardo (n 20) 447. ‘[A] firm may obtain a monopoly through an exercise of Bainian market power – that is, Bainian market power may lead to Stiglerian’: ibid. 71 Steven C Salop, ‘The First Principles Approach to Antitrust, Kodak and Antitrust at the Millennium’ (2000) 68 Antitrust Law Journal 187, 189. 72 See generally Krattenmaker and Salop (n 53) 230–240. 73 Ibid 230. 70

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 18 /

Date: 7/9

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 127

(a)

‘Bottleneck’: whereby a party obtains exclusive rights from all (or a significant number of) the lowest-cost suppliers of a key input for the product in question, ‘where those suppliers determine the input’s market price’.74 Competitors of the party purchasing these exclusive rights are forced to turn to high-cost suppliers or to employ less efficient substitutes in place of the preferred input. It is well illustrated by the Terminal Railroad case,75 in which a group of railroad operators obtained rights over the only railroad bridge crossing the Mississippi River at a key point. Rival operators were only permitted access to the bridge on discriminatory terms – as such, their costs were increased (whether they elected to pay the higher access fee or to incur the cost of arranging other means of crossing the river).76 ‘Real foreclosure’: whereby the party purchases77 ‘an exclusionary right over a representative portion of the supply [of an input], withholding that portion from rivals and thereby driving up the market price for the remainder of the input still available to rivals’.78 This is a typical squeeze scenario, where what is significant is the quantity of the input foreclosed (as distinct from the ‘bottleneck’ scenario which arises due to the uniqueness of the input foreclosed). An oft-cited example of such foreclosure was ALCOA’s extraction of promises from electrical utilities not to supply power to its competitors79 – as such, ALCOA’s rivals had to purchase electricity from a reduced range of suppliers, and were thereby forced to pay higher prices. Such an example is dubbed by Krattenmaker and Salop as a ‘naked’ exclusionary right (where the impugned firm purchases nothing other than the exclusion of their competitors – i.e. ALCOA did not itself obtain rights over the electricity it sought to foreclose from its rivals). Such foreclosure,

(b)

74

Ibid 234. United States v Terminal Rail Road Association 224 US 383 (1912). 76 Note the similarity with Dowling v Dalgety Australia Limited & Ors (1992) 34 FCR 109 (bearing in mind that Lockhart J concluded the owners of the facility in question did not have market power). 77 Such a ‘purchase’ may be direct – i.e. involving the payment of money – or may be in exchange for other benefits to the supplier, such as the promise of exclusivity and/or a long-term contractual agreement to purchase from that supplier. 78 Krattenmaker and Salop (n 53) 236. 79 Settled by consent decree in 1912, but discussed in further detail in United States v Aluminum Co of America 148 F2d 416 (2d Cir, 1945). See Krattenmaker and Salop (n 53) 236. 75

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 19 /

Date: 7/9

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

128 Comparative competition law

however, may be exercised more subtly, for example by advance purchasing of key inputs and/or the continual expansion of capacity prior to any market increase.80 In both cases, the conduct raises the costs of the offending party’s rivals (thereby restricting their output) without affecting its own. In such circumstances, the party is able to lift its own price (to reflect the new ‘competitive’ price which prevails in the market) and may be able to increase its market share. It is this creeping accumulation of market share and gradual structural change to the market that may, in time, give rise to market power of the classical variety.81 It is also significant that this form of foreclosure inhibits the rivals’ ability to produce the competing product in the first place: it concerns the foreclosure of inputs, not customers. As such, the standard Chicago School response to foreclosure – that customers will only agree to exclusive arrangements if it is in their interests to do so82 – does not hold. In cases of input foreclosure, the customer has no say as to which producer will have its costs increased or (in a relative sense) reduced; accordingly, efficient producers may not be rewarded by the ebb and flow of market forces. In respect of the examples giving rise to collusion (Krattenmaker and Salop’s second category of exclusionary conduct), in each case the collusion occurs amongst the suppliers of the input who – by virtue of the effective shrinking of the market through the removal of alternate 80

As indeed alleged in subsequent proceedings against ALCOA: United States v Aluminum Co of America 148 F2d 416 (2d Cir, 1945). 81 See further Krattenmaker, Lande and Salop (n 2) 267–268. Cf Arthur (n 25) 41–42: ‘Unlike the classic monopolist, a seller with nonstructural market power has power only over its own product; it lacks the power to control overall market output and price … A seller with nonstructural market power from either of these market imperfections [asset specificity or informational imperfections] can exploit a buyer only to the extent the buyer is indifferent to the higher price. The power is thus self-correcting, as is structural market power, but with the key difference that there are no structural barriers which make the correction available only in the long run. A sophisticated purchaser thus can be exploited only to a trivial degree, and even unsophisticated customers may then be protected by sophisticated ones, by education programs, and by sellers’ concern for their reputations. As a result, the degree of market power will tend to be both slight (although it may vary by customer) and unlikely to endure, and the group of exploitable customers is too small to be significant for antitrust purposes.’ 82 As noted by Salop and Romaine ‘[o]f course, where the competitive instrument of customer foreclosure is solely offering low prices to consumers, antitrust law is rightfully skeptical’: Salop and Romaine (n 17) 628.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 20 /

Date: 7/9

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 129

suppliers – obtain (or enhance any existing) power over the price of the input. As such, the exclusionary conduct ‘increase[s] the probability that the remaining unrestrained suppliers can successfully collude, expressly or tacitly, to raise price’.83 The party which has engaged in the exclusionary conduct can be – but need not be – aware of such collusion. Some key conclusions can be drawn from this brief analysis of potential exclusionary exercises of strategic market power: (a)

The means by which the power is exercised are quite subtle. As noted by Lombardo, ‘the rival is not “muscled out”’.84 Rather its profitability is gradually undercut such that, in time, it may consider its resources could be better used elsewhere. Even where competitors remain in the market, the offending party may still reap the rewards of high prices and expanding market share. Such conduct does not require structural market power. While all exercises of strategic market power require a degree of patience as well as a sound understanding of the incentives and likely responses of both suppliers and (actual and potential) competitors, such exercises may be essentially opportunistic. As such, they can seek to take advantage of some market imperfection – including one that may be short lived – in order to reap longer-term benefits and, ideally, persistent power. Such imperfections may include: cyclical conditions which effectively create short-term barriers to entry; changing entry conditions (including changes to the regulatory environment); and information asymmetries. In some cases, the offending party will need deep pockets (for example, where it is trying to buy the exclusion of its competitors from chains of supply), but this should not be mistaken for a need for structural power. The analysis requires a far more dynamic consideration of the market. By its nature, strategic conduct is patient, focusing upon long-run effects. As such, ‘the motivation behind these activities is not to reap enhanced profits in the near term, but rather to bring about a reduction in competition over time so that the firm may receive greater profits in the future’.85 Thus, as Lombardo observes, ‘[p]art of the attraction for raising rivals’ costs is that it takes

(b)

(c)

83 84 85

Columns Design XML Ltd

Krattenmaker and Salop (n 53) 240. Lombardo (n 20) 429. Ibid 422.

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 21 /

Date: 7/9

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

130 Comparative competition law

advantage of the dynamic nature of markets to disguise its anticompetitive effects’.86 These factors have the combined effect of reducing the risk of detection. The subtlety with which the power is exercised and the length of time over which its benefits are reaped may mean that it escapes regulatory attention. Furthermore, when strategic market power is exercised opportunistically, it is (by definition) less open to prediction. As such, the affected market may not be subject to the same degree of ongoing scrutiny as would occur in a market with the structural features commonly associated with classical market power. As noted by Lombardo, these factors can be very enticing: ‘Business strategies aimed at placing a firm’s rivals at a competitive disadvantage through impairment of demand for a rival’s good or through an increase in a rival’s costs provide the firm employing them with immediate benefits and the camouflage of a healthy marketplace.’87 Given the above analysis, certain types of conduct are likely to be considered as more problematic when examined from the perspective of strategic market power: (a) (b)

Exclusionary conduct is much more likely to be of concern. Similarly, leveraging is more likely to trigger alarm bells, as an analysis of the offending party’s likely incentives will be undertaken. Such analysis may well confirm that leveraging is not undertaken for the purpose of increasing an existing monopoly rent, but may reveal an intention to preserve other market power which is perceived to be under threat.88 Alternatively, the conduct may be designed to transfer existing power to the secondary market following a realisation that the offending party’s position in the primary market is likely to be eroded (perhaps due to a changing regulatory structure). Such leveraging may also lead to long-term customer 86

Ibid 416. Ibid 452. 88 ‘Sometimes the initial monopoly would face a challenge and possible dissipation by new entry or expansion by fringe competitors. In this situation, the monopolist may attempt an exclusionary strategy in order to deter or destroy that emerging competition. Stated differently, the monopoly power might be reduced or disappear, but for the exclusionary conduct. Under these circumstances, the single monopoly profit theory, and its strong policy implications about the efficiency of integration, clearly would not apply’: Salop and Romaine (n 17) 625. 87

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 22 /

Date: 7/9

JOBNAME: Duns PAGE: 23 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 131

foreclosure which – particularly in markets with certain characteristics – may be difficult to break down.89 Finally, conduct which is often dismissed by Chicagoans as irrational (in the sense that it does not appear to be profit maximising) may become explicable. For example, a firm which ostensibly does not have the market share to support predatory activity may – upon closer examination of the prevailing market – be found to have its interests closely aligned with another significant competitor. In such circumstances, not only is predation possible, it will be less costly than if the firm had a large enough market share to support a finding of classical market power.90

(c)

2.2.3 Identifying strategic market power Proponents of strategic market power, however, are alive to criticisms that anti-competitive strategic conduct is easy to confuse with healthy competition. Accordingly, key criteria have been developed in order to assess whether the conduct in question should be condemned. Krattenmaker and Salop posit that certain structural conditions must prevail before exclusionary conduct can be considered to constitute an exercise of market power, namely: First, conditions in the input market must enable the purchaser to raise its competitors’ costs by purchasing exclusionary rights. These exclusionary rights contracts must significantly raise the competitors’ costs. Second, conditions in the output market must enable the purchaser, after its competitors’ costs increase, to increase its price. It will acquire this power only if unexcluded rivals lack the ability or incentive to expand their output in response to the purchaser’s price increase and if potential entrants cannot take up the slack.91

Further, any costs associated with the conduct (from the offending party’s perspective) must be outweighed by the benefit and rivals must lack effective strategies to counteract the conduct.92 Finally, Krattenmaker and 89

For example, ‘in network markets or markets with large minimum viable scale, the monopoly will be difficult to dislodge once it is achieved’: ibid 637. 90 See further the discussion in ibid 641 concerning reasons why predation may be an attractive option. 91 Krattenmaker and Salop (n 53) 251. 92 It will not be sufficient, however, if these counter-strategies are costly as, in such cases, rivals’ costs are still raised, albeit by a different method: see further ibid 269.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 23 /

Date: 7/9

JOBNAME: Duns PAGE: 24 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

132 Comparative competition law

Salop argue that exclusionary conduct should be excused where consumers experience net efficiency benefits.93 Salop has subsequently reworked this approach into a four-step analysis: (1) The first step is the identification of the exclusionary right. Did the firm obtain the right to exclude its rivals? This raises a question about how the exclusionary rights were adopted … (2) Having identified the exclusionary rights, the next step is to inquire into whether competitors were injured. Did the exclusion have the effect of raising rivals’ costs materially? (3) If so, then the court would inquire into injury to competition and market power. Does the excluding firm or firms gain power over price in the market in which they sell? These three steps prove the existence of market power. (4) At this point, the defendant may try to defend its conduct by claiming there are overriding efficiencies that justify the conduct, despite the exclusion … [N]ot all courts will permit this defense.94

Obtaining power over price will generally require certain market imperfections or barriers to entry or expansion. While market imperfections are likely to be largely in a form recognisable by proponents of the classical system (oligopolistic tendencies, information asymmetries or the existence of network effects),95 the approach to barriers to entry or expansion will generally be significantly broader. A search for strategic market power is likely to consider the four elements of market structure which Bain identified as affecting the ability of established firms to ward off entry in the face of supra-normal profits. These elements, as summarised by Tirole, are as follows: Economies of scale (e.g. fixed costs) … [I]f the minimum efficient scale is a significant proportion of the industry demand, the market can sustain only a small number of firms that make supranormal profits without inviting entry … 93 Although arguably this should reflect the applicable legal framework, as it is closely related to the issue of the purpose of the legislation. 94 Steven C Salop, ‘New Economic Theories of Anti-competitive Exclusion’ (1987) 56 Antitrust Law Journal 57, 59–60. 95 See also Mark L Glassman, ‘Can HMOs Wield Market Power? Assessing Antitrust Liability in the Imperfect Market for Health Care Financing’ (1996) 46 American University Law Review 91, 114–115. Note, however, a strategic market power approach will recognise customer lock-in as a relevant imperfection, whereas a classical approach is less likely to do so.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 24 /

Date: 7/9

JOBNAME: Duns PAGE: 25 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 133 Absolute cost advantages The established firms may own superior production techniques, learned through experience (learning by doing) or through research and development (patented or secret innovations). They may have accumulated capital that reduces their cost of production. They may also have foreclosed the entrants’ access to crucial inputs through contracts with suppliers … Product-differentiation advantages Incumbents may have patented product innovations (which, of course, can be seen as a cost advantage relative to the product), or they may have cornered the right niches in the product space, or they may enjoy consumer loyalty … Capital requirements According to this controversial element of entry barriers, entrants may have trouble finding financing for their investments because of the risk to the creditors …96

In stark contrast to classical analyses, a search for strategic market power will make some allowance for imperfections (and barriers to entry) that are not enduring or which are short lived, but recurrent. Stiroh and Rapp, for example, note that deregulation imposes a ‘before’ and ‘after’ element on a dynamic market structure, noting that after deregulation, ‘the dominant firm will still have a substantial market share although it will no longer be protected’.97 As such, the dominant firm may be able to take advantage of its market position in the period immediately after deregulation such that it shores it up in the long run – thereby effectively creating market power in the classical form. A classical analysis, however, is less likely to scrutinise the dominant firm’s conduct immediately following deregulation as – in a search for persistent market power – its dominance may be considered temporary. It is at the point of assessing the power over price that proponents of strategic market power have made a significant contribution (even for those parties who do not seek to move beyond a classical approach): a thoroughgoing articulation of the competitive price. While there may have been a tendency in market power analysis to adopt the prevailing price as the competitive price,98 Salop identifies a number of fallacies associated with such an approach (including, but not limited to, the

96

Jean Tirole, The Theory of Industrial Organization (1988) 306. See also Carlton and Perloff (n 36) 79. 97 Stiroh and Rapp (n 44) 68. 98 Where the inquiry concerns a merger, however, the prevailing price will be the relevant benchmark. See further Carlton and Perloff (n 36) 611. See also n 99 below.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 25 /

Date: 7/9

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

134 Comparative competition law

infamous cellophane fallacy).99 In brief, the price which should be considered the correct benchmark is ‘the price that would prevail in the absence of the alleged anti-competitive restraints or conduct’.100 As such, even a structural approach to market power cannot be divorced from the conduct at hand: ‘In order to evaluate market power correctly in an antitrust case, it is necessary to identify the proper competitive benchmark. If the competitive benchmark is defective, then the market power evaluation may be irrelevant or erroneous.’101 A search for strategic market power is not restricted to the single dimension of benchmarking prices. Identifying strategic market power also requires articulation of the relevant firms’ incentives. As such, Brunt notes that evidence of independent rivalry will be evidence against a See Salop’s ‘The first principles approach …’: Salop (n 71) 194–195. Salop identifies the: + marginal cost trap: ‘Mistaking a firm’s inability to profitably raise price above its marginal cost for an inability to exercise market power by excluding rivals; and vice versa, that is, mistaking a firm’s ability to profitably raise price above its marginal cost for an ability to exercise additional market power by adopting alleged anti-competitive restraints’; + the cellophane trap: ‘Mistaking a firm’s inability to exercise market power by raising price above the current price for an inability to have already exercised market power by raising price up to the current level …’; + the price-up trap: ‘Mistaking a firm’s inability to profitably raise price above the current level for an inability to exercise market power by preventing competitors’ conduct that otherwise would reduce price below the current level …’; + the threshold test trap: ‘Mistaking a firm’s inability to profitably raise price above the current level because of current competitive constraints from certain rivals for an inability to exercise market power even after those rivals are excluded’; and + the unilateral SSNIP trap: ‘Mistaking a firm’s inability to profitably raise price above the current level unilaterally (i.e. assuming that rivals do not change their prices or outputs) for an inability to exercise market power by conduct that affects rivals’ output and price responses’. 100 Ibid 196. Nonetheless, note ‘[t]his precise benchmark, however, depends on the type of antitrust allegation being made. For example, if the claim is that certain conduct will permit a firm to raise its price above the current price level in the future, as in a typical merger analysis, then the proper benchmark is the pre-restraint, current price. The current price, however, is not necessarily the proper benchmark for other kinds of anti-competitive allegations, and ignoring this distinction can lead to the other traps identified [at n 99] above …’: ibid 196–197. 101 Ibid 195. 99

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 26 /

Date: 7/9

JOBNAME: Duns PAGE: 27 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 135

finding of market power, but such rivalry must be ‘non-discretionary … and thorough-going’, extending to ‘all dimensions of the price-product service packages offered to consumers and customers’.102 Such an approach recognises the notion of ‘co-opetition’,103 where alliances form and dissolve as interests coincide and diverge. In considering how the market power enquiry differs under a strategic, as distinct from classical, approach, it should be noted that certain factors remain unchanged. As is already apparent, barriers to entry remain key (although a more expansive view of what constitutes such barriers prevails). Similarly, market share still has a role. Krattenmaker et al state that ‘[m]arket share … provides some useful information for the analysis of Bainian market power. However, it should be one factor to consider, not the focus of the analysis.’104 Further, a considerably lower share will be significant than would be likely under a classical analysis.105 2.3 Conclusions: Comparing Classical and Strategic Approaches In considering these two approaches to market power, key distinctions emerge around the issues of duration (the extent to which the market power is required to be enduring) and scope (the extent of discretion granted to the monopolist or, put another way, when and how the conduct of the monopolist is constrained by actual or potential rivals). As discussed above, while a classical approach emphasises that market power must be persistent,106 a strategic approach is more likely to consider problematic short-lived or recurring imperfections, which are open to exploitation. Similarly, proponents of a strategic approach will recognise circumstances when the interests of key players coincide, effectively giving rise to a lack of constraint by reason of the strategic choices of the alleged monopolist’s main competitors. In the case of predation, for example, imperfections in the form of a spatial oligopoly based on product differentiation may give rise to a scenario where ‘localised competition means the entrant can be targeted. Few (if any) 102 Brunt (n 21) 99, citing Re Queensland Co-operative Milling Association Ltd (1976) 8 ALR 481, 515. 103 See the book of the same name by Adam M Brandenburger and Barry J Nalebuff (1998). 104 Krattenmaker, Lande and Salop (n 2) 260. 105 See for example the discussion in Glassman (n 95) 114–115. 106 For example, McFalls states, ‘[t]o have market power … implies having the ability to maintain prices profitably above competitive levels for a significant period of time’: McFalls (n 28) 656.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 27 /

Date: 7/9

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

136 Comparative competition law

other firms are hurt, and all will eventually benefit.’107 As these differences emphasise, however, the fundamental starting point will differ depending on the approach adopted. Whereas a strategic analysis will begin with the impugned conduct and work backwards, under a classical approach, market power is a threshold test which can, and should, be resolved before the conduct is examined. Nonetheless, it is important to remember that the sources of market power under either approach are largely the same. It is the case, however, that certain factors may be paid higher regard than others, depending upon one’s preferred approach, and may carry a different significance. For example, under either analysis, concentration is an important consideration. While a classical proponent may see concentration measures as a fair proxy in the absence of accurate information on marginal cost, a strategic proponent is more likely to look for evidence of oligopolistic interdependence. Similarly, both approaches will consider barriers to entry to be extremely important, although a classical approach is likely to be more restrictive in its identification of barriers, and to require such barriers to be more enduring than would be the case under a strategic approach. In a more marked point of departure, however, a strategic analysis is more likely to adopt a broader approach in identifying, and then to place greater significance upon, market imperfections that reduce elasticity of demand (or supply). Such imperfections include product differentiation, information asymmetries, the existence of network effects,108 switching costs and customer lock-in. At an extreme, the particular relationship between two players – e.g. franchisor and franchisee – could be such that, under a strategic approach, the existence of ‘relational’ market power may be asserted.109 On close analysis, however, the distinction between classical and strategic market power is not as clear cut as once was the case. In time, proponents of the classical approach have come to incorporate – to varying degrees – some of the insights of the strategic approach. As such, a classical analysis may take into account a broader (more strategic) view of barriers to entry. Similarly, such an analysis is likely to be more alive to the difficulty of identifying the appropriate ‘competitive’ price against which to benchmark the power of the alleged monopolist. Accordingly, 107

Campbell (n 38) 1652. See generally Salop and Romaine (n 17). 109 See Warren S Grimes, ‘Market Definition in Franchise Antitrust Claims: Relational Market Power and the Franchisor’s Conflict of Interest’ (1999) 67 Antitrust Law Journal 243. 108

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 28 /

Date: 7/9

JOBNAME: Duns PAGE: 29 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 137

by way of an iterative process, each approach appears to be learning from the other.

3. PRACTICAL EXAMPLES OF THE EXERCISE OF MARKET POWER Different jurisdictions regulate market power in different ways. This section uses Australian case law to demonstrate some of the ways market power can be exercised. In Australia existing market power is principally regulated via section 46 of the CCA. Section 46 currently has three elements, providing that a corporation with substantial market power cannot take advantage of that power for an anti-competitive purpose (specifically, one of three proscribed purposes as set out in the legislation). While there has been ongoing debate as to whether this is the ‘right’ test, all relevant case law in Australia to date is based on this test. 3.1 Refusals to Deal 3.1.1 Queensland Wire and NT Power Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd 110 was a private action and was the first section 46 case to go to the High Court (Australia’s ultimate appellate court). The relevant conduct was a constructive refusal to supply. BHP’s subsidiary competed with QWI at the retail level. When QWI sought to acquire an input (Y-bar) from BHP, BHP would sell only at the retail price of the (transformed) final product. It was clear that BHP had substantial market power. At the time BHP produced 97 per cent of the steel made in Australia and supplied about 85 per cent of the country’s requirements for steel and steel production. While all judges in the High Court agreed that BHP had contravened section 46, there were different approaches taken to assessing take advantage. One approach was direct inference – BHP’s conduct was unquestionably exclusionary and its exclusionary purpose could only be achieved if it had (and thus used) its market power. Other judges approached the question by asking how BHP would have behaved if Y-bar were sold in a competitive market: these judges concluded that BHP would not have forgone sales by refusing to deal with QWI. Thus they also concluded that only by reason of BHP’s market power was the conduct commercially feasible. 110

Columns Design XML Ltd

(1989) 167 CLR 177.

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 29 /

Date: 13/10

JOBNAME: Duns PAGE: 30 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

138 Comparative competition law

Fifteen years later, many of the same issues were again considered by the High Court in the NT Power case.111 Another private action, this concerned the denial of access to the electricity transmission and distribution system of the Northern Territory by a government-owned entity (PAWA). Both the first instance judge and the Full Court decided that PAWA benefited from crown immunity, such that section 46 did not apply. But the High Court (by majority) determined that crown immunity did not protect the conduct, and found – on largely identical reasoning to its decision in Queensland Wire – that all elements of section 46 had been established, notwithstanding arguments as to PAWA’s special obligations to the community. 3.1.2 Melway In between these two cases, yet another private action involving a refusal to deal went to the High Court, but this time the outcome was different. In Melway Publishing Pty Limited v Robert Hicks Pty Limited,112 Melway operated an exclusive distribution system for its extremely popular Melbourne street directory. The individuals behind Robert Hicks previously distributed Melway products via a particular company; following a breakdown in relations, one of the individuals left that company and set up a new one. Melway continued to supply the first company, but refused to supply the second (new) company. At both first instance and on appeal, Melway was found to have contravened section 46, with the judges applying the reasoning in Queensland Wire. On appeal to the High Court, the key issues were ‘taking advantage’ and purpose (as market power was effectively conceded – Melway had a 90 per cent market share which was protected by copyright in the directories and a very strong brand name). The majority noted that Melway’s conduct was in accordance with its long-standing distribution system under which each distributor was allocated a segment of the market. That system had been implemented before Melway had market power and indeed continued to operate in other geographic markets where it had no power. Critically, the majority found that, in refusing to supply Robert Hicks, Melway was not declining additional sales – rather, Robert Hicks’ business proposal primarily relied on it taking sales from the rival distributor. Accordingly, while the majority accepted the existence of a proscribed purpose (given their broad 111 112

Columns Design XML Ltd

/

(2004) 219 CLR 90. (2001) 205 CLR 1.

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 30 /

Date: 13/10

JOBNAME: Duns PAGE: 31 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Understanding market power 139

drafting), it considered that Melway was not taking advantage of its market power. Had it been subject to competitive forces, it would still have acted in the same way. 3.2 Bundling: Baxter In Australian Competition and Consumer Commission v Baxter Healthcare Pty Ltd 113 the ACCC alleged contraventions of sections 46 and 47 (exclusive arrangements which substantially lessen competition) in relation to Baxter’s bundling of ‘PD products’ with sterile fluid products. Baxter had a monopoly over sterile fluids.114 Its bundling strategy – which occurred via tender responses to state government purchasing authorities – was said to eliminate the effectiveness of any competition from rival PD suppliers. In simple terms, Baxter offered, for example, product X for $1 or products X and Y (when purchased together) for $1.10. Its pricing strategy meant that either the bundled price was predatory or the stand-alone offer amounted to monopoly pricing. The cases progressed through the lower courts in Baxter’s favour before the High Court determined that derivative crown immunity did not protect the conduct. It was then remitted to the Full Court for proper consideration. The Full Court’s judgment is very confusing. Ultimately, however, Baxter was found to have contravened section 46 where it lodged ‘non-compliant’ tender responses – i.e. responses that were outside the terms of the request for tender. Where its response was considered to fall within the terms of the request for tender, however, the Full Court found there to be no taking advantage. 3.3 Predatory Pricing: Boral In Boral Besser Masonry v Australian Competition and Consumer Commission115 the ACCC alleged that Boral engaged in predatory pricing in contravention of section 46, in circumstances of: (1) aggressive new entry; (2) cyclical downturn; (3) extensive expansion of capacity by the respondent; and (4) common interests with the other key player in the market. Boral’s pricing was below average variable cost for significant periods of time, and its conduct was shown to be directed to forcing out some of the smaller players in the market. That said, there was also 113 114 115

Columns Design XML Ltd

[2008] FCAFC 141. In the literal sense of the word. (2003) 215 CLR 374.

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 31 /

Date: 7/9

JOBNAME: Duns PAGE: 32 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

140 Comparative competition law

extensive evidence that Boral customers were able to play suppliers off against each other. On appeal, the Full Court found that section 46 had been contravened, as a consequence of a detailed strategic analysis of the conduct of the key players in the market. The High Court, however, overturned this decision and adopted a more structuralist approach. It found that there was no market power due to the extent of the constraints on Boral’s conduct. The majority considered it improper to look at the elements of taking advantage and purpose prior to establishing market power – therefore, the fact that another key player in the market would probably choose not to act as a constraint (as its interests coincided with Boral’s) was not considered relevant. 3.4 Raising Rivals’ Costs: Safeway In Australian Competition and Consumer Commission v Australian Safeway Stores Pty Limited 116 the ACCC alleged a range of contraventions by Safeway, including under section 46 in relation to its conduct vis-à-vis various bakers from which it acquired stock. Safeway objected to bakers selling discounted bread to other retailers, allowing them to (periodically) undercut Safeway. It consequently devised a ‘deletion’ policy, whereby it would delete the product of an offending baker where it engaged in this conduct. Along with Boral, Safeway is one of the few cases in Australia in which the existence (as opposed to use) of substantial market power has been controversial. Following an appeal from the first instance judgment, a majority of the Full Court found contraventions of section 46 on four occasions where Safeway ‘over-deleted’ the brand in question (i.e. deleted all products supplied by the relevant baker, not just the ‘offending’ product). The bases for the finding of market power included: the lack of alternative sources of supply for rival bread acquirers, the barriers to entry for establishing a business of similar size to Safeway (i.e. a state-wide grocery retailing network), Safeway’s market share and the existence of excess baking capacity. One member of the Full Court dissented, finding that the High Court’s reasoning in Boral, which required market power to be established as a threshold issue, meant that Safeway could not be considered to have substantial market power at all.

116

Columns Design XML Ltd

(2003) 129 FCR 339.

/

Job: Duns-Comparative_competition_law

/

Division: 07_Chapter6

/Pg. Position: 32 /

Date: 7/9

JOBNAME: Duns PAGE: 1 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

7. Antitrust treatment of intellectual property rights Michael A. Carrier *

1. INTRODUCTION One of the most challenging and important issues in antitrust law today involves the appropriate treatment of intellectual property (IP) rights. By its very nature, IP grants its owners rights to exclude. How antitrust (or, as it is referred to outside the United States (US), competition) law should address such a regime is far from clear. This chapter first explains the tension between the laws. It next traces the history of the intersection in the US. It then discusses several types of activity that demonstrate the tension, focusing on the law in the US and European Union (EU), the jurisdictions that have most closely examined these issues. In particular, it analyzes refusals to license, conduct in the pharmaceutical industry including settlements between brand and generic companies and ‘product hopping’ to new versions of drugs, innovation markets, standard-setting (in general and in the smartphone setting), and patent pools.

* Portions of this chapter are adapted from Michael A Carrier, Innovation for the 21st Century: Harnessing the Power of Intellectual Property and Antitrust Law (Oxford University Press 2009); Michael A Carrier, ‘Competition Law and Enforcement in the Pharmaceutical Industry’ in Ariel Ezrachi (ed), Research Handbook on International Competition Law (Edward Elgar 2012); Michael A Carrier, ‘A Roadmap to the Smartphone Patent Wars and FRAND Licensing’ (2012) 2 Competition Policy International: Antitrust Chronicle 4; Michael A Carrier, ‘An Antitrust Framework for Climate Change’ (2011) 9 Northwestern Journal of Technology and Intellectual Property 513; Michael A Carrier, ‘Solving the Drug Settlement Problem: The Legislative Approach’ (2010) 40 Rutgers Law Journal 83; and Michael A Carrier, ‘Introduction’ in Michael A Carrier (ed), Critical Concepts in Intellectual Property Law: Competition (Edward Elgar 2009).

141

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

142 Comparative competition law

2. CONFLICT For the past century, the intersection of the antitrust and IP laws has presented difficult issues for courts and commentators. Although the two systems attempt to increase total societal welfare, they pursue this goal through different paths. The foundation of the IP system – the patent and copyright laws in particular – is the right to exclude. Such a right is designed to allow innovators to charge prices high enough to recover their investment costs and gain profits, thereby encouraging future innovation. The exclusion at the core of IP nonetheless may be punished under the antitrust laws. These laws scrutinize activity that restricts competition, because such conduct could lead to higher prices, lower output, and (often) less innovation. Monopolists, for example, typically lack the constraints provided by competitive markets. Similarly, agreements between IP holders and licensees could restrict competition by their very nature. IP holders could impose limitations on licensees such as (1) quantity restrictions, which limit the number of products that can be sold; (2) royalty payments, which determine the royalties that can be received; (3) grantbacks, by which licensees extend rights to use improvements to licensors; and (4) territorial restrictions, which confine licensees to certain areas. On a larger scale, owners could combine their IP. Patentees could jointly set royalties for multiple patents in a patent pool. Copyright holders could offer blanket licenses that provide access to numerous (even millions of) copyrighted works. And IP owners could enter into joint ventures and mergers. This broad range of activity may make perfect sense from the standpoint of dispersing or exploiting the protected innovation. Crosslicensing and patent pools avoid bottlenecks. Blanket licenses reduce transaction costs. Licensing allows owners to maximize the use of their creations. Refusals to license are consistent with IP’s right to exclude. But the greater need for cooperation from the perspective of IP law could trigger the suspicion of the competition laws.

3. HISTORY Over the course of the past century, US courts have reconciled the antitrust and IP laws without any compass to guide them. As a result, their analysis has shifted dramatically. In the period immediately following the enactment of the Sherman Antitrust Act in 1890, courts treated IP

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 143

as paramount. By the middle of the 20th century, they had adopted an approach hostile to IP. By the 1980s, the pendulum had swung back, with courts deferring once again to IP. 3.1 IP Dominance In the period from 1890 to 1912, courts refused to impose antitrust liability for patent-based activity. They treated IP as owners’ private property and licensing arrangements as activity lying within the owners’ discretion. For example, in E. Bement & Sons v National Harrow Co., the Supreme Court (Court) upheld price fixing among competitors, explaining that ‘[t]he very object of the [patent] laws is monopoly’ and that ‘any conditions which are not in their very nature illegal … will be upheld by the courts.’1 Similarly, in A.B. Dick v Henry, the Court upheld a tying arrangement that required those who wished to license a patented machine to use certain unpatented supplies, concluding that ‘[a]rguments based on suggestions of public policy not recognized in the patent law are not relevant.’2 3.2 Antitrust Ascendance Congress responded to the Court’s decision in A.B. Dick by enacting the Clayton Act, which prohibited the tying of patented and unpatented products.3 Courts also began to limit the power of patentees. In Standard Sanitary Manufacturing Co. v United States, the Court recognized that IP rights were subject to the ‘positive prohibitions’ of the Sherman Act.4 And in Motion Picture Patents Co. v Universal Film Manufacturing Co., the Court asserted that the rights of patent owners flowed not from patent law but from the ‘general law [of] the ownership of … property.’5 In the mid 20th century, the Supreme Court adopted an even more aggressive stance toward IP. Three examples reveal its hostile treatment of patent ties, cross-licensing agreements, territorial restraints, and other arrangements. In International Salt Co. v United States, the Court found a tying arrangement to be per se illegal and refused to analyze the issue 1

186 US 70, 91 (1902). 224 US 1, 19 (1912). 3 15 USC §§ 12–27 (1914); see generally Herbert Hovenkamp, IP and Antitrust Policy: A Brief Historical Overview (University of Iowa Legal Studies Research Paper, Number 05-31, 2005) 11. 4 226 US 20, 49 (1912). 5 243 US 502, 513 (1917). 2

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

144 Comparative competition law

of market power since ‘the tendency of the arrangement to accomplishment of monopoly seems obvious.’6 In United States v Loew’s, the Court struck down ‘block booking,’ by which a party conditions the license or sale of desired movies on the buyer’s acceptance of a package containing unwanted films, explaining – in a presumption it would follow for more than 40 years – that ‘[t]he requisite economic power is presumed when the tying product is patented or copyrighted.’7 And in United States v Line Material Co., the Court attacked cross-licensing arrangements as price fixing even though the public could ‘obtain the full benefit of the efficiency and economy of the inventions’ only by using both products.8 Aligned with the Court’s increasingly hostile approach to tying and cross-licensing arrangements were the ‘Nine No-No’s,’ announced by the Department of Justice Antitrust Division in 1970 and followed for approximately a decade.9 The Nine No-No’s encompassed IP licensing activities that had no economically harmful aspects but which the agency regarded as suspect under the antitrust laws.10 3.3 IP Re-emergence Beginning in the 1970s, the courts retreated from this overtly hostile approach to various IP arrangements. They began to follow a more economics-based approach, analyzing the competitive effects of arrangements. Scholars affiliated with the Chicago School of Economics played a pivotal role in the transformation. One important case was Continental T.V., Inc. v GTE Sylvania, Inc., in which the Court concluded that a vast array of restraints affecting competition between different brands should be analyzed not as strict ‘per se’ violations but rather under a more lenient ‘rule-of-reason’ analysis.11 This holding has had a significant effect on IP licensing arrangements, which are nearly always upheld today.

6

332 US 392, 396 (1947). 371 US 38, 45 (1962). 8 333 US 287, 291 (1948). 9 Bruce B Wilson, Patent and Know-How License Agreements: Field of Use, Territorial, Price and Quantity Restrictions, Address Before the Fourth New England Antitrust Conference (November 6, 1970); Willard K Tom and Joshua A Newberg, ‘Antitrust and Intellectual Property: From Separate Spheres to Unified Field’ (1997) 66 Antitrust Law Journal 167, 178–79 n 67. 10 Ibid 179–81. 11 433 US 36, 51, 55, 59 (1977). 7

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 145

Other cases also strengthened IP rights. In BMI Music, Inc. v Columbia Broadcasting, Inc., the Court upheld blanket licenses, which allowed licensees to perform any of the millions of copyrighted musical works in the package.12 The Court in Berkey Photo, Inc. v Eastman Kodak Co. refused to punish Kodak’s failure to ‘predisclose’ its products to competitors, explaining that ‘withholding … advance knowledge of one’s new products … ordinarily constitutes valid competitive conduct.’13 And in Cal. Computer Prods., Inc. v IBM Corp., the Court found that IBM could redesign its products ‘to make them more attractive to buyers’ rather than ‘constrict[ing] its product development so as to facilitate sales of rival products.’14 In addition to the courts, Congress played a role in the rehabilitation of IP. In the Patent Misuse Reform Act of 1988, for example, Congress limited the range of activities that led to a finding of patent misuse, which prohibits a patentee from enforcing its patent against or recovering damages from an infringer. The legislation declared that two types of activity would no longer be considered misuse: (1) a refusal to license a patented item; and (2) the tying of a patented good to a second product when the inventor lacked market power in the tying product.15 Finally, and perhaps most important, Congress passed the Federal Courts Improvement Act of 1982, which created the Federal Circuit. This court, which has exclusive jurisdiction over appeals from final decisions of district courts on claims of patent infringement, unified disparate patentability standards and has tended to favor patents in cases arising at the patent–antitrust intersection.16 The transition to an increasingly dominant IP was crystallized in 1995 when the antitrust agencies issued the Antitrust Guidelines for the Licensing of Intellectual Property. The Guidelines embodied three principles: (1) IP is ‘essentially comparable’ to any other form of property; (2) IP does not automatically create market power in the antitrust context; and (3) IP licensing is generally procompetitive.17 12

441 US 1, 5, 19, 21–22 (1979). 603 F.2d 263, 281 (2d Cir. 1979). 14 613 F.2d 727, 744 (9th Cir. 1979). 15 Pub. L. No. 100-703, 102 Stat. 4676 (1988), codified at 35 U.S.C. § 271(d) (1994). 16 House of Representatives, No 97-321, 97th Cong., 1st Sess. 18 (1981). 17 US Department of Justice and Federal Trade Commission, Antitrust Guidelines for the Licensing of Intellectual Property (1995) ¶ 2.0 (hereafter ‘IP Guidelines’). 13

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

146 Comparative competition law

These principles made clear that IP should not be treated more harshly than real property, that IP does not necessarily demonstrate market power, and that licensing ‘can lead to more efficient exploitation’ of IP, which benefits consumers through ‘the reduction of costs and introduction of new products’ and ‘greater investment in research and development.’18 Finally, the Guidelines applied general principles to categories of activity such as horizontal restraints, tying arrangements, cross-licensing, patent pools, and the acquisition of IP.19

4. ACTIVITIES Analysis of the patent–antitrust intersection typically has occurred in the context of particular types of conduct. This section examines how courts have applied antitrust analysis to various patent-based activities. 4.1 Refusals to License: USA Refusals to license, which are undertaken by a single firm, most directly present the conflict between IP and antitrust law. A refusal to license is a natural consequence of IP’s right to exclude. But a court may view a firm’s refusal to share IP-protected products as predatory conduct constituting monopolization. This section explores the three most important US opinions, which were articulated by federal appellate courts at the end of the 20th century. Of the three, the most IP-friendly decision seems to have generated the longest coattails. In the first case, Data General Corp. v Grumman Systems Support Corp., the Court held that a party’s ‘desire to exclude others from use of its [IP-protected] work is a presumptively valid business justification.’20 The First Circuit found that such a presumption was not rebutted when a company refused to license a program diagnosing computer problems. In the second case, Image Technical Services, Inc. v Eastman Kodak Co. (Kodak II), the Ninth Circuit affirmed a monopolization verdict, finding that the Data General presumption could be rebutted by evidence of pretext.21 The Court found that such evidence existed where ‘the proffered business justification played no part in the [defendant’s] decision to act’ since ‘Kodak photocopy and micrographics equipment 18

Ibid §§ 2.1, 2.2, 2.3. Ibid §§ 5.1, 5.3, 5.5, 5.7. 36 F.3d 1147, 1187 (1st Cir. 1994). 125 F.3d 1195 (9th Cir. 1997).

19 20 21

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 147

require[d] thousands of parts, of which only [65] were patented,’ and Kodak’s parts manager testified that patents ‘did not cross [his] mind’ when the company instituted its parts policy.22 The Federal Circuit provided the third example in In re Independent Service Organizations Antitrust Litigation (Xerox), carving out three limited categories in which a patentee would not be immune from antitrust liability: (1) tying patented and unpatented products; (2) obtaining a patent through knowing and willful fraud; and (3) engaging in sham litigation.23 Because the Court concluded that Xerox’s refusal to sell its patented parts did not exceed the scope of the patent and did not fall within any of the three exceptions, it concluded that Xerox did not violate the antitrust laws.24 The Xerox case has been cited by lower courts in support of antitrust immunity for refusals to license. The Court in Townshend v Rockwell International Corporation, for example, found that ‘a patent owner has the legal right to refuse to license his or her patent on any terms’ and thus ‘a predicate condition to a license agreement cannot state an antitrust violation.’25 The Xerox case also is consistent with a line of cases that grant immunity as long as the challenged activity lies within the ‘scope’ of the patent.26 Finally, Xerox is consistent with Verizon Communications v Trinko, in which the Supreme Court not only held that an incumbent telephone company’s refusal to share its network with rivals did not constitute monopolization but also employed aggressive language that lauded the benefits of monopoly power; lamented antitrust law’s ‘considerable disadvantages,’ false positives, and negative investment effects; and bemoaned courts’ supervision of decrees and ‘carte blanche’ to force monopolists to ‘alter [their] way of doing business.’27

22

Ibid 1218–20. 203 F.3d 1322, 1326–28 (Fed. Cir. 2000). 24 Ibid. 25 No. C99-0400, 2000 US Dist. LEXIS 5070, 26 (N.D. Cal. Mar. 28, 2000). 26 See for example Zenith Radio Corp. v Hazeltine Research, Inc. 395 US 100, 136 (1969); Ethyl Gasoline Corp. v United States 309 US 436, 456 (1940); Motion Picture Patents Co. v Universal Film Mfg. Co. 243 US 502, 510 (1917); United States v Studiengesellschaft Kohle, m.b.H. 670 F.2d 1122, 1135 (D.C. Cir. 1981); United States v Westinghouse Elec. Corp. 648 F.2d 642, 647 (9th Cir. 1981); SCM Corp. v Xerox Corp. 645 F.2d 1195, 1206 (2d Cir. 1981). 27 Verizon Commc’ns Inc. v Law Offices of Curtis v Trinko, LLP 540 US 398, 412, 415 (2004). 23

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

148 Comparative competition law

4.2 Refusals to License: EU Refusals to license IP are more likely to be successfully challenged in the EU. In RTE & ITP v Commission (Magill), the European Court of Justice found that a refusal to deal (regarding copyrighted daily program listings) could constitute an abuse of dominance in ‘exceptional circumstances,’ which were met since: (1) the information was indispensable in creating a comprehensive weekly TV guide; (2) the refusal prevented the appearance of a new product; (3) there was no justification for the refusal; and (4) the stations ‘reserved to themselves the secondary market of weekly television guides by excluding all competition on that market.’28 In another important case, IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG, IMS refused to supply information on sales of drug products in a large number of small areas called ‘bricks.’29 Such a system allowed IMS to offer data without identifying sales by individual pharmacies. After finding that the criteria identified in Magill needed to be satisfied, the European Court of Justice found an abuse of dominance.30 The Court of First Instance in Microsoft Corp. v Commission of the European Communities synthesized these cases, stating that exceptional circumstances would be met when a refusal relates to a product indispensable to behavior on a neighboring market, excludes effective competition on that market, and prevents the appearance of a new product for which there is potential consumer demand.31 The Court found such circumstances met in the Microsoft case since the refusal: (1) covered indispensable interoperability information; (2) threatened to eliminate competition in the market for work group server operating systems; and (3) limited technical development.32 The Court’s recitation expanded liability from the Magill and IMS cases. It indicated that it could find liability even in the absence of one of the exceptional circumstances. And it extended the reach of the third factor from preventing a new product to limiting technical development.33 28

Joined Cases C-241/91 P & C-242/91 P, 1995 E.C.R. I-808 ¶¶ 49–50,

54–56. 29

Case C-418/01, 2004 E.C.R. I-5069 ¶¶ 10–11. Ibid ¶¶ 38, 52. 31 Case T-201/04, 2007 O.J. C269/80, ¶ 332. 32 Ibid ¶¶ 436, 620, 647–49. 33 Ibid ¶¶ 336, 649–58. See generally Renata B Hesse, ‘Counseling Clients on Refusal to Supply Issues in the Wake of the EC Microsoft Case’ (2008) 22 Antitrust 32, 33–34. 30

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 149

4.3 Pharmaceutical Industry The next category of conduct occurs in the pharmaceutical industry. Because of the importance of patent-antitrust issues in this setting and the complicated regulatory structure, background on the industry is helpful. For several reasons, the pharmaceutical industry presents unique competition issues. First, patents are particularly important in the industry. Each of the major studies that economists have undertaken in the past 50 years has shown that patents are the primary motivator for innovation in only a handful of industries, with pharmaceuticals at the top of the list.34 Second, generic entry significantly alters the market. The entry of generic versions of brand drugs dramatically lowers price. The first generic entrant prices its product, on average, approximately 10–30 percent lower than the brand drug.35 The presence of a second generic lowers the price to at least half the brand price, and often more than that.36 In markets in which several or more generics enter, the price falls to a quarter of the brand price or less.37 Third, in contrast to many markets, in which the party purchasing the good is the same as the one who decides whether to purchase it, the pharmaceutical industry is characterized by a complicated structure. The most fundamental disconnect, as the Federal Trade Commission (FTC) explained decades ago, is that ‘the consumer who pays does not choose, 34

Edwin Mansfield, ‘Patents and Innovation: An Empirical Study’ (1986) 32 Management Science, 174–75; Richard C Levin et al., Appropriating the Returns from Industrial Research and Development (Brookings Papers on Economic Activity 1987) 783, 802, 809; Wesley M Cohen et al, Protecting their Intellectual Assets: Appropriability Conditions and Why US Manufacturing Firms Patent (or Not) (NBER Working Paper 7552, 2000) 9. 35 Richard G Frank and David S Salkever, ‘Generic Entry and the Pricing of Pharmaceuticals’ (1997) 6 Journal of Economics & Management Strategy 75, 84 (generic-to-brand price ratio with one generic was approximately 69%); David Reiffen and Michael R Ward, ‘Generic Drug Industry Dynamics’ (2005) 87 The Review of Economics and Statistics 37, 44 (generic priced 12% below brand if only one on market); Atanu Saha et al, ‘Generic Competition in the US Pharmaceutical Industry’ (2006) 13 International Journal of Economic and Business Management 15, 28 (average generic-to-brand price ratio is 76% one month after generic entry, 54% by the end of first year, and 41% by end of second year). 36 US Department of Health and Human Services, Generic Competition and Drug Prices (FDA Center for Drug Evaluation and Research 2006). 37 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

150 Comparative competition law

and the physician who chooses does not pay.’38 Patients typically are not informed or educated about the product before purchase, with this information asymmetry reducing choice.39 This disconnect between the chooser and the user creates a gap that can be exploited. Fourth, a complex regulatory framework governs the introduction of drugs into the marketplace. In the US, the Hatch-Waxman Act affects numerous aspects of generic market entry. Congress passed the Act for two reasons: to foster generic competition and to promote brand-firm innovation. Congress sought to promote generic competition by creating a new type of drug application that allowed the generic to rely on the brand’s studies, resuscitating the ‘experimental use’ defense (which allowed generics to experiment on the drug even when it was still protected by patent), and giving a 180-day period of marketing exclusivity to the first generic to challenge a brand firm’s patent that filed a ‘paragraph IV’ certification, claiming that the patent was invalid or not infringed.40 At the same time, the Act assisted brand firms by extending the patent term, providing non-patent market exclusivity for new chemical entities and new clinical investigations, and granting an automatic 30-month stay of the Food and Drug Administration (FDA) approval for brand firms that sued generics that had challenged a patent’s invalidity or claimed non-infringement.41 Although the Hatch-Waxman regime is limited to the US, analysis in Europe also is complicated by regulatory requirements such as marketing authorizations (by Member States or by the European Commission (EC)) and pricing and reimbursement determinations.42 This chapter focuses most directly on settlements between brand and generic firms, and ‘product hopping’ that involves switches from one version of a drug to another. In addition to these two activities, drug companies have engaged in an array of other activities that could raise antitrust concern. Much of this behavior occurs near the end of the patent 38

Bureau of Consumer Protection, Drug Product Selection: Staff Report to the FTC (1979) 2–3. 39 Maura J Monaghan and Michael S Monaghan, ‘Do Market Components Account for Higher US Prescription Prices?’ (1996) 30 Annals of Pharmacotherapy 1489, 1490. 40 Michael A Carrier, Innovation for the 21st Century: Harnessing the Power of Intellectual Property and Antitrust Law (Oxford University Press 2009), 348–49, 351–52. 41 Ibid 349–50. 42 European Commission, Pharmaceutical Sector Inquiry Final Report (European Commission Staff Working Paper 15-49, 2009).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 151

term. The activity has included: (1) fraud in obtaining a patent, including false representations to the US Patent and Trademark Office; (2) sham litigation that discourages entry through frivolous patent claims against generic firms; (3) authorized generics, which are identical to brand-name drugs but which brand firms market as generics to discourage entry;43 (4) improper listings of patents in the ‘Orange Book’ (which lists drugs and related patents) to block generic substitutes; and (5) filing citizen petitions with the FDA to delay the approval of generic drugs.44 To pick one example, citizen petitions allow the public to express concerns about a drug’s safety. Brand firms, however, have used this process to delay generic entry. My recent study of the activity found that petitions have recently increased, that the FDA has denied 81 percent of petitions filed by brand firms, and that congressional legislation has not reduced the number of petitions.45 4.4 Settlements: USA The most pressing antitrust issue in the pharmaceutical industry today involves settlements by which brand-name drug companies pay generic firms to settle patent litigation and delay entering the market. Many commentators have voiced concern with these agreements, in particular with large ‘reverse payments’ from brand to generic firms. These payments, which differ from typical licensing payments that flow from challengers to patentees, may even exceed what the generic could have earned by entering the market. If a patent is valid and infringed, a patentee could rely on the patent itself to restrict competition, and an agreement allowing a generic to enter before the end of the patent term could increase competition. On the other hand, if a patent is invalid or not infringed, there is no legitimate justification for delaying competition, with the agreement serving as a cover for market allocation. Further increasing concern in this setting are the parties’ aligned incentives. Because the brand typically makes more by keeping the generic out of the market than the two parties would receive by 43 Federal Trade Commission, Authorized Generic Drugs: Short-Term Effects and Long-Term Impact (2011). 44 See generally David Balto, ‘Roadblocks on the Pharmaceutical Competition Highway: Strategies to Delay Generic Competition’ [2008] Duke Law & Technology Review. 45 Michael A Carrier and Daryl Wander, ‘Citizen Petitions: An Empirical Study’ (2012) 34 Cardozo Law Review 249.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

152 Comparative competition law

competing in the market, the parties have an incentive to split the monopoly profits, making each better off than if the generic had entered. But a central difficulty in analyzing this behavior is that the appropriate antitrust-law treatment of settlements depends on the validity of the patent and existence of infringement, and these issues require the analysis of complex issues such as patent claim construction and infringement analysis, which courts cannot easily consider in the midst of an antitrust case. In the first case in which a US appellate court examined the issue, the Sixth Circuit in In re Cardizem CD Antitrust Litigation found ‘a classic example of a per se illegal restraint of trade’ when a firm agreed not to market any generic version of a brand drug (including those not covered by the patent).46 Courts, however, quickly retreated from such analysis. The Eleventh Circuit in Schering-Plough Corp. v FTC upheld settlements that fell ‘within the protections of the … patent’ and stated that it ‘cannot be the sole basis for a violation of antitrust law’ for a brand firm with a patent to pay a generic competitor.47 The Second Circuit in In re Tamoxifen Citrate Antitrust Litigation concluded that as long as ‘the patent litigation is neither a sham nor otherwise baseless’ or beyond the patent’s scope, the patentee can enter into a settlement ‘to protect that to which it is presumably entitled: a lawful monopoly over the manufacture and distribution of the patented product.’48 The Federal Circuit in In re Ciprofloxacin Hydrochloride Antitrust Litigation found that settlements fell ‘well within’ the patentee’s rights, that patents bestowed ‘the right to exclude others,’ that a ‘long-standing policy’ favored settlements, and that the crucial inquiry was ‘whether the agreements restrict competition beyond the exclusionary zone of the patent.’49 Finally, the Eleventh Circuit in FTC v Watson Pharmaceuticals, Inc. stated that, in the absence of sham litigation or fraud in obtaining a patent, ‘a reverse payment settlement is immune from antitrust attack so long as its anti-competitive effects fall within the scope of the exclusionary potential of the patent.’50 The Court also rejected the FTC’s proposed rule that a plaintiff can state a claim if it is ‘more likely than not’ that the 46 47 48 49 50

Columns Design XML Ltd

/

332 402 466 544 677

F.3d F.3d F.3d F.3d F.3d

896, 907–08 (6th Cir. 2003). 1056, 1076 (11th Cir. 2005). 187, 208–09, 213 (2d Cir. 2006). 1323, 1332–33, 1336, 1337 (Fed. Cir. 2008). 1298, 1312 (11th Cir. 2012).

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 12 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 153

patent ‘would not have blocked generic entry earlier than the agreedupon entry date.’51 The trend from 2005 until early 2012 thus was strongly in favor of settlements. Courts emphasized the importance of settlements, the link between settlements and innovation, the presumption of patent validity, the scope of the patent, and the ‘natural’ status of reverse payments in upholding the payments. But in July 2012, in In re K-Dur Antitrust Litigation, the Third Circuit applied the most aggressive scrutiny to these agreements in years. It rejected the ‘scope of the patent’ test that other courts had relied on, explaining that such a test assumes the validity at issue in the case and is not relevant where the issue is infringement (where the patentee bears the burden of proof).52 The Court also concluded that a reverse payment was ‘prima facie evidence of an unreasonable restraint of trade’ that could be rebutted only if the settling parties could show that the payment was for a purpose other than delay or that it ‘offers some pro-competitive benefit.’53 To resolve the split among the courts, the Supreme Court granted certiorari to review the Eleventh Circuit’s ruling, and in FTC v Actavis,54 rejected the ‘scope of the patent’ test that the Federal, Second, and Eleventh Circuits had followed. The Court found it ‘incongruous’ to ‘determine antitrust legality by measuring the settlement’s anticompetitive effects solely against patent law policy, rather than by measuring them against procompetitive antitrust policies as well.’55 The Court also found that the specific restraint at issue had the ‘potential for genuine adverse effects on competition’ since ‘payment in return for staying out of the market … keeps prices at patentee-set levels.’56 In addition, the Court highlighted the harms from a brand payment to a generic, which ‘in effect amounts to a purchase by the patentee of the exclusive right to sell its product, a right it already claims but would lose if the patent litigation were to continue and the patent were held invalid or not infringed by the generic product.’57 51

Ibid. 2012 WL 2877662 (July 16, 2012); Michael A Carrier, ‘Unsettling Drug Patent Settlements: A Framework for Presumptive Illegality’ (2009) 108 Michigan Law Review 37, 66. 53 2012 WL 2877662, 16. 54 133 S. Ct. 2223 (2013). 55 Ibid 2231. 56 Ibid 2235. 57 Ibid 2234. 52

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

154 Comparative competition law

The Court revealed its strong preference for determining patent strength by examining the payment rather than the patent. The ‘size of the unexplained reverse payment can provide a workable surrogate for a patent’s weakness, all without forcing a court to conduct a detailed exploration of the validity of the patent itself.’58 Even strong patents are not immune from the concern with payments, because an unexplained payment on a ‘particularly valuable patent … likely seeks to prevent the risk of competition.’59 The Court concluded that ‘the FTC must prove its case as in other rule-of-reason cases.’60 It found that the ‘size of the payment’ serves as ‘a strong indicator of power,’ thereby allowing the plaintiff to dispense with required showings of market power.61 And it instructed future courts to analyze payments’ ‘size, … scale in relation to the payor’s anticipated future litigation costs, … independence from other services for which it might represent payment, and … lack of any other convincing justification.’62 In the period since Actavis, lower courts have begun to interpret the decision. One issue they are currently wrestling with is whether payment includes only cash transfers or whether it includes other consideration, such as a brand promise not to introduce its own generic.63 4.5 Settlements: EU The issue of settlements has taken a different path in Europe. There have been fewer enforcement actions and, until recently, less attention on the issue. One of the most important activities was the 2009 pharmaceutical sector inquiry. The EC commenced this inquiry in January 2008, seeking to determine the reasons for delayed generic entry and for the decline in new medicines reaching the market.64 The inquiry concentrated on ‘those 58

Ibid 2236–37. Ibid 2236. 60 Ibid 2237. 61 Ibid 2236. 62 Ibid 2237. 63 Compare In re Lamictal Direct Purchaser Antitrust Litigation, 2014 WL 282755 (D.N.J. Jan. 24, 2014) (limiting payment to cash) with In re Niaspan Antitrust Litigation, 2014 WL 4403848 (E.D. Pa. Sept. 5, 2014) (stating that payment is not limited to cash). 64 European Commission, Pharmaceutical Sector Inquiry Final Report (European Commission Staff Working Paper 3, 2009). 59

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 155

practices which companies may use to block or delay generic competition as well as to block or delay the development of competing originator products.’65 The report found that 22 percent of settlements from 2000 to 2008 involved payments from the innovator to the generic firm. The remainder involved agreements that did not limit the generic’s ability to market the drug (52 percent) as well as settlements that limited generic entry but did not involve payment (26 percent).66 The inquiry noted differences between settlements in the US and in the EU. Most significant, the 180-day exclusive marketing period granted to the first-filing generic in the US does not have a counterpart in the EU. In addition, settlements in the EU have not been marked by brand firms’ promises not to launch authorized generics that would compete with the generics. Finally, there have been fewer ‘side deals’ in Europe in which the parties enter into agreements on products unrelated to the patent at the heart of the settlement.67 As of the time this chapter went to press, the inquiry has been followed up by four monitoring exercises. The first, analyzing settlements between the middle of 2008 and end of 2009, found that the number of settlements that involved a restriction on generic entry and value transfer from the originator to the generic fell from 22 percent of settlements (between 2000 and 2007) to 10 percent.68 In addition, the amount of money involved in the settlements fell from more than €200 million to less than €1 million.69 The report offered potential reasons for the decline, which included increased awareness of competition law scrutiny, continued monitoring of settlements, and the opening of proceedings in a case against France’s Les Laboratoires Servier.70 The second monitoring exercise, one year later, traced an even more significant decline in the most concerning agreements. Of the 89 patent settlements in the period, only three limited the generic’s ability to market its product and included a value transfer from originator to 65

Ibid 4. Ibid 267–85. 67 Ibid 290–91. 68 European Commission, First Report on the Monitoring of Patent Settlements (period: mid 2008–end 2009) (July 2010) 12–13. 69 European Commission, ‘Antitrust: Commission Welcomes Decrease of Potentially Problematic Patent Settlements in EU Pharma Sector’ (Press Release, 5 July 2010) . 70 European Commission (n 68) 13. 66

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 15 /

Date: 13/10

JOBNAME: Duns PAGE: 16 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

156 Comparative competition law

generic.71 The report found that the companies continued to settle cases, but took different, less concerning, forms: 61 percent did not restrict generic entry and 36 percent limited entry but did not involve a value transfer to the generic.72 While the third monitoring exercise found an increase (to 11 percent) in settlements involving restrictions on generic entry and a value transfer, this was partially explained by a case (based on an ‘erroneously assumed expiry of exclusivity rights’ caused by ‘an administrative error by public authorities’) that did not ‘trigger potential competition law scrutiny.’73 The fourth monitoring exercise found a reduction to 7 percent of settlements involving payment for delayed entry.74 The EC also targeted individual companies. It fined Lundbeck and generic competitors €146 million for violating Article 101 of the Treaty on the Functioning of the European Union (TFEU) by agreeing ‘to delay the market entry of cheaper generic versions of Lundbeck’s branded Citalopram, a blockbuster antidepressant.’75 It also fined Servier and generic rivals €427 million for settlements that delayed generic entry on Perindopril, a blockbuster blood pressure medicine.76 4.6 Product Hopping: USA Another type of activity that has gained attention on both sides of the Atlantic involves product hopping. This activity involves switches from one version of a drug (eg, capsule) to another (eg, tablet). Brand firms often make minor changes to their products as their patents are about to expire. One reason for such conduct involves companies’ interests in shifting patients from the old version of a drug (which faces generic 71

European Commission, Second Report on the Monitoring of Patent Settlements (period: January–December 2010) (July 2011) 11. 72 Ibid 12. 73 European Commission, 3rd Report on the Monitoring of Patent Settlements (period: January–December 2011) (25 July 2012). 74 European Commission, 4th Report on the Monitoring of Patent Settlements (period: January–December 2012) (9 December 2013). 75 European Commission, ‘Antitrust: Commission fines Lundbeck and other pharma companies for delaying market entry of generic medicine’ (19 June 2013) . 76 European Commission, ‘Antitrust: Commission fines Servier and five generic companies for curbing entry of cheaper versions of cardiovascular medicine’ (9 July 2014) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 16 /

Date: 13/10

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 157

competition) to the new version (which does not, thereby providing additional exclusivity). Two important cases in the US reveal the conduct. The first case involves TriCor, a drug for lowering cholesterol and triglycerides. In 1998 brand firm Abbott received FDA approval of its 200 mg capsule version of TriCor. But it then made changes to the drug just as two generic firms were about to enter the market, first lowering the drug’s dosage strength to 160 mg and switching from a capsule to a tablet, and later moving to a 145 mg tablet. It also changed the code for TriCor capsules in the National Drug Data File (NDDF) (comprehensive set of drug database elements, drug pricing, and clinical information) to obsolete which, combined with the other conduct, ‘prevented pharmacies from filling TriCor prescriptions with a generic capsule formulation.’77 By the time the generics received FDA approval for their version of the 200-mg capsule in 2002, generic substitution was no longer possible. In Abbott Laboratories v Teva, the district court applied a ‘rule-ofreason’ approach in denying Abbott’s motion to dismiss. It found that Abbott had prevented a choice between products ‘by removing the old formulations from the market while introducing new formulations.’78 The court also found that through Abbott’s ‘allegedly manipulative and unjustifiable formulation changes,’ Teva and Impax were not able to offer generic substitutes for TriCor, which was the alleged ‘cost-efficient means of competing’ in the market.79 In short, Abbott’s activity directly threatened the price-lowering benefits of laws allowing pharmacists to substitute generic versions of brand drugs. A second example of product hopping in the US is offered by AstraZeneca’s conversion from heartburn drug Prilosec to Nexium.80 The two drugs, which contained the same active ingredient, worked in nearly an identical manner. Less than a year before the Prilosec patent expired, AstraZeneca obtained a patent for Nexium, providing it with 13 additional years of protection.81 At that point, AstraZeneca ceased promoting Prilosec and aggressively encouraged doctors to switch from Prilosec to

77

Abbott Laboratories v Teva 432 F. Supp. 2d 408, 416 (D. Del. 2006). Ibid 422. 79 Ibid 423. 80 Walgreen Co. v AstraZeneca Pharmaceuticals LP 534 F. Supp. 2d 146 (D.D.C. 2008). 81 The facts are taken from: ibid 148–49. 78

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 17 /

Date: 8/9

JOBNAME: Duns PAGE: 18 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

158 Comparative competition law

Nexium (by comparing Prilosec unfavorably to Nexium). It also caused managed care organizations not to cover the cost of generic Prilosec.82 The district court granted AstraZeneca’s motion to dismiss. It concluded that ‘there is no allegation that AstraZeneca eliminated any consumer choices.’ To the contrary, the company actually ‘added choices.’83 This conclusion, however, would appear to oversimplify the reality of pharmaceutical markets. As explained above, a doctor that selects a medication does not pay, and a party that pays (patient, insurance company, etc.) does not choose. 4.7 Product Hopping: EU There have also been two important product-hopping cases in Europe. In 2010 the European General Court upheld findings of the EC that AstraZeneca had abused its dominant position by blocking and delaying market access to the generic version of ulcer medication Losec. The Court found that AstraZeneca gave misleading information to patent offices so it could get a Supplemental Protection Certificate (SPC), which provided it with an additional period of patent protection.84 The Court concluded that the Commission correctly found an abuse of dominance from ‘objectively misleading representations by an undertaking in a dominant position’, which led the offices to grant SPCs that AstraZeneca was not entitled to, which ‘result[ed] in a restriction or elimination of competition.’85 The Court also found that AstraZeneca deregistered capsule marketing authorizations to ‘delay and make more difficult’ the marketing of generics.86 The Court concluded that the Commission ‘did not err in finding that AZ’s deregistration of the Losec capsule marketing authorizations in Denmark, Norway and Sweden, in conjunction with the swing in AZ’s sales from Losec capsules towards Losec tablets in those countries, amounted to an abuse of a dominant position’ since it 82 Walgreen Co v AstraZeneca Pharmaceuticals LP Plaintiffs’ Statement of Points of Authority in Opposition to Defendants’ Motion to Dismiss 1:06-cv02084-RWR, 10 (DDC 21 May 2007). 83 534 F. Supp. 2d 149, 151 (D.D.C. 2008). 84 General Court of the European Union, The General Court essentially upholds the decision of the Commission which found that the AstraZeneca Group abused its dominant position by preventing the marketing of generic products replicating Losec (Press Release No 67/10, 1 July 2010). 85 Case T-321/05 AstraZeneca v Commission [2010] OJ C 221/33. 86 General Court of the European Union (n 84).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 18 /

Date: 8/9

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 159

‘restrict[ed] access to the market of generic products in those countries.’87 Similarly, ‘the Commission did not err in taking the view that that conduct constituted an abuse of a dominant position in Sweden, inasmuch as it was such as to restrict parallel imports of Losec capsules in that country.’88 A second example of product hopping is provided by the case involving Gaviscon, a drug used to treat heartburn and acid reflux. The UK National Health Service uses software that allows doctors to substitute generics for brand drugs if a generic name is published.89 The publication of the generic name increases price competition, as drug companies ‘have a strong incentive to compete on price to persuade pharmacies to choose to dispense their products.’90 Reckitt Benckiser, in contrast, sought to prevent this substitution, even though ‘it would have been commercially irrational’ to withdraw its profitable medicine from the market if not for ‘the gains [it] expected to derive from hindering the development of full generic competition.’91 In 2011 the UK Office of Fair Trading (OFT) found that the objective of Reckitt Benckiser was to ‘delay for as long as possible the introduction of a generic name’ and to ‘replace/cannibalize all current … sales’ with ‘the new patent protected variant.’92 It found that Reckitt Benckiser had sought to ensure that doctors ‘would be unable to identify prescription packs of Gaviscon products against which open scripts could be issued and against which pharmacies could choose to dispense a Gaviscon product or equivalent alternative.’93 Reckitt Benckiser’s withdrawal of a profitable medicine was not ‘competition on the merits’ but ‘tended to restrict competition or was capable of having that effect.’94 The OFT concluded that Reckitt Benckiser’s withdrawal of its initial product constituted an abuse of dominance.95 The EC pharmaceutical sector inquiry also emphasized product hopping. It explained that originator companies launch follow-on products 87

Case T-321/05 AstraZeneca v Commission [2010] OJ C221/33, 864. Ibid. 89 Graeme Wearde, ‘Gaviscon Prescriptions Investigated by Office of Fair Trading’, The Guardian (23 February 2010). 90 Decision of the OFT, Abuse of a Dominant Position by Reckitt Benckiser Healthcare (UK) Limited and Reckitt Benckiser Group plc, OFT Decision CA98/02/2011 ¶ 1.8 (2011). 91 Ibid 6.114. 92 Ibid 136, 2.164. 93 Ibid 1.10. 94 Ibid 6.162. 95 Ibid. 88

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

160 Comparative competition law

shortly before the first-generation product loses exclusivity.96 The brand firm facilitates such a switch by ‘channelling … demand from the first product to the follow-on product’ and by ‘delay[ing] or prevent[ing] generic entry for the sensitive period of the product switch.’97 In particular, the Commission underscored the importance of the timing of the switch, with brands suffering reduced sales and prices if generics entered the market before or at the same time as the follow-on product.98 For 13 of the 22 second-generation products surveyed in the report, the new product was launched before the first lost its exclusivity, with an average lead time of 17 months.99 4.8 Innovation Markets While an array of challenged activity in the pharmaceutical industry has taken place in settings in which there are already products on the market, other conduct has occurred in earlier stages. One of those stages involves research and development (R&D), which takes place in what is known as an ‘innovation market.’100 In the 1990s the US antitrust agencies began to challenge mergers that affected these markets. The theory behind innovation markets is that a merger between the only two (or two of a few) firms conducting R&D in a particular area might increase the incentive to suppress at least one of the research paths. With no other firms ready to enter the market, the merging firms might choose not to introduce a second product that would reduce sales of the first. The concept of innovation markets has been criticized. It may be difficult to identify all the firms in a particular innovation market. The relationship between R&D and innovation is unclear. After a half-century of debate between adherents of Joseph Schumpeter (favoring concentration) and Kenneth Arrow (favoring competition), no clear answer has emerged to the question of the type of market structure most conducive to innovation.101 96

European Commission (n 64), 351. Ibid 356. 98 Ibid. 99 Ibid 361, fig. 138. 100 IP Guidelines (n 17) ¶¶ 3.2.2, 3.2.3. 101 Joseph A Schumpeter, Capitalism, Socialism and Democracy (Routledge 1942) 88, 103, 105, 106; Kenneth J Arrow, ‘Economic Welfare and the Allocation of Resources for Invention’ in Kenneth J Arrow, Essays in the Theory of Risk-Bearing (3rd edn, North-Holland 1976) 144, 157–58. 97

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 20 /

Date: 13/10

JOBNAME: Duns PAGE: 21 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 161

I have argued that the critiques leveled against innovation markets are less relevant in the pharmaceutical industry: barriers to entry provided by patents and a lengthy regulatory process restrict the universe of potential innovators; innovation-market analysis targets activity that could not be challenged at a later time; and competition plays a role in innovation.102 In addition, the realities of pharmaceutical innovation would seem to call for a recognition of the wildly varying odds of success of reaching the market. On average, compounds in preclinical development face a less than 1 in 4,000 chance of reaching the market while those in Phase III of clinical studies (the last pre-market stage) have a greater than 1 in 2 chance.103 Mergers involving compounds in late clinical studies thus are more likely to threaten the suppression of research paths. In recent years, most of the agencies’ challenges to mergers in innovation markets have involved the pharmaceutical industry. In particular, the FTC challenged: (1) the Roche–Genentech merger affecting certain drugs to treat HIV/AIDS; (2) the American Home Products– American Cyanamid merger for the vaccine for rotavirus (a form of diarrhea causing children’s deaths); (3) the Glaxo Wellcome merger for non-injectable treatment for migraines; (4) the Upjohn–Pharmacia merger for chemotherapy drugs targeting colon cancer; (5) the Baxter–Immuno merger for fibrin sealants, which stop bleeding and heal wounds; (6) the Ciba-Geigy–Sandoz merger for gene therapy products; (7) the Pfizer– Warner-Lambert merger for inhibitors for solid cancerous tumors; and (8) the Glaxo Wellcome–Smith Kline Beecham merger for a prophylactic herpes vaccine.104 While enforcement in innovation markets remains controversial, it could continue to play a role in antitrust enforcement in the pharmaceutical industry. 4.9 Standard-setting: General One IP-related activity that has gained attention in recent years is standard-setting. Standards, which allow products to work together, are 102

Michael A Carrier, ‘Two Puzzles Resolved: Of the Schumpeter-Arrow Stalemate and Pharmaceutical Innovation Markets’ (2008) 93 Iowa Law Review 393. 103 See Food & Drug Administration Department of Health and Human Services, FDA and the Drug Development Process: How the Agency Ensures that Drugs Are Safe and Effective (Just the Facts Publication No. FS 02-5, 2002); Carrier, ‘Two Puzzles Resolved’ (n 102), 417–19. 104 See Carrier, ‘Two Puzzles Resolved’ (n 102) 448.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 21 /

Date: 13/10

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

162 Comparative competition law

often set by private industry groups known as standard-setting organizations (SSOs). These standards (often called interface or interoperability standards) are ubiquitous in our economy and are especially needed in network effects markets, in which users benefit from an increase in the number of other users in the system. A telephone or e-mail system, for example, becomes more valuable as more users connect to it. Even though standards are vital, competition law traditionally viewed the process of setting standards with suspicion. SSOs tend to be composed of industry rivals discussing sensitive information such as price. Despite this concern, competitors frequently have good reason to engage in such discussions. Before the selection of a standard, an SSO often can choose from an array of alternative technologies. But after the SSO selects a standard, particularly if it incorporates a patented technology, the owner can block others from using the standard by obtaining injunctive relief or imposing royalties high enough that members are effectively prevented from using the standard. The excessive licensing terms could reflect not just the value of the patent but also the significant costs of switching to a new technology. The patentee, in other words, could ‘hold up’ the standard’s implementation. This threat of holdup explains why SSOs have required members to provide certain information before a standard’s selection. Some SSOs have mandated that participants disclose IP (typically patents and sometimes patent applications and other IP rights) that could be implicated by the standard. Others have adopted search rules that require members to search for relevant IP rights. A popular way to address the holdup problem is to require patentees to agree before a standard is selected to license their technologies on reasonable terms. This set of licensing promises is known as FRAND (fair, reasonable, and nondiscriminatory) in Europe and RAND (reasonable and nondiscriminatory) in the US. One even stricter policy requires members to announce their most restrictive licensing terms (including maximum royalty rates) before a standard’s adoption.105 Some standard-setting cases have involved participants that engaged in manipulation of the process. In Allied Tube & Conduit Corp. v Indian Head, Inc., manufacturers packed an SSO meeting to defeat a rival’s attempt to certify its product for the standard.106 In American Society of Mechanical Engineers v Hydrolevel Corp., a vice president of the firm 105

Letter from Thomas O Barnett, Assistant Attorney General, Department of Justice to Robert A Skitol (30 October 2006) 5–6 (hereafter ‘VITA letter’). 106 486 US 492, 496–97 (1988).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 163

dominating the market, in his capacity as vice chairman of the relevant SSO subcommittee, interpreted a code so that a rival’s product was deemed unsafe.107 Further, in In re American Society of Sanitary Engineering, the SSO protected existing manufacturers by refusing to approve a new toilet tank fill valve that was safer, cheaper, and more durable.108 Courts or the FTC found an antitrust violation in each of the three cases. Other cases have involved a participant’s deceptive conduct. In In re Dell, an SSO adopted a standard based on a Dell representative’s promise that a proposal did not infringe its IP. But after the standard became widely adopted, Dell sought payment for the use of its IP.109 The FTC also challenged the misrepresentation of the Union Oil Company of California (Unocal), which failed to disclose relevant patents and applications. After manufacturers had spent billions of dollars to comply with the new standards, Unocal sought to collect royalties that would have cost consumers more than $500 million annually.110 The FTC obtained a consent decree halting the behavior in both cases. In In re Rambus, the FTC found that Rambus had ‘engaged in representations, omissions, and practices likely to mislead JEDEC members,’ which ‘significantly contributed to its acquisition of monopoly power.’111 In 2008 the DC Circuit, focusing on causation, reversed this conclusion, finding that JEDEC might have adopted Rambus’s technology even absent any deception. As a result, any ‘loss of an opportunity to seek favorable licensing terms’ did not, without more, constitute ‘antitrust harm.’112 In the final case, the FTC filed a complaint against Negotiated Data Solutions (N-Data) in January 2008. N-Data licensed patents used in equipment employing Ethernet, a ubiquitous networking standard. N-Data’s predecessor had committed to license its technology for a

107

456 US 556, 559–63, 577–78 (1982). 106 F.T.C. 324 (1985). 109 In re Dell 121 F.T.C. 616, 617, 623–24 (1996). 110 In re Union Oil Co. of California, No. 9305, ¶¶ 9, 10, 26, 31, 61, 63 (F.T.C. Nov. 25, 2003), www.ftc.gov/os/2003/11/031126unionoil.pdf, rev’d, No. 9305 (F.T.C. July 7, 2004); Federal Trade Commission, ‘Statement In re Union Oil Co. of California, Docket No. 9305’ (June 10, 2005) . 111 In re Rambus, No. 9302, 2006 WL 2330117, at 5–6, 8, 50, 67, 104, 118 (Aug. 2, 2006). 112 Rambus Inc. v F.T.C. 522 F.3d 456, 466–67 (D.C. Cir. 2008). 108

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

164 Comparative competition law

one-time royalty of $1,000 per licensee, but N-Data later demanded royalties ‘far in excess of that commitment.’113 By a vote of 3-2, the FTC challenged N-Data’s action. It did not allege a violation of the Sherman Act, but instead claimed an unfair method of competition and unfair act or practice under Section 5 of the Federal Trade Commission Act. The majority asserted that N-Data’s behavior harmed consumers and businesses and explained that its exercise of its ‘unique’ authority was needed to ‘preserv[e] a free and dynamic marketplace.’114 Two dissents were filed. Former Chairman Deborah Majoras bemoaned a lack of meaningful guideposts for liability and subsequent former Chairman William Kovacic lamented a failure to distinguish between unfair methods of competition and unfair acts or practices.115 4.10 Standard-setting: Smartphones Standard-setting issues today frequently appear in the setting of smartphones. Many issues arise because of RAND or FRAND promises, which typically are made for standard essential patents (SEPs), which are patented technologies essential to a standard’s implementation. In the smartphone setting, standardized technologies have included wireless broadband technologies (such as WiFi), video compression technologies (H.264), and telecommunications standards (4G LTE).116 Other patents – such as a ‘slide-to-unlock’ technique or method of e-mail notification

113 Federal Trade Commission, ‘FTC Challenges Patent Holder’s Refusal to Meet Commitment to License Patents Covering “Ethernet” Standard Used in Virtually All Personal Computers in US’ (23 January 2008) ; Federal Trade Commission, ‘Statement in re Negotiated Data Solutions LLC, File No. 0510094’ (23 September 2008) . 114 Ibid. 115 Dissenting Statement of Chairman Majoras, in re Negotiated Data Solutions LLC, File No. 0510094, (lack of guideposts); Dissenting Statement of Commissioner Kovacic, in re Negotiated Data Solutions LLC, File No. 0510094, (lack of distinction). 116 Department of Justice, ‘Statement of the Department of Justice’s Antitrust Division on Its Decision to Close Its Investigations of Google Inc.’s Acquisition of Motorola Mobility Holdings Inc. and the Acquisitions of Certain Patents by Apple Inc., Microsoft Corp., and Research in Motion Ltd’ (13 February 2012) 3 (hereafter DOJ letter).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 24 /

Date: 10/9

JOBNAME: Duns PAGE: 25 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 165

service – are non-essential patents (non-SEPs) and do not typically allow an industry to be held up since they are not needed to use the standard. The assurance of FRAND licensing for SEPs plays a crucial role in reducing the likelihood of injunctions that would remove products from the market. But in some cases, companies have sought injunctions even though they had promised to license their SEPs on reasonable terms. In other cases, patentees have challenged competitors’ uses of non-SEPs that do not threaten to block the product from the market but could lead to less desirable workarounds. In 2012 the EC began an investigation of Samsung, determining whether it breached its obligation to the European Telecommunications Standards Institute (ETSI) to license its patents related to third-generation (3G) mobile and wireless telecommunications systems on FRAND terms. In particular, the EC examined whether Samsung’s attempts to obtain injunctions against competitors constituted an abuse of a dominant position prohibited by Article 102 of the TFEU.117 In April 2014 the EC accepted a settlement by which Samsung ‘committed not to seek any injunctions in the European Economic Area (EEA) on the basis of any of its SEPs, present and future, that relate to technologies implemented in smartphones and tablets against any company that agrees to a particular framework for licensing the relevant SEPs.’118 The framework provides for ‘a negotiation period of up to 12 months,’ and if the parties cannot agree, it calls for ‘a third party determination of FRAND terms by a court if either party chooses, or by an arbitrator if both parties agree.’119 The EC also investigated Motorola Mobility (MMI) for its conduct related to injunctions after promising to license its mobile and wireless communications patents on FRAND terms.120 In April 2014 the EC found that ‘it was abusive for Motorola to both seek and enforce an injunction against Apple in Germany on the basis of an SEP which it had 117

European Commission, ‘Antitrust: Commission opens proceedings against Samsung’ (31 January 2012) . 118 European Commission, ‘Antitrust: Commission accepts legally binding commitments by Samsung Electronics on standard essential patent injunctions’ (29 April 2014) . 119 Ibid. 120 European Commission, ‘Antitrust: Commission sends Statement of Objections to Motorola Mobility on potential misuse of mobile phone standardessential patents’ (6 May 2013) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 25 /

Date: 13/10

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

166 Comparative competition law

committed to license on FRAND terms’ where ‘Apple had agreed to take a licence and be bound by a determination of the FRAND royalties by the relevant German court.’121 The Commission also ‘found it anticompetitive that Motorola [had] insisted, under the threat of the enforcement of an injunction, that Apple give up its rights to challenge the validity or infringement by Apple’s mobile devices of Motorola SEPs.’122 The EC nonetheless decided not to levy a fine because there was ‘no case-law by the European Union Courts dealing with the legality under Article 102 TFEU of SEP-based injunctions,’ and ‘national courts have so far reached diverging conclusions on this question.’123 Additional guidance for analyzing FRAND issues appears in the EC’s 2011 Guidelines on horizontal cooperation agreements. The Guidelines explain that FRAND licensing prevents patentholders from ‘making the implementation of a standard difficult’ by refusing to license, requesting ‘excessive’ fees, or imposing ‘discriminatory’ royalties.124 In particular, the obligation requires participants ‘to provide an irrevocable commitment in writing’ to offer to license patents on FRAND terms.125 Where there are disputes, analysis of whether fees are ‘unfair or unreasonable’ is to be based on whether the fees ‘bear a reasonable relationship’ to the patent’s economic value.126 The Guidelines provide a safe harbor from Article 101(1) of the TFEU (which targets anti-competitive agreements) for standard-setting conduct that is transparent, does not impose an obligation to comply with the standard, and provides access on FRAND terms.127 In the US, issues related to smartphone patent licensing arose in the approval by the US Department of Justice (DOJ) of three transactions. In February 2012 the DOJ found that competition was ‘unlikely to [be] substantially lessen[ed]’ by (1) Google’s acquisition of Motorola’s portfolio of 17,000 patents and 6,800 patent applications; (2) Apple’s acquisition of the nearly 900 patents originally held by Novell and 121

European Commission, ‘Antitrust: Commission finds that Motorola Mobility infringed EU competition rules by misusing standard essential patents’ (29 April 2014) . 122 Ibid. 123 Ibid. 124 European Commission, Guidelines on the Applicability of Article 101 of the Treaty on the Functioning of the European Union to Horizontal Co-operation Agreements (2011 OJ (C 11) 1) ¶ 287. 125 Ibid ¶ 285. 126 Ibid ¶ 289. 127 Ibid ¶ 280.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 27 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 167

purchased in 2010 by CPTM (a coalition including Apple, EMC, Microsoft, and Oracle); and (3) the acquisition by the ‘Rockstar’ group (made up of Apple, Microsoft, and RIM) of the 6,000 patents and applications available in the Nortel bankruptcy auction. Central to the DOJ’s approval were the promises made by the acquiring parties to license SEPs. Apple, for example, stated that ‘[s]eeking an injunction would be a violation of the party’s commitment to FRAND licensing,’ and Microsoft promised to ‘not seek an injunction or exclusion order against any firm on the basis of … essential patents.’ The DOJ concluded that ‘Apple and Microsoft made clear that they will not seek to prevent or exclude rivals’ products from the market in exercising their SEP rights.’128 One concern with these promises, however, is that they might not bind successors. For example, the CEO of the Rockstar consortium, which acquired the Nortel patents, has claimed that it is not bound by Apple and Microsoft’s promises since ‘we are separate’ and ‘that does not apply to us.’129 The FTC turned to the SEP issue for the first time in 2012, entering into a settlement with Bosch, which had threatened an injunction against what it found to be a willing licensee. The FTC found that Bosch’s predecessor had made a FRAND pledge but had ‘allegedly reneged on these commitments and pursued injunctions blocking competitors from using the standardized technologies, even though the competitors were willing to license the technology on FRAND terms.’130 Shortly thereafter, as part of a wide-ranging settlement that centered on Google’s search-engine behavior, the FTC required Google (as part of its acquisition of MMI) to follow certain procedures in relation to SEPs. Before seeking an injunction, Google was required to provide a potential licensee with a written offer containing the material license terms and also provide an offer of binding arbitration to determine terms not agreed upon.131 The agreement additionally made clear that a licensee could negotiate the terms of a license with Google for at least six months and 128

DOJ letter (n 116) 2–3. Robert McMillan, ‘How Apple and Microsoft Armed 4,000 Patent Warheads’ (Wired, 21 May 2012) . 130 Federal Trade Commission, ‘FTC Order Restores Competition in US Market for Equipment Used to Recharge Vehicle Air Conditioning Systems’ (26 November 2012) . 131 Federal Trade Commission, ‘Analysis of Proposed Consent Order To Aid Public Comment, In the Matter of Motorola Mobility LLC and Google Inc, File 129

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 27 /

Date: 8/9

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

168 Comparative competition law

that (if the licensee did not choose this path) Google was not able to seek an injunction unless it provided license terms for at least six months and an option to arbitrate for at least 60 days.132 4.11 Patent Pools A patent pool involves a single organization – either a new entity or one of the original patentholders – that licenses the patents of two or more companies to third parties as a package.133 As the antitrust enforcement agencies have recognized, patent pools tend to be procompetitive in ‘integrating complementary technologies, reducing transaction costs, clearing blocking positions, and avoiding costly infringement litigation.’134 For that reason, the agencies have upheld pools related to: (1) MPEG-2, a video compression technology underlying the transmission, storage, and display of digitized moving images and sound tracks;135 (2) DVD-ROM and DVD-video formats describing ‘the physical and technical parameters for DVDs for read-only-memory and video applications;’136 and (3) 3G wireless communication systems.137 These pools were composed of essential patents, which means that the product or standard at issue in the pool could not have been produced without infringing the patent. Essential patents do not have substitutes and typically are complementary, possessing a greater value if the licensee can use other essential patents. On the other hand, patent pools can present anti-competitive harm when they facilitate the combination of substitute patents, which are not necessary for the use of a technology in the pool and which present alternate ways of creating products that otherwise would have been used in competition with each other. For example, competing patents made up No. 121-0120’ (3 January 2013) 6. 132 Ibid 7–8. 133 Carl Shapiro, ‘Navigating the Patent Thicket: Cross Licenses, Patent Pools, and Standard Setting’ in Adam B Jaffe et al (eds), Innovation Policy and the Economy (MIT Press 2001) 119, 132. 134 IP Guidelines (n 17) ¶ 5.5. 135 Letter from Joel I Klein, Acting Assistant Attorney General, Department of Justice Antitrust Division, to Gerrard R Beeney (26 June 1997) (hereafter ‘MPEG-2 pool letter’). 136 Letter from Joel I Klein, Assistant Attorney General, Department of Justice Antitrust Division, to Gerrard R Beeney (16 December 1998). 137 Letter from Charles A James, Assistant Attorney General, Department of Justice Antitrust Division, to Ky P Ewing (12 November 2012).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 28 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Antitrust treatment of intellectual property rights 169

the Summit-VISX pool, which consisted of lasers used in photorefractive keratectomy (PRK), a form of eye surgery employed to correct vision disorders. In its Complaint, the FTC explained that, if not for the pool, the two firms would have competed against each other ‘by using their respective patents, licensing them, or both.’138 These substitute patents were not necessary for the use of a technology in the pool, but presented alternate ways of creating products that otherwise would have competed with each other. Patent pools also could reduce innovation. One context in which this has arisen involves grantback clauses. Grantbacks are arrangements by which a licensee agrees to extend to the IP licensor the right to use the licensee’s improvements to the licensed technology.139 The concern is that grantbacks could reduce innovation incentives by reducing the return from follow-on inventions.140 This concern has been addressed through requirements that grantback clauses be limited as narrowly as possible (such as to essential, not substitute, patents).141 The distinction that the US agencies have drawn between essential and substitute patents parallels that in the EU. The EC, in its Guidelines on the application of Article 81 (now Article 101) of the EC Treaty to technology transfer agreements, explained that ‘[t]he inclusion in the pool of substitute technologies restricts inter-technology competition,’ that pools ‘substantially composed of substitute technologies … amount[ed] to price fixing between competitors,’ and that ‘[w]hen a pool is composed only of technologies that are essential and therefore by necessity also complements,’ it does not violate Article 81.142 Even if it is not always obvious whether patents are substitutes or complements,143 the concept is valuable to focus the analysis on the relationship among the patents, and is more nuanced than many other frameworks that have been used to analyze antitrust and IP issues. 138

In re Summit Tech., Inc., F.T.C. No. 9286 ¶ 8. IP Guidelines (n 17) ¶ 5.6. 140 Hartmut Schneider, ‘Analyzing Patent Pools: Will They Pass Muster Abroad?’ (16 August 2010) Law 360. 141 MPEG-2 pool letter, at 13 (restriction of grantback clauses to essential patents makes it ‘unlikely that there is any significant innovation left to be done that the grantback could discourage’). 142 European Commission, Guidelines on the application of Article 81 of the EC Treaty to technology transfer agreements (¶¶ 219, 220) 2004. 143 Herbert Hovenkamp et al, IP and Antitrust: an Analysis of Antitrust Principles Applied to Intellectual Property Law (2nd edn, Aspen Publishers 2002) 34.2, 34.8–34.10. 139

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 29 /

Date: 8/9

JOBNAME: Duns PAGE: 30 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

170 Comparative competition law

5. CONCLUSION For more than a century the intersection of the antitrust and IP laws has presented an array of challenging and important issues. This chapter has focused on the tension between the laws and how that tension has been revealed in an array of conduct including refusals to license, conduct in the pharmaceutical industry, innovation markets, standard-setting, and patent pools. As we go forward, these issues will continue to play a crucial role in the success and effectiveness of the antitrust laws.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 10_Chapter7

/Pg. Position: 30 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

8. Current issues in merger law Julie Clarke *

1. INTRODUCTION The globalization of markets, combined with the extraordinary expansion of merger control laws over the past two decades,1 has resulted in an increasing number of mergers inviting multiple regulatory responses.2 This has had a significant impact on the complexity, time and cost associated with transnational mergers and has highlighted the differences in law, policy and procedure employed by more than 70 jurisdictions now adopting targeted merger regimes.3 * Part of the research work conducted for this chapter was completed while I was a visitor at the Melbourne Law School in November 2012 and a visiting fellow at the Institute of Advanced Legal Studies, London from December 2012 to March 2013 and I would like to thank those institutions for their support during that time. 1 Dabbah observes that the geographic expansion of the field of competition law is one that has not been ‘seen in the case of any other branch of law’: Maher M Dabbah, International and Comparative Competition Law (Cambridge University Press 2010) 1; See also Arvid Fredenberg, ‘Introduction’ in Konkurrensverket – Swedish Competition Authority (ed), More Pros & Cons of Merger Control 2012 (Konkurrensverket – Swedish Competition Authority 2012) 9. 2 See generally Lise Davey and John K Barker, Merger Review Benchmarking Report (Competition Bureau (Canada) 2001) 7; ICPAC, ‘International Competition Policy Advisory Committee to the Attorney General and Assistant Attorney General for Antitrust – Final Report’ (Department of Justice (US) 2000) 2. 3 The precise number of merger regimes is difficult to ascertain and fluctuates. Dabbah and Lasok identify 70 regimes with established merger review laws (Albania, Argentina, Republic of Armenia, Australia, Austria, Barbados, Belgium, Bosnia and Herzegovina, Brazil, Republic of Bulgaria, Canada, Chile, China, Costa Rica, Croatia, Republic of Cyprus, Czech Republic, Denmark, Estonia, European Economic Area, European Union, Finland, France, Federal Republic of Germany, Greece, Hong Kong, Hungary, Iceland, India, Republic of Indonesia, Ireland, Israel, Italy, Japan, Kenya, Korea, Latvia, Lithuania, Republic of Macedonia, Malta, Mexico, the Netherlands, New Zealand, Norway, Pakistan, Peru, Republic of the Philippines, Poland, Portugal, Romania, Russia, Serbia,

171

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

172 Comparative competition law

By contrast with other areas of competition law and policy, the treatment of mergers involves a significant regulatory component, with most jurisdictions adopting ex ante notification and suspension obligations for mergers exceeding defined thresholds. The justification for this lies in the structural change to the market affected by the merging of assets, personnel and intellectual property, which are difficult to reverse.4 However, ex ante regulation also has the consequence of subjecting the vast majority of benign or beneficial mergers to the cost and delay associated with administrative scrutiny. This cost and delay has the potential to jeopardize time-sensitive transactions or postpone expected efficiency gains. Where markets extend beyond domestic borders, these costs are multiplied and the slowest and Singapore, Slovakia, Slovenia, South Africa, Spain, Sri Lanka, Sweden, Switzerland, Taiwan, Thailand, Republic of Tunisia, Turkey, Ukraine, United Kingdom, United States of America, Uzbekistan, Venezuela, Zambia) and several more with what they term ‘nascent merger control regimes’ – in particular, Azerbaijan, Belarus, Kazakhstan, Moldova and Zimbabwe all have mandatory pre-merger notification provisions where certain conditions are met and Luxembourg and Panama have a voluntary system. They further identify more than 100 jurisdictions with some form of competition law and a further 30 currently seeking to implement such policy: Maher Dabbah and Paul Lasok QC (eds), Merger Control Worldwide (2nd edn, Cambridge University Press 2012). Others claim a higher number of merger regimes, although this is likely to include regimes with generic competition laws capable of capturing merger activity; see, for example, claiming that ‘there are more than 100 merger control regimes in force’: Volker Weiss and Michael Mayer, ‘Identifying Filing Obligations and Beyond: Merger Control in Cross-Border Transactions’ in Nigel Parr and Ruth Sander (eds), Merger Control 2013 (International Comparative Legal Guides 2013); claiming 115 jurisdictions adopt merger laws as part of their competition policy: John Arden, ‘Record Number of Jurisdictions Regulate Mergers, New Aspen Publication Finds’ (Trade Regulation Talk, 8 January 2009) http://traderegulation.blog spot.com/2009/01/record-number-of-jurisdictions-regulate.html. See generally Casey Cogut and Sean Rodgers, ‘Global Overview’ in Casey Cogut (ed), Getting the Deal Through, Mergers & Acquisitions in 48 Jurisdictions Worldwide 2003 (Law Business Research Ltd 2003) 3; ICN, ‘Merger Notification Filing Fees’ (2005) International Competition Network Report April 2005, 2 ; Chris Noonan, The Emerging Principles of International Competition Law (Oxford University Press 2008) 62. 4 In the US, for example, the Guidelines expressly state that they ‘reflect the congressional intent that merger enforcement should interdict competitive problems in their incipiency’: Department of Justice and Federal Trade Commission, US Horizontal Merger Guidelines (2010) 1 (hereafter US Horizontal Merger Guidelines).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 173

most prescriptive jurisdiction will influence or determine the time at which the merger can close, if at all, and on what conditions.5 Achieving the optimal balance between preserving the integrity of the relevant market, while keeping to a minimum the costs incurred by parties and agencies in the review of lawful mergers, is one of the greatest challenges facing the international competition law community in relation to merger control. Meeting this challenge demands increased levels of convergence on policy, law and, most importantly, procedure. More importantly, however, it requires increased cooperation in relation to mergers affecting markets which transcend national borders. Although there have been substantial efforts towards convergence and cooperation in merger review over the past decade, particularly through the work of the Organisation for Economic Co-operation and Development (OECD) and the International Competition Network (ICN),6 important differences remain, highlighted in recent years by the emergence of new and significant ex ante merger regimes, most notably in China, India and Brazil.7 This chapter examines the existing similarities and differences in the purpose, substance and administration of merger control regimes and examines the potential for enhanced convergence and increased international cooperation in merger review in the future.

2. THE RATIONALE FOR MERGER LAW AND REGULATION The rapid expansion of merger control laws and competition laws generally has revitalized debate about what competition law is and should 5

In some cases it may be possible to consummate aspects of a merger, carving out a slow or hostile jurisdiction, but this will not always be possible. 6 See for example Rachel Brandenburger, ‘International Cooperation: Taking a Broader View’ (Remarks prepared for the 29th American Chamber of Commerce Annual EU Competition Policy Conference, Brussels, Belgium, 6 December 2012). In 2011 the ICN approved a project on international enforcement cooperation and the OECD have also agreed to a cooperation work program: ICN Steering Group, ‘International Enforcement Cooperation Project’ (ICN Website, 2003) ; OECD, ‘OECD Global Forum on Competition’ (OECD Website) . 7 Brazil has recently introduced a system of pre-merger notification and review, replacing an ex post system of review: see generally Ron Knox, ‘An Anxious Transition’ (2012) 15(6) Global Competition Review 15.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

174 Comparative competition law

be about.8 Despite more than 100 years of modern competition law9 and the global proliferation of competition laws in recent decades, this fundamental question remains highly controversial.10 Very broadly, competition policy is predicated on the notion that a competitive business environment is better for society, both economically and socially, than one which is not; in particular, a competitive market is more desirable than a monopolistic one. The ICN recently identified the protection of competition as the sole goal towards which merger laws ought to be directed11 and it is therefore not surprising that most jurisdictions express the object of their law as being the prevention of conduct that would cause significant harm to competition.12 However, the lack of any universal definition of ‘competition’13 or consensus about what constitutes its promotion, means that regimes

8

Dabbah (n 1) 2. Examples of competition laws date back to at least Roman times. However, ‘modern’ competition law is typically said to date from the introduction of the Sherman Act of 1890 in the US, although Canada introduced the Competition Act 1889 one year earlier. 10 See, for example, Kaplow observing that this issue remained both ‘complex and controversial’: Louis Kaplow, ‘On the Choice of Welfare Standards in Competition Law’ in Daniel Zimmer (ed), The Goals of Competition Law (Edward Elgar 2012) 26. See also Dabbah, who notes the lack of consensus on the answer to the question ‘what should the goals of competition law be …’: Dabbah (n 1) 2. 11 Recommendation 1A and associated comments: ‘The purpose of competition law merger analysis is to identify and prevent or remedy only those mergers that are likely to harm competition significantly’; ‘… merger review law should not be used to pursue other goals’; and the ‘appropriate goal of agency intervention to prohibit or remedy a merger is to restore or maintain competition affected by the merger, not to enhance premerger competition’: ICN, ‘Recommended Practices for Merger Analysis’ (Merger Working Group 2008, 2010) . 12 See generally Henry C Thumann, ‘Multijurisdictional Regulation of Monopoly in the Global Market’ [2008] Wisconsin Law Review 261, 265. See also Richard M Stuer, ‘The Simplicity of Antitrust Law’ (2012) 14(2) University of Pennsylvania Journal of Business Law 543. 13 See, for example, Stucke ‘… no satisfactory comprehensive definition of competition exists’: Maurice Stucke, ‘What is Competition?’ in Daniel Zimmer (ed), The Goals of Competition Law (Edward Elgar 2012) 28; Andreas Fuchs, ‘Characteristic Aspects of Competition and their Consequences for the Objectives of Competition Law – Comment on Stucke’ in Daniel Zimmer (ed), The Goals of Competition Law (Edward Elgar 2012) 53. 9

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 4 /

Date: 13/10

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 175

which identify this objective may nevertheless be doing so with a view to promoting a diverse range of underlying policy goals.14 Identifying this policy is important because of the impact it may have on the interpretation and application of the laws themselves. It is particularly important in the case of mergers which, in most cases, are efficiency enhancing. Assessing whether mergers ought to be permitted therefore involves balancing competing priorities and knowing which factors are intended to carry more weight in this analysis depends upon an understanding of the underlying policy goals.15 It is principally the economic benefits associated with preventing anticompetitive mergers that have provided the genesis for the recent proliferation of competition laws.16 Effective competition policies tend to improve market outcomes, with monopoly, in general, leading to reduction in quality and output and, by extension, consumer harm.17 Merger activity, particularly horizontal, but also to a lesser degree vertical and conglomerate, has the potential to distort markets and impede competition. Most established regimes have, therefore, accepted that a fundamental purpose of competition policy is to promote economic efficiency,18 with merger

14

See, for example, Budzinski, who observes that ‘the absence of a unifying theory of competition … implies that different regimes will legitimately base their theories of competitive harm on diverging economic approaches’: Oliver Budzinski, ‘International Antitrust Institutions’ (2012) Ilmenau Economics Discussion Papers, Vol 17, No 72, July 2012, 7. See also UNCTAD, ‘Model Law on Competition’ (TD/RBP/CONF.7/L.1, UNCTAD 2010) paras 8–9; ICN, ‘Advocacy and Competition Policy Report’ (ICN Conference, Italy, 2002) 32; Kaplow (n 10) 3. 15 Thus, for example, if the predominant goal of merger law is economic welfare, with a focus on total welfare, then merger-generated efficiencies – even those likely to benefit the merging parties alone – will carry more weight than would be the case if the priority was identified as consumer welfare, which considers the distribution of generated efficiencies. 16 See, however, Dabbah, who observes that many policies – including US – clearly intended a ‘broader spectrum of values’: Dabbah (n 1) 28. 17 See, Dabbah, observing that ‘… generally monopoly does seem to lead to poor quality, restriction in output and harm to consumers. …’: ibid 26. See also Donald Hay, ‘The Assessment: Competition Policy’ (1993) 9(2) Oxford Review of Economic Policy 1, 1. 18 American Bar Association, ‘Report on Antitrust Policy Objectives’ (2003) Section for Antitrust Law of the American Bar Association Report .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

176 Comparative competition law

regulation directed towards ensuring that a competitive economic environment is maintained in the relevant domestic economy by preventing creation of entities that have the ‘incentive and ability to exercise market power’,19 either through unilateral or coordinated conduct. Debate continues, however, about precisely which economic welfare standard should be adopted; in particular whether a consumer or total welfare approach is most appropriate for competition policy.20 Much has been, and will continue to be, written on this subject, despite the fact that most of the time little turns on the precise economic welfare standard applied in individual merger analysis.21 While economists typically favour a total welfare approach,22 it is the consumer welfare standard that currently dominates merger assessment in practice.23 In addition to economic welfare considerations, many jurisdictions also protect competition for the purpose of promoting broader social, moral

19 UNCTAD, ‘Model Law on Competition’ (TD/RBP/CONF.7/L.1, UNCTAD 2010) 51. 20 See for example Herbert J Hovenkamp, Implementing Antitrust’s Welfare Goals (University of Iowa Legal Studies Research Paper, Number 12-39, November 2012); Kaplow (n 10) 3–4. See also D Daniel Sokol and William Blumenthal, ‘Merger Control: Key International Norms and Differences’ in Ariel Ezrachi (ed), Research Handbook on International Competition Law (Edward Elgar 2012) 329–330; Debate has not been assisted by the fact that the total welfare model is often referred to by its proponents as ‘consumer welfare’: Robert H Bork, The Antitrust Paradox (Free Press 1978). See also Barak Y Orbach, ‘The Antitrust Welfare Paradox’ (2011) 7 Journal Competition Law and Economics 133; Einer Elhauge, ‘Chapter 1: Introduction and Overview to Current Issues in Antitrust Economics’ in Einer Elhauge (ed), Research Handbook on the Economics of Antitrust Law (Edward Elgar 2012) 13. 21 See Hovenkamp, observing that few ‘if any decisions have turned on the difference’. The issue of standards will be most relevant to the issue of balancing increased efficiencies against any likely reduction in competition: Hovenkamp (n 20) 4; compare Kaplow (n 10) 3. 22 Abraham L Wickelgren, ‘Chapter 10: Issues in Antitrust Enforcement’ in Einer Elhauge (ed), Research Handbook on the Economics of Antitrust Law (Edward Elgar 2012) 272. See also Sokol and Blumenthal (n 20) 329–330. 23 In this respect, Hovenkamp has recently argued that, considering ‘efficiency and administrability consumer welfare emerges as the most practical goal of antitrust enforcement’: Hovenkamp (n 20) 29. Hovenkamp also observes that ‘courts almost invariably apply a consumer welfare test’: Hovenkamp (n 20) 6.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 6 /

Date: 13/10

JOBNAME: Duns PAGE: 7 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 177

and political objectives,24 such as ensuring ‘fair’ rules25 in the market, preventing ‘high levels of industrial concentration’,26 protecting small business, protecting jobs or as a mechanism for wealth distribution.27 At times it has also been used to further national or regional industrial policy.28 This is perhaps not surprising and the existence of conflicting objectives that nevertheless produce similar, or even identical, substantive laws is neither unusual nor paradoxical in the context of competition law, given the capacity for competitive market economies to achieve multiple

24 See, for example Budzinski observing that ‘the goals of competition policy diverge in focusing on different welfare standards (consumer welfare vs total welfare; allocative welfare vs dynamic welfare; etc), on different intermediate goals (different efficiency concepts, protection of the competitive process, liberty of competition, structure-conduct-performance concepts, etc) and on the inclusion of ‘non-welfare’ goals (market integration, economic development, fairness, high employment, international competitiveness, etc).’: Budzinski (n 14) 7. See also Sokol and Blumenthal (n 20) 328–329; ICN, ‘Advocacy and Competition Conference’ (n 14) 32–34; See further Kathryn McMahon, ‘Developing Countries and International Competition Law and Policy’ (2009) Warwick School of Law Research Paper No 2009/11, 17; Eleanor M Fox, ‘Post-Chicago, Post-Seattle and the Dilemma of Globalization’ in Antonio Cucinotta, Roberta Pardolesi and Roger Van Den Bergh (eds), Post-Chicago Developments in Antitrust Law (Edward Elgar 2002) 76–77. See also Robert H Lande, ‘Proving the Obvious: The Antitrust Laws Were Passed to Protect Consumers (Not Just to Increase Efficiency)’ (1999) 50 Hastings Law Journal 959, 963–964; Robert H Lande, ‘Wealth Transfers as the Original and Primary Concern of Antitrust: The Efficiency Interpretation Challenged’ (1982) 34 Hastings Law Journal 65, 68; Damien Geradin and Ianis Girgenson, ‘Industrial Policy and European Merger Control – A Reassessment’ (TILEC Discussion Paper No 2011-053 2011); Noonan (n 3) 63–73; Michael A Utton, The Economics of Regulating Industry (Blackwell 1986) 129. 25 See, for example, Hay noting that there may ‘be ethical and social objections to the absence of competition: it is simply not fair that large firms or cartels should be able to oppress smaller competitors and/or customers by charging prices that greatly exceed the costs of supply’: Hay (n 17) 1. 26 American Bar Association, ‘Report on Antitrust Policy Objectives’ (2003) Section for Antitrust Law of the American Bar Association Report . See also Stucke (n 13) 29–30; Dabbah (n 1) 28. 27 This is directly related to the economic debate over total welfare versus consumer welfare. 28 See for example Geradin and Girgenson (n 24).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

178 Comparative competition law

economic and social goals.29 These broader ‘goals’ are also frequently considered more politically appealing than pure economic objectives.30 Given the complementary nature of many of these goals, it is only when they are brought into direct conflict that it is necessary to delve deeper to identify the underlying policy objectives. National laws are, however, often far from clear about the specific purpose of their competition law31 and more work needs to be done, by government and agencies, to clarify appropriate objectives. This does not mean that the goals identified for different jurisdictions should necessarily be homogeneous32 or that the articulation of identical goals will necessarily produce identical outcomes in different jurisdictions. There is no universal consensus, among government, regulators or commentators, as to an ‘optimal’ approach33 and, even where identical goals do exist, what should be considered harmful by reference to those goals may legitimately differ depending on the jurisdiction. This may reflect the size (small markets may be less able to self-correct than large ones),34 geographic location (greater concentration might be tolerated in countries with ready access to imports compared with those more geographically isolated),35 the desire for regional consistency (as is the

29

Antitrust Modernization Commission, ‘Report and Recommendations’ (2007) 26, fn 22; See also discussion in Sokol and Blumenthal (n 20) 327–330. 30 They are also more easily understood than goals expressed in pure economic terms, particularly when economists tend to conflict over the meaning to be given to particular terms or predicted economic outcomes: see, for example, Dabbah noting that ‘economists have not been the most consensusbuilding community in the world of competition law … different economic approaches speak of identical issues differently’: Dabbah (n 1) 28. 31 See for example Lande, ‘Wealth Transfers as the Original and Primary Concern of Antitrust’ (n 24) 83. 32 Dabbah (n 1) 3. 33 See for example Budzinski (n 14) 4, 7. 34 McMahon noting that EU rules have generally been aimed at achieving ‘short-run competitive rivalry rather than the Chicago School goal of economically efficient outcomes based on “self-correcting” markets’: McMahon (n 24) 17. 35 See Gal, who claims that in small economies the ‘concern for ensuring that a sufficient number of competitors operate in each market should be subordinated … to the more compelling necessity of serving a small population efficiently … [In small economies] protection of competition would blockade many mergers that have positive welfare effects’: Michal S Gal, ‘Extra-territorial Application of Antitrust – The Case of a Small Economy (Israel)’ (NYU Center for Law, Economics and Organization 2009) 200–201.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 8 /

Date: 13/10

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 179

case among most European states), agency experience and resources,36 market concentration and level of economic development and industrial advancement. It might also be the case that different tests (or at least different methodology in applying those tests) must be used within the same economies for different industries,37 with hi-tech industries often singled out as in need of special attention.38 The challenges for government in defining appropriate goals arise not only as a result of political considerations, but also because of the lack of academic and economic consensus about the benefits of competition law or the desirability and legitimacy of utilizing competition policy to pursue non-economic objectives. Despite these difficulties and despite the fact that, in many cases, merger outcomes may be identical regardless of the specific goals identified, the onerous procedural and substantive compliance obligations imposed on parties, often supported by significant penalties, demand clearly articulated objectives. Identifying these goals helps justify the regulatory intrusion, supports consistent interpretation of the laws and limits the potential for competition law to be used as a political tool to garner support of specific interest groups.

3. NATIONAL GOALS FOR INTERNATIONAL MARKETS Related to the issue of policy goals for competition law generally is whether or not a different standard ought to apply where mergers produce effects in more than one jurisdiction. Competition policy has followed ‘a national rather than an international agenda’,39 with decisions on mergers tending to be ‘based almost exclusively on the market effects within the 36 Although ideally agencies should be resourced and skilled enough to conduct thorough, sophisticated and timely merger analysis, this is not always the case and needs to be factored into any assessment of the appropriate law and policy to be applied. 37 See for example Mark A Glick and Donald Campbell, ‘Market Definition and Concentration: One Size Does Not Fit All’ (2007) 52 Antitrust Bulletin 229. 38 This is particularly relevant to assessment of dynamic efficiencies: See for example Gregory Sidak and Hal Singer, ‘Evaluating Market Power with TwoSided Demand and Pre-emptive Offers to Dissipate Monopoly Rent: Lessons for High-Technology Industries from the Antitrust Division’s Approval of the XM-Sirius Satellite Radio Merger’ (2008) 4 Journal of Competition Law and Economics 697. 39 Utton (n 24) 19.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

180 Comparative competition law

country concerned’.40 This is partly historical, in that the early development of competition laws preceded the globalization of markets, but it is also the case that international mergers have the ‘capacity to rouse strong nationalistic feelings’41 as evidenced by some of the infamous transAtlantic conflicts of the past decades. Pure nationalistic approaches to merger regulation need to be reviewed in light of increased globalization of markets.42 The present focus on national markets means that where efficiency gains are achieved in one country and anti-competitive effects felt in another, the latter is likely to block the merger (in whole or in part) while the former would provide clearance.43 The national competition response in such cases ‘will be in conflict with the optimal global policy’44 and the asymmetry of welfare effects is likely to disproportionately prejudice lesser developed countries.45 This conflict is likely to increase in the future.46 Consequently, while theoretically a ‘global’ welfare approach ought to be the preferred objective where multiple markets are affected by a single merger,47 this has the potential to produce adverse effects at a national level and there is little to suggest states would be prepared to subordinate their national interests to global ones. Nevertheless, improvements in global welfare outcomes can be achieved to a degree through active cooperation, which ensures that, to the extent it does not conflict with 40

Ibid. Ibid 87. 42 See generally Andrew Guzman, ‘Is International Antitrust Possible?’ (1998) 73 New York University Law Review 1501; Utton (n 24) ch 3; Budzinski (n 14) 3. 43 Utton (n 24) 78–79. 44 Ibid 94. 45 This is because of ‘asymmetric allocation of producers and consumers among the jurisdictions of the relevant international market’: Budzinski (n 14) 4. This asymmetry of effect also makes it more difficult to reach agreement on a way forward. 46 Utton (n 24) 73–74, 92–93. 47 See for example Noonan (n 3) 94–96: Joseph Wilson, ‘Globalization and the Limits of National Merger Control Laws: Gaps in Global Governance and the Need for an International Merger Control Regime’ (Doctor of Civil Law Thesis, McGill University 2002) 17. See also Utton (n 24) 78; Eleanor M Fox and Andreas F Lowenfeld, ‘Letters to the Editor: Boeing Affair’s Valuable Lessons’ Wall Street Journal (New York, 5 August 1997) A19; ‘… people are better off by maximizing world economic welfare rather than by maximizing their nation’s power vis-à-vis the power of other nations in the world’: Eleanor M Fox, ‘Competition Law and the Agenda for the WTO: Forging the Links of Competition and Trade’ (1995) 4 Pacific Rim Law & Policy Journal 1, 12. 41

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 181

national interest, the interests of other states and global consumers are taken into account. This is particularly important where merger remedies are proposed, to avoid unnecessary accumulation or conflict of remedies that might be generated by a purely inward looking focus. To a significant degree, jurisdictions do already consider the interest of others when applying remedies, or take account of remedies applied elsewhere when assessing whether a merger ought to be prohibited or subjected to further remedies. The expansion of merger jurisdictions and increased interconnection of global markets means that the need for such cooperation and consideration will increase in the future.

4. SUBSTANTIVE LAW AND ANALYSIS The legal test adopted by countries which have specific merger regimes fall broadly into two categories: (1) those which apply a substantial lessening of competition test (‘SLC’ test);48 and (2) those which adopt a market dominance test (‘dominance’ test). In some cases both tests are adopted. There is ‘no general consensus concerning the overall superiority of either’ of these tests,49 although the former has the benefit of focussing attention towards the core objective of protecting competition. It has also become more common since 2004 when the European Union (EU) moved from a dominance to a competition-based test.50 A public interest test is also sometimes invoked, though normally as an integral part of one of the primary tests or as a means by which an otherwise anti-competitive or dominance-generating merger might be allowed to proceed.51 48

For example, the US, the EU and Australia all adopt a competition-based test. In the EU and many other European countries, the ‘competition’-based test takes the form of a ‘significant impediment to effective competition’ (SIEC) test. This will be discussed further, below, but for purposes of broad classification it will be considered together with the SLC tests. 49 OECD, ‘Substantive Criteria Used in the Assessment of Mergers’ (DAFFE/COMP(2003)5, Best Practice Roundtable on Competition Policy 2003) 7. 50 See generally OECD, ‘Policy Roundtables: The Standard for Merger Review, with a Particular Emphasis on Country Experience with the change of Merger Review Standard from the Dominance Test to the SLC/SIEC Test’ (DAF/COMP(2009)21, Roundtable No 102, OECD Policy Round Tables 2009). 51 No OECD country adopts a ‘pure’ public interest test. In the United Kingdom, the public interest test, which had operated since 1965, was abolished

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

182 Comparative competition law

The tests adopted may reflect different objectives or, more commonly, arise because of the history, different economic size, industrial advancement or geographic location of the country, including the desire for regional consistency.52 The same test is normally applied regardless of whether the merger is national or transnational in scope, although the practical application of the law may vary where international factors are involved. Statistics suggest that there is generally little difference in the outcome of merger reviews conducted in multiple jurisdictions that apply different tests.53 In this respect, given the largely administrative nature of the merger review process in most countries, the precise wording of the prohibition is far less important than the analytical guides and processes used to assess the merger’s likely market impact.

5. ANALYTICAL APPROACH Despite differences in policy goals or language adopted in relation to substantive merger legislation, a cursory review of merger guidelines produced by many of the competition agencies responsible for their enforcement demonstrates that national competition agencies are increasingly adopting a similar analytical approach, at least on paper, to guide their assessment of merger-generated competitive effects.54 This has been by the Enterprise Act 2002 and replaced with a competition test. Most jurisdictions allow some sort of interference on public policy grounds, although some are more expansive than others in this respect. 52 See for example Michal S Gal, Competition Policy for Small Market Economies (Harvard University Press 2003) 200–201. Compare Richard Whish, ‘Substantive Analysis under the EC Merger Regulation: Should the Dominance Test Be Replaced by “Substantial Lessening of Competition”’ in J William Rowley (ed), International Merger Control: Prescriptions for Convergence (International Bar Association 2001) 102. 53 OECD, ‘Substantive Criteria Used in the Assessment of Mergers’ (DAFFE/COMP(2003)5, Best Practice Roundtable on Competition Policy 2003) 173. See also SFE Corporation Limited, ‘Submission to the Review of the Competition Provisions of the Trade Practices Act 1974’ (Public Submission 92, Trade Practices Act Review 2002); Warren Pengilley, ‘Submission to the Review of the Competition Provisions of the Trade Practices Act 1974’ (Public Submission 8, Trade Practices Act Review 2002). 54 See, for example, recent new and revised guidelines including Australia (2008), Canada (2011), Chile (2012), France (2009), Germany (2012), Indonesia (2010), Japan (2010), United Kingdom (2010), US (2010). In relation to competition law generally, Stuer observes that ‘there is more convergence today

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 12 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 183

aided by the work of the ICN,55 which has recently developed a set of ‘Recommended Practices for Merger Analysis’.56 These recommendations provide a legal framework for substantive merger analysis, including identification of theories of harm and recommendations on the use of market shares and entry and expansion considerations. Convergence has also been aided by the fact that many new competition law regimes have been modelled on more established ones57 and that a large number of regimes have reviewed their guidelines in recent years and, in so doing, have paid regard to emerging international best practice, including ICN recommendations on the approaches adopted by other agencies. The increasing similarity of merger guidelines outlining competition agencies’ approach to merger analysis has, therefore, naturally converged over the last decade and is likely to continue to do so as agencies refresh their guidelines over time.58 than seemed possible not long ago. Virtually every competition regime limits anti-competitive bullying and ganging up, and while the language and procedures vary enormously, the direction is largely the same’: Richard M Stuer, ‘The Simplicity of Antitrust Law’ (2012) 14(2) University of Pennsylvania Journal of Business Law 543, 551 . See also Larry Fullerton and Megan Alvarez, ‘Convergence in International Merger Control’ (2012) 26(2) Antitrust 20, 20–22. See also Stephen Calkins, ‘The Global Conversation about Merger Review and Powerful Buyers’ (Third Annual Baxt Lecture, Melbourne Law School, 2012). 55 This includes not only recommendations and best practice work product, but its role in facilitating and promoting more communication and collaboration between its members. See, for example, Galloway discussing the influence of the ICN recommendations on merger control laws in India and China: Jonathan Galloway, ‘Convergence in International Merger Control’ (2009) 5(2) The Competition Law Review 179. 56 ICN, ‘Recommended Practices for Merger Analysis’ (Merger Working Group 2008, amended 2009 and 2010). The core areas covered are The Legal Framework for Competition Merger Analysis (2008), Use of Market Shares: Thresholds & Presumptions (2008), Entry and Expansion (2008), Competitive Effects Analysis in Horizontal Merger Review: Overview (2009), Unilateral Effects (2009) and Coordinated Effects (2009). See also Christine A Varney, ‘Our Progress Towards International Convergence’ (Paper presented at the 36th Annual Fordham Competition Law Institute Conference on International Antitrust and Policy, New York, 24 September 2009) 5. 57 New regimes have usually modelled themselves on either the European or US models: Richard M Stuer, ‘The Simplicity of Antitrust Law’ (2012) 14(2) University of Pennsylvania Journal of Business Law 543, 553. 58 The ICN has recommended the periodic review of merger control provisions and consideration of reforms promoting ‘convergence towards recognized best practices’: ICN, ‘Recommended Practices for Merger Notification

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

184 Comparative competition law

Key areas of consideration in analytical approach include the relevance of one of the firms being a ‘failing firm’ or, conversely, one of the firms being a ‘maverick’,59 the role of efficiencies in merger analysis and the issue of market definition. 5.1 Theories of Harm Most jurisdictions focus on the potential for mergers to have adverse unilateral60 or coordinated effects61 and many guidelines adopt a similar approach to these theories of harm,62 which are increasingly informed by sophisticated economic modelling.63 There is also a broad consensus that Procedures’ (recommendation XIII(A) and (B) (originally recommendation VII, then recommendation XI), Merger Working Group 2002). See also Calkins, highlighting similarities in terminology adopted in recent agency guidelines: Calkins (n 54). 59 This is not intended to be an exhaustive list, but highlights key areas of convergence and divergence. Other areas include, for example, the relevance of buyer power in merger analysis: Calkins (n 54). 60 ICN Merger Working Group: Analytical Framework Subgroup, ‘Project on merger guidelines’ (2004) Report for the third ICN annual conference in Seoul April 2004, Ch 1, para 26 . Although the core approach is similar where a merger leads to monopoly, there is greater controversy when the merger combines smaller firms into an oligopoly (para 29). Merger guidelines in Australia, Brazil, Canada, Finland, France, Ireland, Japan, New Zealand, the UK and US, among others, all adopt the common terminology of ‘unilateral effect’. In addition the EC guidelines refer to ‘non-coordination’ and the German guidelines refer to dominance (Germany retains a dominance, rather than competition test for mergers): Calkins (n 54). 61 ICN Merger Working Group: Analytical Framework Subgroup, ‘Project on merger guidelines’ (2004) Report for the third ICN annual conference in Seoul April 2004, Ch 1, para 43 . 62 See for example Calkins (n 54). See also Deborah L Feinstein, ‘Process Divergence as an Obstacle to Substance Convergence?’ (2012) 26(3) Antitrust 5, 5. 63 The increase in economic evidence was particularly notable in the European Union following the modification to the Merger Regulation in 2004. See further Sokol and Blumenthal (n 20) 327–328. Note this can have positive and negative effects. It has been suggested that over-reliance on economic evidence has ‘raised the bar’ for authorities wishing to demonstrate that a merger will result in anti-competitive consequences and therefore reduced, below an optimal level, the effective enforcement of merger laws: see, in particular, Oliver Budzinski, ‘An Institutional Analysis of the Enforcement Problems in Merger

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 185

market shares and concentration, while relevant to a competition assessment, should not be considered conclusive. However, differences remain as to what is necessary in order to establish harm pursuant to these theories, particularly in relation to vertical and conglomerate mergers64 and there is also debate regarding the extent to which these assessments should depend on economic concepts and models, particularly where full and accurate market information may not be available.65 The weight attached to pure economic evidence by the judiciary also varies, with courts in some jurisdictions embracing and even demanding rigorous economic evidence to support theories of harm and others downplaying the value of economic testimony.66 These differences are likely to continue and reflect both judicial and agency expertise and confidence in economic modelling. 5.2 Failing Firms and Mavericks Many jurisdictions recognize a ‘failing firm’ defence,67 either formally or informally, and the concept appears frequently in merger guidelines. It is generally difficult to establish, normally requiring proof that the ‘failing firm’ would exit the market absent the merger, that its assets would also

Control’ (2010) 6(2) European Competition Journal 445, esp from 460. See also Mat Hughes and David Wirth, ‘A Tale of Three Mergers: The Use of Quantitative Techniques in UK and EU Merger Control’ in Nigel Parr and Ruth Sander (eds), Merger Control 2013 (International Comparative Legal Guides 2013) ; OECD, ‘OECD Policy Roundtables: Economic Evidence in Merger Analysis’ (DAF/COMP(2011)23, OECD 2012). 64 See for example Sokol and Blumenthal (n 20) 331–332. 65 See for example Lars Sørgard, ‘Merger Screening in Markets with Differentiated Products’ in Konkurrensverket – Swedish Competition Authority (ed), More Pros & Cons of Merger Control 2012 (Konkurrensverket – Swedish Competition Authority 2012). 66 Courts in the US and Europe increasingly demand extensive economic evidence for competitive effects (see, Oliver Budzinski, ‘An Institutional Analysis of the Enforcement Problems in Merger Control’ (2010) 6(2) European Competition Journal 445, 460ff). Conversely, Australian courts have taken a more cautious approach to economic evidence in recent years. See for example ACCC v Metcash Trading Limited [2011] FCA 967 (25 August 2011); [2001] FCAFC 151 (30 November 2011). 67 See generally OECD, ‘Policy Roundtables: Failing Firm Defence’ (DAF/ COMP(2009)38, Roundtable No 103, OECD 2009).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

186 Comparative competition law

exit the market and that there was no less anti-competitive alternative than the proposed merger. For example, in the United States (US), the new Horizontal Merger Guidelines maintain a specific section on failing and exiting firms68 which provides that failing firm claims will generally not be considered unless it can be demonstrated that the firm would be unable to meet financial obligations in the near future, would not be able to successfully reorganize under the Bankruptcy Act 1898 and has made ‘unsuccessful good-faith efforts to elicit reasonable alternative offers that would keep its tangible and intangible assets in the relevant market and pose a less severe danger to competition than does the proposed merger’.69 Similar criteria exist in other jurisdictions. For example, the European Commission (EC) guidelines refer to the ‘basic requirement … that the deterioration of the competitive structure that follows the merger cannot be said to be caused by the merger [and] where the competitive structure of the market would deteriorate to at least the same extent in the absence of the merger’.70 In Australia, agency guidelines make clear that mere speculation of exit will not be sufficient, that there must be imminent danger of failure without the merger and that, absent the merger, assets would leave the industry and that competition remaining after the merger must not be substantially less than it would have been had the firm merely exited the market.71 Convergence surrounding the treatment of failing firms demonstrates both recognition of the different effect such acquisitions might have on the future state of competition, while making clear that the mere existence of a failing firm should not be sufficient to ensure clearance of a merger if there are less anti-competitive alternatives, which may even include allowing the firm to fail. 68

US Horizontal Guidelines (n 4) §11.0. Ibid. 70 The Commission considers three criteria particularly relevant: (1) ‘the allegedly failing firm would in the near future be forced out of the market because of financial difficulties if not taken over by another undertaking’, (2) ‘there is no less anti-competitive alternative purchase than the notified merger’ and (3) ‘in the absence of a merger, the assets of the failing firm would inevitably exit the market’. This is almost identical to the formulation of the defence in the US: Council Regulation (EC) XX/2004 Guidelines on the Assessment of Horizontal Mergers under the Council Regulation on the Control of Concentrations between Undertakings of 5 February 2004 [2004] OJ C 31, 5-18 paras 89–90. 71 ACCC, Merger Guidelines (November 2008) 14, paras 3.22–3.23. 69

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 187

Conversely, many jurisdictions now also make explicit reference to ‘maverick’ firms,72 those firms that play a ‘disruptive role in the market to the benefit of customers’.73 Guidelines generally now explicitly identify the elimination of a ‘maverick’ firm as being likely to raise more competition concerns than elimination of a ‘normal’ competitor.74 5.3 Market Definition There is a high level of consistency in merger guidelines relating to the methodology for determining market definition,75 with agencies generally adopting the hypothetical monopolist test – or SSNIP76 test77 – to determine the boundaries of the market. Beyond this, however, different approaches are taken by the agencies in relation to the techniques adopted to apply these tests.78 72 According to a survey of recent guidelines conducted by Stephen Calkins, the word ‘maverick’ is now used in Australia, Canada, the European Community, Finland, France, Indonesia, Ireland, New Zealand, Tanzania, the UK and the US: Calkins (n 54) slide 20. 73 US Horizontal Merger Guidelines (n 4) §2.1. 74 Ibid §7.1. In Australia, legislation expressly requires the courts to consider, as part of its competition assessment, whether the merger would remove a ‘vigorous and effective competitor’, which the guidelines equate to a ‘maverick’ firm: Competition and Consumer Act 2010 (Cth) s 50(3)(h); ACCC, Merger Guidelines (November 2008) 49, para 7.56. 75 Compare Joseph Farrell and Carl Shapiro, ‘Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition’ (2008) Working Paper, 25 November 2008 . 76 Small but significant and non-transitory increase in price. 77 Utton observes that there is ‘general agreement that the policymakers are using the correct analytical approach’ and that is the hypothetical monopolist approach, first applied to determine product and then geographic area: Utton (n 24) 74; Reference to SSNIP is included in, among others, the guidelines in Australia, Brazil, Canada, the European Community, Finland, France, Ireland, Japan, New Zealand, Tanzania, the UK and the US, according to Calkins (referring to a review of 13 merger guidelines which were introduced or revised between 2001 and 2012), 12 of which (the exception being Germany) refer to SSNIP. The HHI index is referred to in 10 of those guidelines, including Germany, but excluding Brazil, New Zealand and Tanzania. All of the surveyed jurisdictions (with the exception of Tanzania) refer to ‘close substitutes’: Calkins (n 54). 78 See Hughes and Wirth discussing the varied use of qualitative techniques in select merger cases: Hughes and Wirth (n 63). See also OECD, ‘OECD Policy Roundtables: Market Definition’ (DAF/COMP(2012)24, OECD 2012); Michael

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 17 /

Date: 13/10

JOBNAME: Duns PAGE: 18 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

188 Comparative competition law

There is also little real controversy surrounding the proposition that the breadth of the market defined should accurately represent its full economic breadth and should not be artificially restricted by national boundaries. Any such restriction will distort the analysis of competitive effects and will advance neither national nor global welfare.79 Although this is well recognized by competition authorities, in some cases legislation mandates that markets be defined within national boundaries.80 In these cases the agencies conducting the analysis often formulate less optimal or informal mechanisms for overcoming this distortion in order to produce accurate competition analysis. The clear preference is for the legislation itself to provide appropriate recognition of the market as an economic and not national construct, with national considerations taken into account when assessing local effects. One aspect of market definition about which debate has recently been reignited relates to its precise role in merger review.81 In particular, debate surrounds whether or not market definition should be used as a threshold step in merger analysis or whether it should be more integrated into the competition analysis. This debate has largely been triggered by the release of the new US Merger Guidelines in 2010 which provide that the agencies’ competition analysis ‘need not start with market definition’.82

Katz and Carl Shapiro, ‘Critical Loss: Let’s Tell the Whole Story’ (Spring 2003) Antitrust 49; Daniel O’Brien and Abraham Wickelgren, ‘A Critical Analysis of Critical Loss Analysis’ (2003) 71 Antitrust Law Journal 161; Øystein Daljord, Lars Sørgard and Øyvind Thomassen, ‘The SSNIP Test and Market Definition with the Aggregate Diversion Ratio: A Reply to Katz and Shapiro’ (2008) 4(2) Journal of Competition Law and Economics 263. 79 See for example Utton (n 24) 74. See also Cam Truong, ‘Transnational Commerce and the Problem of Market Definition’ (2010) 18 Trade Practices Law Journal 254. 80 For example, Australia defines market as a market within Australia (or a region of Australia): Competition and Consumer Law Act 2010 (Cth) s 50(6). 81 See generally Rhonda Smith, ‘Market Definition: Going, Going, Gone? Developments in the United States’ (2010) 18 Competition and Consumer Law Journal 119. 82 US Horizontal Merger Guidelines (n 4) 7; see for example Sokol and Blumenthal noting the concerns around relying too heavily on market definition at the early stages of merger review: Sokol and Blumenthal (n 20) 331. See also Smith (n 81).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 18 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 189

5.4 The Role of Efficiencies Another key area of debate surrounds the role efficiencies should play in merger analysis.83 The potential for merger-generated efficiencies is well recognized.84 Efficiency gains have the potential to mitigate against use of increased market power in some cases and to deliver resource savings. Where a merger produces dynamic efficiencies, consumers may benefit from the development of new or better products and producers might achieve productive efficiencies through technological advances in production techniques. Despite recognition of efficiency benefits that might be achieved as a result of a merger, the role these predicted efficiency gains play in merger review varies. Efficiencies may be considered a factor for consideration as part of a competition analysis or as a defence,85 or both, and the value attached to efficiency claims in that analysis varies considerably. Recognizing the lack of consensus on this point, the ICN has observed that ‘no one modality for the treatment of merger efficiencies is necessarily correct or appropriate for all countries’, but that efficiencies should be

83

See for example Herbert J Hovenkamp, ‘Merger Policy and the 2010 Merger Guidelines’ (2010) University of Iowa Legal Studies Research Paper No 10-34 (Revised) December 2010, 15. See also Pál Szilágyi, ‘How to Give a Meaningful Interpretation to the Efficiency Defence in European Competition Law?’ (Competition Law Research Centre Péter Pázmány Catholic University 2011) . 84 The EU was initially hostile towards efficiencies, but has given them more attention in recent years: see for example Sokol and Blumenthal (n 20) 332. See also Mitja Kocmut, ‘The Role of Efficiency Considerations Under the EU Merger Control’ ((L) 05/05, The University of Oxford Centre for Competition Law and Policy, 2005). 85 This is overtly the case in Canada, where s 96(1) of the Competition Act 1985 provides that the Tribunal shall not prohibit a merger where the proposed merger would ‘bring about gains in efficiency that will be greater than, and will offset, the effects of any prevention or lessening of competition …’, provided those efficiencies would not be attained absent the merger. See also generally An Renckens, ‘Welfare Standards, Substantive Tests, and Efficiency Considerations in Merger Policy: Defining the Efficiency Defense’ (2007) 3 Journal of Competition Law and Economics 149; Lin Bian and DG McFedtridge, ‘The Efficiencies Defence in Merger Cases: Implications of Alternative Standards’ (2000) 33 Canadian Journal of Economics 297. Compare the position in the US: FTC v Procter & Gamble Co 383 US 586, 580 (1967).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

190 Comparative competition law

taken into consideration in merger analysis because of the role mergers can play in promoting economic growth and development.86 Incorporating efficiencies into merger analysis is extraordinarily difficult87 because gauging the value of predicted efficiencies is and will remain a very difficult task. While the capacity for mergers to generate efficiency gains is acknowledged, it is also recognized that mergers may ‘have adverse consequences for the pace of innovation’88 reduce allocative efficiency, with some empirical studies suggesting that it is ‘unlikely that most mergers enhance efficiency’.89 Difficulties of proof and scepticism surrounding self-serving efficiency predictions means that in most jurisdictions, regardless of the precise role of efficiencies in the analytical process, it is difficult for parties to an otherwise anti-competitive merger to demonstrate efficiency gains to a level necessary to alleviate or overcome competition concerns. This is particularly so in those countries, including the US, the UK90 and the EU, which require that efficiencies be merger-specific and ‘verifiable’ and produce benefits for consumers.91 As a result, while the theoretical benefits of efficiencies are increasingly recognized, the challenges of proof mean that it remains rare that

86 ICN, ‘ICN Merger Guidelines Workbook’ (Prepared for the Fifth Annual ICN Conference in Cape Town by the Merger Working Group: Investigation and Analysis Subgroup 2006) 1. 87 See ‘Making a prospective determination about whether a merger will lead to static efficiencies and how such efficiencies measure up against any anti-competitive effects that the merger is expected to cause can be very challenging. Dynamic efficiencies pose an even greater measurement problem than static efficiencies because dynamic effects will occur – if at all – over several time periods and may be more abstract in nature than static effects.’: OECD ‘Policy Roundtables: Dynamic Efficiencies in Merger Analysis’ (DAF/ COMP(2007)41, Roundtable No 77 OECD 2007) 9. 88 Richard Gilbert and Steven Sunshine, ‘Incorporating Dynamic Efficiency Concerns in Merger Analysis: The Use of Innovation Markets’ (1995) 63 Antitrust Law Journal 569, 574. 89 OECD ‘Policy Roundtables: Dynamic Efficiencies in Merger Analysis’ (DAF/COMP(2007)41, Roundtable No 77, OECD 2007) 11. 90 Renckens (n 85) 164. 91 This suggests a focus on consumer rather than producer welfare: Utton (n 24) 76–77. See also OECD ‘Policy Roundtables: Dynamic Efficiencies in Merger Analysis’ (DAF/COMP(2007)41, Roundtable No 77 OECD 2007) 1; Renckens (n 85) 162, 168. Belgium, Finland, France, Germany, Hungary, Iceland, Spain and Switzerland all require efficiencies to benefit consumers before they will be considered as part of the competition analysis.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 20 /

Date: 13/10

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 191

an otherwise anti-competitive merger will be justified on efficiency grounds.92

6. ADMINISTRATIVE TREATMENT OF MERGERS Merger law will be most effective at deterring and preventing anticompetitive mergers where appropriate procedural mechanisms are in place to detect, prevent or unravel mergers which contravene the substantive laws.93 Procedural regulation of mergers should therefore aim to facilitate the identification and prevention of mergers which might contravene substantive law94 while keeping to a minimum the cost to parties and regulators. The extraterritorial reach of merger laws in most jurisdictions means that regardless of where the merger takes place, the location of the merging parties, likely competitive impact, or the procedural requirements of all jurisdictions potentially affected by the merger, need to be considered by firms proposing to merge.95 It is also in the area of administrative structures and procedures that there exists the most significant divergence in relation to merger regulation.96 This divergence has been highlighted by the introduction of several relatively new but

92 See for example Renckens observing that in Europe no cases have yet been cleared purely on efficiency grounds. Similarly Renckens observes that in the US, in most cases ‘where efficiencies were considered important, they most often have been found insufficient to counterbalance the negative effects of the merger’: Renckens (n 85) 161. 93 Sound merger policy can also deter the negotiation of anti-competitive mergers. See generally Seldeslachts, Clougherty and Barros, who conclude that blocking mergers is effective in deterring future mergers (although negotiated settlements are ineffective in deterring such mergers): Jo Seldeslachts, Joseph A Clougherty and Pedro Pita Barros, ‘Settle for Now but Block for Tomorrow: The Deterrence Effects of Merger Policy Tools’ (2009) 52 Journal of Law and Economics 607, 631. 94 See for example Frederick G Hilmer, Mark Rayner and Geoffrey Taperell, National Competition Policy (Report by the Independent Committee of Inquiry, Commonwealth of Australia 1993) (‘Hilmer Report’) 83. 95 See for example Gal (n 52) 199. 96 Deborah L Feinstein, ‘Process Divergence as an Obstacle to Substance Convergence?’ (2012) 26(3) Antitrust 5.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

192 Comparative competition law

strategically important merger regimes, particularly in the BRIC countries,97 which all now adopt mandatory and suspensory PMN (pre-merger notification) regimes.98 Key issues relevant to administrative treatment of mergers are the structure and powers of the agencies involved (including the role of the judiciary), notification obligations (including criteria, form and information required), process and time frame for assessment, negotiation of remedies and transparency of decision-making.99

7. ADMINISTRATIVE STRUCTURE AND AGENCY POWERS As a result of the predominance of pre-merger review systems, knowledge of the nature of the administrative and judicial structures governing competition law enforcement is important to understanding the processes by which a merger may be permitted or blocked or a decision reviewed. A distinction must be drawn between jurisdictions, like the EU, in which the enforcement agency has the power to block a merger (subject to judicial appeal) and, more commonly, those jurisdictions in which competition agencies only have the power to challenge a merger before the courts who hold the ultimate decision-making power. In some cases other public authorities may also be given an influencing role or there might be separate structures in place relevant to specific sectors of the economy, such as the banking or telecommunication industries. In addition to different processes for contesting mergers, variation also exists in the nature and effect of agency clearance. In some cases agency ‘clearance’ of a merger renders the merger free from further challenge. In other cases third parties may still possess the right to bring private action challenging the proposed merger. 97 ‘The BRIC (Brazil-Russia-India-China) cluster has been a major magnet as it represents about 25% of global GDP and over 60% of global growth’: Karel Kool, Nicolas Harlé and Philippe Ombregt, ‘Merger Control and Practice in the BRIC Countries vs. the EU and the US: Review Thresholds’ (Insead Blog, 28 August 2012) . 98 See further Emily Roche, ‘BRIC Merger Control – The New Regulatory Frontier’ (2012) 5(19) International In-house Counsel Journal 1. 99 Deborah L Feinstein, ‘Process Divergence as an Obstacle to Substance Convergence?’ (2012) 26(3) Antitrust 5, 5–7.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 193

Agencies that have the ability to block a merger without resort to the courts have a clear advantage and, as a result, are likely to have more bargaining power when it comes to negotiating remedies, although this may be mitigated to a degree by the prospect of judicial review. The role of the courts in the process also affects the nature of information gathering that is required or appropriate. In judicial systems agencies have sometimes been criticized for gathering excessive information to aid them in the event of future litigation. Agencies also differ significantly in relation to their resources and expertise. This has led to concerns that some of the newer merger regimes have been established without the necessary resources, expertise or public support to ensure effective enforcement. Thus, for example, it has been suggested that the Administrative Council for Economic Defense (CADE), responsible for merger assessment in Brazil, lacks capacity to deal with the new regime effectively100 and that the Ministry of Commerce (MOFCOM) in China is still inadequately resourced to handle the volume of merger notifications now falling within its ambit,101 with the result that some larger deals are likely to be delayed. The Director General of the Antimonopoly Bureau of MOFCOM, Shang Ming, also recently acknowledged the ‘weak’ competition culture in China, which MOFCOM is seeking to correct through press conferences and related endeavours.102 All of these factors are likely to impact on the effectiveness of agencies in producing efficient and predictable decisions in merger cases.

8. NOTIFICATION OBLIGATIONS The difficulty in correcting the structural damage to the market caused by anti-competitive mergers makes detection and prevention in their incipiency particularly desirable. It is therefore not surprising that PMN regimes have emerged to support the substantive merger laws. While the type of PMN is dominated by mandatory, suspensory systems, there remains significant divergence with respect to notification 100 See generally Knox also observing that CADE is ‘one of the world’s most swiftly developing competition authorities’: Knox (n 7) 16, 17. 101 The number of mergers notified in China rose from 16 in 2008 to 171 in 2011: Shang Ming, ‘Ministry of Commerce Press Conference of the 2012 Antitrust Work Progress’ (The State Council The People’s Republic of China 27 December 2012) . 102 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

194 Comparative competition law

thresholds, timing of notification and review and information requirements for parties required to notify. It is in this area that the ICN directed its initial work on mergers, making early recommendations as to jurisdictional nexus,103 thresholds,104 notification timing,105 review periods106 and information requirements for initial notification.107 While these have assisted in promoting convergence,108 particularly for new regimes and jurisdictions which have reviewed their guidelines in recent years, the recommendations themselves are sufficiently vague that it is possible for countries to adhere while still applying significantly different standards109 with the result that significant areas of divergence remain. 8.1 Type of Regime PMN regimes generally take one of two forms; a mandated formal clearance process in which the consummation of mergers is prohibited without agency clearance (suspensory regime) or a voluntary system in which parties are free to merge without clearance, but remain subject to substantive laws. In voluntary systems parties are often given the option of applying for a formal or informal opinion about the legality of their merger prior to consummation110 in order to improve certainty about the prospect, or otherwise, of the merger being challenged in the courts. All competition authorities, whether imposing mandatory PMN requirements or not, recognize the benefits of detecting potentially anti-competitive mergers at an early stage so as to prevent the economic 103

ICN, ‘Recommended Practices for Merger Notification Procedures’ (Merger Working Group 2002, amended 2003, 2004, 2005, 2006), recommendation I. 104 Ibid, recommendation II. 105 Ibid, recommendation III. 106 Ibid, recommendation IV. 107 Ibid, recommendation V; See further Julie Clarke, ‘Multi-jurisdictional Merger Review Procedures: A Better Way’ (2006) 14 Trade Practices Law Journal 90–109. 108 Eleanor M Fox and Merit E Janow, ‘A Report of the Second Annual Conference of the International Competition Network’ (International Competition Network 2003) 33–34 . See also ICN, ‘Implementation of the ICN Recommended Practices for Merger Notification and Review Procedures’ (ICN April 2005) Annex B, 2 . 109 See further, Fox and Janow (n 108) 33. 110 There are variations on these approaches, such as post-merger notification and notification requirements that are non-suspensory, but mandatory suspensory regimes clearly dominate.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 24 /

Date: 10/9

JOBNAME: Duns PAGE: 25 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 195

and social harm from structural change that a merger might otherwise bring about. Although the view that post-consummation remedies are always, or almost always, unsatisfactory is not unanimously held,111 it remains the dominant view and enjoys significant evidentiary support as well as the benefit of being intuitively correct. Following the integration of corporate assets and structure, the task of effectively unravelling these changes, commonly analogized as ‘unscrambling the eggs’, will be difficult and disruptive to those involved, even if it can effectively put an end to the anti-competitive harm occasioned by the merger.112 Mandatory PMN was first formally adopted in the US in 1976 at a time when mergers were predominantly local in nature. There is no doubt that, since that time, PMNs have had a profound effect on the enforcement of merger laws.113 In recent decades both the volume and range of transnational mergers have dramatically increased and the number of regimes imposing such requirements (currently in excess of 70)114 continues to expand. As a consequence, the administrative time burdens on business and authorities have burgeoned, arguably beyond the optimal level to achieve the initial goals. The key alternative to a mandatory regime is a voluntary one.115 While rare, voluntary PMN regimes share the same broad objectives as mandatory notification and can achieve similar results.116 Empirical evidence 111

See for example Jonathan Green and Gianandrea Staffiero, ‘Economics of Merger Control’ in ‘The 2007 Handbook of Competition Economics: Global Competition Review Special Report’ (Global Competition Review 2007) 8 and J William Rowley, Omar K Wakil and A Neil Campbell, ‘Streamlining International Merger Control’ (Paper presented at the EC Merger Control 10th Anniversary Conference, Brussels, 14 September 2000) 8. 112 See generally Richard Burnley, ‘An Appropriate Jurisdictional Trigger for the EC Merger Regulation and the Question of Decentralisation’ (2002) 25 World Competition 263, 276. 113 See generally William J Baer, ‘Twenty Years of Hart-Scott-Rodino Merger Enforcement: Reflections on Twenty Years of Merger Enforcement Under the Hart-Scott-Rodino Act’ (1997) 65 Antitrust Law Journal 825. 114 See Choe Chongwoo and Chander Shekhar, ‘Compulsory or Voluntary Pre-merger Notification? Theory and Some Evidence’ (Working Paper No 13450, MPRA 2009) 1; ICN, ‘Merger Notification Filing Fees’ (Mergers Working Group, 2005) 2; Noonan (n 3) 62. 115 There are other alternatives, including post-merger notification, which was the approach previously adopted in Brazil, but these are now clearly in the minority. 116 As there are very few voluntary regimes (including the UK, Australia and New Zealand) most empirical research into the efficiency of notification regimes presupposes that the regime will be mandatory. As a result, ‘… there is a paucity

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

196 Comparative competition law

suggests that, provided the incentives for notifying potentially anticompetitive mergers are sufficiently high,117 a voluntary notification regime can result in most problematic mergers being notified, or otherwise detected, without imposing onerous obligations on those mergers that do not raise anti-competitive concerns.118 Voluntary regimes are often considered more suitable for smaller economies119 where the detection of mergers may be less problematic and where significant global mergers which may have national impact are likely to be subjected to scrutiny by other agencies. Nevertheless, a number of smaller economies have opted for the more common mandatory regime. This has attracted criticism in some quarters by subjecting multi-billion-dollar global transactions to additional regulatory requirements in circumstances where there is only a peripheral connection to the reviewing country.120 There is, however, no obvious ‘optimal’ approach, given the heterogeneous nature of countries adopting merger laws. As the number of pre-merger regimes continues to burgeon and markets continue on their of theoretical studies that analyze the optimal merger notification’: Chongwoo and Shekhar (n 114) 5. See also Michal S Gal, ‘Merger Policy for Small and Micro Jurisdictions’ in Konkurrensverket – Swedish Competition Authority (ed), More Pros & Cons of Merger Control 2012 (Konkurrensverket – Swedish Competition Authority 2012) 61. 117 For example, where the regulator has ‘developed a strong reputation for imposing heavy costs on parties that fail to notify such mergers’: Chander Shekhar and Philip Williams, ‘Should the Pre-Notification of Mergers Be Compulsory in Australia?’ (2004) 37 The Australian Economic Review 382, 383–384. 118 In this respect, empirical evidence suggests that mergers notified pursuant to a voluntary regime are more likely to be objected to by regulators, supporting the view that ‘the voluntary notification system allows the parties to self-sort the potentially problematic proposals’: Shekhar and Williams (n 117) 385, 387. 119 For example, both Australia and New Zealand adopt voluntary regimes. Although the UK adopts a voluntary regime, this must be considered in the context of the EU, which imposes mandatory notification obligations where thresholds relating to the internal market are exceeded. See further Gal (n 116). 120 This has been the case in Jersey and Guernsey. Jersey has had a mandatory merger regulation for several years and Guernsey introduced a mandatory regime in 2012. This was despite recommendations that a voluntary regime would be more appropriate, given the size of the economy, and appears to have been brought about by a desire for consistency with Jersey (with both now falling within the ambit of a single regulator, the Channel Islands Competition and Regulatory Authorities), rather than because of a genuine belief that a mandatory regime was necessary.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 27 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 197

unabated trend towards globalization, the issue as to whether mandatory PMN is appropriate for all jurisdictions or whether greater use should be put to the voluntary option, particularly for smaller economies, is likely to re-emerge as a significant policy issue. 8.2 Thresholds for Notification While justifiable conceptually, the appropriateness and effectiveness of PMN will rest, to a significant degree, on the thresholds adopted to trigger notification obligations. In this respect, the ICN has recommended that notification thresholds be clear and understandable,121 based on objectively quantifiable criteria122 and based on information that is readily accessible to the merging parties.123 Most guidelines now satisfy the second of these recommendations and there is increasing adherence to the other two, although work needs to be done to improve clarity and reduce information burdens in some cases.124 The majority of PMN regimes now incorporate some form of local market nexus, often coupled with a worldwide turnover figure, as a threshold requirement for notification. Recognizing the importance of objective criteria, suspensory jurisdictions now generally impose a turnover and/or test to determine nexus and notifiability, rather than relying on market share or concentration levels. However, the preference for objective thresholds, while desirable for business certainty, means that the thresholds set pay little or no regard to the specific circumstances of the transaction;125 whether notification is required will therefore depend entirely on the size of the transaction rather than its potential effects on 121

ICN, ‘Recommended Practices for Merger Notification Procedures’ (Merger Working Group 2002, amended 2003, 2004, 2005, 2006), recommendation II(A). The Working Group comments that an essential feature of notification thresholds should be ‘clarity and simplicity’: Working Group comment 1 to recommendation II(A). 122 Ibid, recommendation II(B). 123 Ibid, recommendation II(C). 124 See, for example, Shroff and Uberoi, noting that the Competition Commission of India (CCI) ‘has considerable ground to cover in relation to bringing about clarity in the law’, including in relation to ‘determination of turnover under the Act’: Cyril Shroff and Nisha Kaur Uberoi, ‘The Future of Competition Law in India: Reading The Portents’ (2013) 1(2) CPI Antitrust Chronicle 3. 125 See for example Malcolm R Pfunder, ‘Twenty Years of Hart-Scott-Rodino Merger Enforcement: Some Reflections on, and Modest Proposals for Reform of, the Hart-Scott-Rodino Premerger Notification Program’ (1997) 65 Antitrust Law Journal 905, 905. See also Shekhar and Williams (n 117) 385.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 27 /

Date: 13/10

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

198 Comparative competition law

competition. While there will, in many cases, be a correlation between size and potential effect on competition, this is not invariably the case. Consequently, these crude thresholds will inevitably be both underinclusive and over-inclusive.126 The specific threshold levels vary considerably and in many cases are frequently reviewed. When setting thresholds for notification there is a clear preference among law-makers and regulators for over-inclusiveness, evidenced by the fact that only a very small proportion of mergers notified under PMN, typically less than 5 per cent,127 raise any significant concerns for regulators.128 This is the case despite the fact that many jurisdictions have reviewed their thresholds in recent years.129 One of the lowest thresholds is Russia’s,130 with the result that in 2011 Russia reviewed significantly more mergers than any other agency (including the US DOJ (Department of Justice) and FTC (Federal Trade Commission) combined), both on an absolute and per capita basis.131

126 They will be under-inclusive because they will ignore small but competitively significant transactions, although these might still be the subject of agency review if the agency becomes independently aware of the merger. In the US more than one-third of all mergers challenged in the courts in the 2008 fiscal year fell below notification thresholds: Timothy P Daniel, ‘Whither Merger Review? Looking Forward While Looking Back’ (2009) 8(6) The Antitrust Source 1, 3. See also Federal Trade Commission, noting that the ‘FTC now devotes more attention to the identification of unreported, usually consummated, mergers that could harm consumers’. They will be over-inclusive because, in order to capture most anti-competitive mergers, they will also necessarily capture a high proportion of mergers unlikely to cause competitive harm: Federal Trade Commission, ‘Performance and Accountability Report: Fiscal Year 2004’ (Federal Trade Commission 2004) 44. 127 See for example Pfunder (n 125). 128 See also GCR, ‘Rating Enforcement: The Annual Ranking of the World’s Leading Competition Authorities’ (2012) 15(5) Global Competition Review 14. 129 See generally ICN, ‘Setting Notification Thresholds for Merger Review’ (Report to the ICN Annual Conference, Kyoto, Japan, Merger Working Group, Notification and Procedures Subgroup, 2008) 7. 130 Kool, Harlé and Ombregt (n 97). 131 GCR reports that Russia reviewed 5,406 mergers in 2011 (one for every 25,546 citizens). The next highest was the US, with 2,864 (1,450 at the DOJ and 1,414 at the FTC) equating to one for every 109,567 citizens: GCR, ‘Rating Enforcement: The Annual Ranking of the World’s Leading Competition Authorities’ (2012) 15(5) Global Competition Review 14. Russia did, however, revise its merger thresholds in 2012: Kool, Harlé and Ombregt (n 97).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 28 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 199

Brazil’s notification threshold has also been criticized as low and, after Russia, it reviews more mergers than the other BRIC countries.132 In addition to these monetary thresholds, merger laws and guidelines also differ as to the type of transaction required to be notified.133 Most regimes apply to ‘transactions that lead to a change of control’,134 but this is not always the case and, even where it is, the definition of ‘control’ can vary.135 Countries also differ in their treatment of joint ventures. Clearly articulating what these benchmarks are is important for transparency and business certainty given that the consequences for failure to notify can be severe. To the extent that this is not the case in some jurisdictions work must be directed towards providing guidance designed to improve clarity for parties.136 8.3 Timing of Notification Notification can generally only be made following a ‘triggering event’ and these continue to vary. ‘Asynchronous triggering events’137 can cause significant and unnecessary problems138 and may prevent realization of synergies that would otherwise be possible where mergers are reviewed in multiple jurisdictions, such as coordination between jurisdictions. In this respect the ICN has recommended that parties ‘should be permitted to notify proposed mergers upon certification of a good-faith 132

Kool, Harlé and Ombregt (n 97). For a detailed discussion on this issue see ICN, ‘Defining Merger Transactions for Purposes of Merger Review’ (International Competition Network Sixth Annual Conference, Moscow, Russia May–June 2007). 134 Weiss and Mayer (n 3). 135 These include Germany, the UK and Canada: Weiss and Mayer (n 3). 136 See for example Natalya Morozova, ‘Extraterritorial Application of Merger Control Requirements Under Russian Competition Law’ (V&E Antitrust Update E-communication, 11 February 2011) . For an overview of key differences to consider see Weiss and Mayer (n 3). 137 Joseph Wilson, ‘Globalization and the Limits of National Merger Control Laws: Gaps in Global Governance and the Need for an International Merger Control Regime’ (Doctor of Civil Law Thesis, McGill University, 2002) 47. 138 ‘… whether you have to file, and then lining up all the time frames that are the tricky problems’: ICN, ‘Report on the Costs and Burdens of Multijurisdictional Merger Review’ (Mergers Working Group, Notification and Procedures Subgroup 2004) 17, quoting Donna Patterson testimony on 11 September 1998, 100. 133

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 29 /

Date: 8/9

JOBNAME: Duns PAGE: 30 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

200 Comparative competition law

intent to consummate the proposed transaction’.139 Although variation remains, most jurisdictions now allow notification at early stages of a proposed merger, assisting the coordination of reviews between jurisdictions.140 8.4 Information Requests One of the most significant areas of difference, and one which contributes most significantly to parties’ costs, relates to the nature and level of information required to be submitted to the relevant agency. Informational requirements have expanded and grown in complexity, owing in part to the development of digital technologies,141 as well as to advances in economic analysis, which depend upon extensive data. The EU has typically required much more extensive information in the early stages of review142 than the US (although it reviews far fewer mergers). However, where mergers proceed beyond initial review, the US agencies may seek additional information143 and this can be extremely onerous.144 One of the most onerous regimes is also one of the newest. 139 ICN, ‘Recommended Practices for Merger Notification Procedures’ (Merger Working Group 2002, amended 2003, 2004, 2005, 2006), recommendation III. 140 This is not universally the case; for example, even within the OECD, Finland, Greece, Hungary, Ireland and Portugal all require notification within a relatively short period of a nominated event, such as a decision to merge or publication of a public bid. 141 See Sims, Jones and Hollman observing that the ‘average number of electronic pages gathered from a merging party for the transactions we have worked on in the 2006–08 time period was over 6 million pages’: Joe Sims, Robert C Jones and Hugh H Hollman, ‘Merger Process Reform: A Sisyphean Journey?’ (2009) 23 Antitrust 60, 62. 142 European Commission, Communication Pursuant to Article 3(2) of Regulation 802/2004 Implementing Council Regulation (EC) No 139/2004 on the Control of Concentrations Between Undertakings [2006] OJ C 251/2, Annex I. See also Dave Poddar, ‘Recent Changes to the Australian Merger Control Regime – How the Changes Have Operated in Practice’ (2009) 32 University of New South Wales Law Journal 275. 143 Rules, Regulations, Statements and Interpretations Under the Hart-ScottRodino Antitrust Improvements Act of 1976, 16 CFR §803.20 (2005). 144 See, for example, PriceWaterhouseCoopers, ‘A Tax on Mergers? Surveying the Time and Costs to Business of Multi-jurisdictional Merger Reviews’ (International Bar Association and American Bar Association 2003) 4. See also Joe Sims, Robert C Jones and Hugh H Hollman, ‘Merger Process Reform: A Sisyphean Journey?’ (2009) 23 Antitrust 60, which notes the average cost of

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 30 /

Date: 13/10

JOBNAME: Duns PAGE: 31 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 201

Brazil’s new merger law requires CADE to ‘review more company documentation and data than almost any other competition enforcer’.145 Considerable variation exists among other agencies.146 The burden of extensive information requirements has been mitigated in many jurisdictions, though not all, by an increase in the availability of early termination procedures147 and the use of ‘stage 1’ or ‘short form’ filing forms which normally impose fewer demands for information and therefore can reduce initial compliance costs, particularly for those mergers unlikely to raise genuine competition concerns. The opportunity for pre-filing discussion with agencies may also help to reduce information requirements or clarify notification obligations, but this is not an option in all jurisdictions. At least some of this cost may be unavoidable, given the increasing complexity of merger analysis, stemming in part from ‘greater recognition of economic principles and analysis’.148 In particular, accurate market definition, which is relevant to any assessment of likely competitive impact, requires a great deal of data collection and interpretation to ensure this accuracy.149 Nevertheless, wherever possible, efforts should continue to be made to reduce the burden of onerous information requirements, particularly at early stages of merger review.

notification where a second request investigation is initiated is $5m with a time frame of between 6 and 7 months. 145 Knox (n 7) 17. See also, Fiona Schaeffer and Michael Culhane Harper, ‘A Fundamental Shift: Brazil’s New Merger Control Regime and Its Likely Impact on Cross-Border Transactions’ (August 2012) The Antitrust Source 1. 146 Knox (n 7) 17. 147 In some jurisdictions early termination requests, which allow parties to consummate a merger despite the stipulated period of delay not having passed, are available. For example, in the US in the 2008 fiscal year, of 1,726 transactions notified, 1,385 included requests for early termination and 1,021 of those requests were granted: Federal Trade Commission and Department of Justice, ‘Hart-Scott-Rodino Annual Report Fiscal Year 2008: Section 7A of the Clayton Act, Hart-Scott-Rodino Antitrust Improvements Act of 1976’ (Thirtyfirst Annual Report, Federal Trade Commission and Department of Justice 2008) Appendix A. 148 Baer (n 113) 825, 842. 149 Utton (n 24) 74–75.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 31 /

Date: 8/9

JOBNAME: Duns PAGE: 32 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

202 Comparative competition law

8.5 Fees In many jurisdictions parties are required to pay a fee for notification. These fees vary significantly.150 Some countries impose fees that are designed to cover all or part of the costs of the review process,151 or to fund merger enforcement generally.152 Some use the fees to crosssubsidize other enforcement activity153 and yet others, like the EU, do not impose any fees or impose fees only where a second stage review is initiated or competition concerns are raised.154 In some cases the level of fees will be variable,155 adjusted for the size of the merger, likely competitive impact, or based on the number of hours actually taken to review the merger, but in many cases a flat fee will apply.156 Given the differences in administrative structures and agency funding practices in different jurisdictions, it is unlikely that any ‘model’ or consistent approach to fee-setting for merger review will emerge, at least in the foreseeable future.

9. TIME FRAME AND PROCESS FOR ASSESSMENT The timeline for merger review includes the time associated with determining whether notification is required at all, sometimes involving

150

See generally ICN, ‘Merger Notification Filing Fees’ (Mergers Working Group, April 2005); Julie Clarke, ‘The International Regulation of Transnational Mergers’ (PhD thesis, Queensland University of Technology 2010) 393–397. 151 ICN, ‘Merger Notification Filing Fees’ (Mergers Working Group 2005) 17. 152 It may, for example, also fund litigation against those parties who merge without notifying (whether required to notify or not): ICN, ‘Merger Notification Filing Fees’ (Mergers Working Group 2005) 4. 153 Ibid; J William Rowley, Omar K Wakil and A Neil Campbell, ‘Streamlining International Merger Control’ (EC Merger Control 10th Anniversary Conference, Brussels, 14 September 2000) 7. 154 See generally ICPAC (n 2) 107. 155 ICN, ‘Merger Notification Filing Fees’ (Mergers Working Group 2005) 7. 156 Of the 21 OECD states that impose notification fees, nine impose a flat fee. See also Alexandr Svetlicinii, ‘Competitiveness and Competition: International Merger Control from the Business Prospective’ in Vinko Kandžija and Andrej Kumar (eds), 50 Years of European Union (University of Rijeka, Faculty of Economics 2008) 7; ICN, ‘Merger Notification Filing Fees’ (Mergers Working Group 2005) 7–14.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 32 /

Date: 8/9

JOBNAME: Duns PAGE: 33 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 203

pre-notification discussion with relevant agencies,157 preparation of notification material and the formal review period adopted by the agencies. The time frames for merger review are significant because of their potential to delay the consummation of mergers or contribute to business uncertainty and because of the challenges they may present to transnational cooperation. In relation to formal review periods, there remains significant divergence between agencies, both in relation to notional time frames, which are often prescribed by legislation, and the actual review period, which may be shorter for simple mergers where effective ‘phase 1’ processes are in place, or much longer where the ability exists for ‘clock-stopping’ events which enable the agency to gather further information.158 Thus, while most larger jurisdictions allow 100–150 days for in-depth review (including the time taken, if any, for preliminary review), in practice it can take significantly longer than this in complex cases. Legislated review periods are, therefore, often a poor indicator of agencies’ actual review times. The use of ‘stop-the-clock’ mechanisms in some jurisdictions, combined with different workloads and agency efficiencies, mean that the ‘theoretical’ legislative time frame can give the false impression that some jurisdictions are significantly more efficient at dealing with merger reviews than others.159 Currently Brazil has the longest legislated review period among suspensory jurisdictions with CADE given ‘240 days to review a merger, which can then be extended by as much as 90 days’.160 As a result, it has been suggested that Brazil may ‘overtake China as the antitrust enforcer most likely to hold up the closing of a deal’.161 The Competition Commission in India (CCI) also has a relatively high legislated maximum

157 Agencies also differ in relation to the opportunity they provide parties for pre-filing discussion, noting the contrast between the approach in the EU, where extensive pre-filing discussion is common and that in the US, where it is rare: Deborah L Feinstein, ‘Process Divergence as an Obstacle to Substance Convergence?’ (Summer 2012) 26(3) Antitrust 5, 6. 158 Clarke (n 150) 402–407. 159 For a survey of ‘actual’ review periods see PriceWaterhouseCoopers (n 144) ch 3. 160 Knox (n 7) 17. Compare, suggesting that experience to date has suggested CADE has been efficient in conducting its merger reviews: Paulo Furquim de Azevedo, ‘A Promising, But Tough, Transition: Perspectives on the New Brazilian Competition Law’ (2013) 2 CPI Antitrust Chronicle 4. 161 Knox (n 7) 17 referring to Jones Day client bulletin.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 33 /

Date: 8/9

JOBNAME: Duns PAGE: 34 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

204 Comparative competition law

review period of 210 days, but to date has concluded all merger reviews under its ‘phase I’ process in 78 days or less.162 It is clear that excessive actual review periods can have serious impacts on time-sensitive mergers. This impact applies to mergers ultimately determined to raise competition concerns as well as to those that do not and, particularly for the sake of the latter, it is important that agencies be sufficiently resourced to enable effective and efficient review of all mergers falling within their review thresholds.

10. REMEDIES In most cases, regardless of the administrative structure in place, agencies will have the power to negotiate ‘remedies’ designed to alleviate competition concerns and therefore allow a proposed merger to proceed in modified form, rather than block it (or seek to do so) outright. This is more advantageous to parties (at least in most cases) than having a merger successfully (or even unsuccessfully) challenged by the regulator, in that it involves much less disruption to corporate structure than a subsequent divestiture order. By providing parties with ‘the opportunity to negotiate with the regulator before the merger is consummated’,163 negotiated remedies can help to avoid costly litigation. Moreover, the negotiation may result in a higher welfare outcome, by conditionally permitting a largely efficient merger that might otherwise have been prohibited.164 Agencies must, however, take care to ensure that they do not take advantage of the parties’ desire for approval to leverage remedies which do more than the substantive law would require.165 To this end the 162

Shroff and Uberoi (n 124) 2–3. Chongwoo and Shekhar (n 114) 3 (footnote omitted). 164 Some claim that this is the ‘main rationale for, and the potential benefit of, compulsory notification’: ibid 3. See generally William J Baer and Ronald C Redcay, ‘Solving Competition Problems in Merger Control: The Requirements for an Effective Divestiture Remedy’ (2001) 69 George Washington Law Review 915. 165 Some argue that mandatory pre-merger notification has, at least in the US, ‘removed antitrust law relating to mergers from the courts and transferred it to regulatory agencies who use the power they have gained to improperly extract concessions from applicants’: Shekhar and Williams (n 117) 384 citing Joe Sims and Deborah P Herman, ‘Twenty Years of Hart-Scott-Rodino Merger Enforcement: The Effect of Twenty Years of Hart-Scott-Rodino On Merger Practice: A Case Study In The Law of Unintended Consequences Applied To Antitrust Legislation’ (1997) 65 Antitrust Law Journal 865. 163

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 34 /

Date: 13/10

JOBNAME: Duns PAGE: 35 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 205

ICN has recommended that merger remedies ought to be directed towards the restoration or maintenance of competition and not at improving pre-merger competition and that they should only be considered in circumstances where it is reasonable to believe the merger would contravene the relevant law.166 Remedies take one of two forms: (1) structural (generally requiring some form of divestiture of existing assets or carve-out of proposed acquisition); or (2) behavioural (for example, requiring parties to continue to supply to existing consumers on similar terms). The preference among established regimes is normally for the former, both because structural remedies are often considered more appropriate to address structural conduct and because it alleviates the need for ongoing monitoring of behavioural commitments. Nevertheless, behavioural remedies are used strategically in some cases167 and have been adopted more regularly by some newer regimes, most notably in China, which has utilized extensive behavioural and structural remedies to address competition concerns. Where mergers are assessed in multiple jurisdictions there is clear potential for inconsistent and cumulative remedies which might have a broader impact than necessary to alleviate local concerns. In recent years this has been highlighted by the different approaches taken by China, the US and the EU in relation to several multi-billion dollar transactions. In 2011 two multi-billion dollar deals in the hard disk drive (HDD) market were announced, the first involving a proposed merger between Seagate and Samsung and the second a proposed merger between Western Digital (WD) and Hitachi. The Seagate/Samsung merger received unconditional approvals in several jurisdictions, including Australia, the EU, the US and Japan. China, while approving the merger in March 2012, did so subject to significant behavioural conditions, including a requirement that Seagate increase production capacity of Samsung HDDs and invest at least $800m annually in additional research and development over three years. In a recent press conference, the Director General of the Antimonopoly Bureau of MOFCOM, Shang Ming, stated that there had been widespread concern about the WD merger in China because China is the largest computer manufacturer and highest user of hard drives.168 166 ICN, ‘Recommended Practices for Merger Notification and Review Procedures’ (Merger Working Group 2002) Recommendation XI (‘Remedies’). 167 See for example Sokol and Blumenthal (n 20) 334. 168 The number of mergers notified rose from 16 in 2008 to 171 in 2011: Ming (n 101).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 35 /

Date: 13/10

JOBNAME: Duns PAGE: 36 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

206 Comparative competition law

Similar considerations were evident in the Western Digital/Hitachi merger, which China also approved subject to significant behavioural conditions. This merger was also reviewed in several other jurisdictions, with the US and EU imposing structural undertakings and China’s approval was subject to those undertakings being performed in addition to their own additional behavioural undertakings, which required WD to maintain Hitachi as an independent competitor in the global HDD market.169 These remedies have been described as ‘highly unusual in that they permitted the deals to close, but then prohibited the parties from integrating the acquired businesses and required the acquired businesses to operate separately as if no acquisition had taken place’.170 More recently, on 19 May 2012, MOFCOM announced its decision in the proposed Google/Motorola vertical merger. This deal, worth an estimated US$12.5 billion,171 provided Google with access to thousands of patents controlled by Motorola. Google anticipated that although it would run Motorola as a separate business, the vertical integration of Google’s ‘Android’ with Motorola’s business would enhance its ability to compete with other mobile operating systems.172 The US DOJ and the EC both approved the merger unconditionally in February 2012. MOFCOM, however, reached different conclusions than the DOJ and EC on the likely competitive impact, holding that the merger would increase Google’s ability to foreclose competition. MOFCOM approved the deal subject to a number of behavioural conditions, including that Google continue open source licensing and non-discriminatory treatment in relation to Android.173 169

Henry LT Chen and Frank Schoneveld, ‘China Conditionally Clears Western Digital’s Acquisition of Hitachi’s Hard Disk Drive Business’ (National Law Review, 2012) . 170 ‘Chinese Antitrust Authorities Approve Western Digital/Hitachi GST Deal with Significant Conditions’ (Davis Polk, 9 March 2012) . 171 Brian Womack and Zachary Tracer, ‘Google to Buy Motorola Mobility for $12.5 Billion to Gain Wireless Patents’ (Bloomberg, 15 August 2011) . 172 For further background see Xiao Yong and Li Zhaohui, ‘China: Chinese Agency Conditionally Approves Google’s Acquisition Of Motorola Mobility’ (Mondaq, 18 July 2012) . 173 For further background see ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 36 /

Date: 8/9

JOBNAME: Duns PAGE: 37 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 207

These decisions represent a significant divergence in the approach taken to remedies which cannot be attributed to local conditions alone.174 Recognizing the potential for conflict and duplication, the ICN has made a number of recommendations designed to guide authorities in relation to merger remedies. In particular, it has recommended that remedies be ‘tailored to cure domestic competitive concerns’ and should ‘endeavour to avoid inconsistency’ with remedies imposed in other jurisdictions.175 Despite this, there remains clear potential for divergent outcomes, either because of genuinely different competitive effects in different regions or because of differences in the analysis of competitive effects. More work needs to be done on assessing the appropriate form of remedies in merger cases and the extent to which consideration is or should be given to outcomes in other jurisdictions. This will be assisted by increased cooperation and interaction between agencies, both at a bilateral and multilateral level, and by improving transparency with respect to agency decision-making methodology.

11. TRANSPARENCY The nature of merger review as largely an administrative concern, not reaching the courts, if at all, unless challenged, means that a body of precedent has not developed in most jurisdictions surrounding the circumstances in which mergers will be cleared, blocked or subject to remedies (or the form of those remedies). The processes of merger review are also often secretive, taking account of the commercially sensitive nature of the deals involved, which raises additional concerns about the integrity of process in some cases. One of the difficulties, both for parties in assessing the appropriateness of any proposed remedy, and for practitioners and other agencies, is that, as remedies generally constitute a pure administrative decision, most jurisdictions do not 174

Smriti Arora, ‘China’s Ministry of Commerce Conditionally Clears Google’s Acquisition of Motorola’ (In Competition Blog, 24 May 2012) observing that the decision in Google ‘represents another example of China’s divergence from EU and US competition law regimes’. 175 ICN, ‘Recommended Practices for Merger Notification and Review Procedures’ (Merger Working Group 2002) Recommendation X(E).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 37 /

Date: 8/9

JOBNAME: Duns PAGE: 38 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

208 Comparative competition law

provide any published rationale for their decision to provide the remedy and in some cases do not even publish notice of the remedy itself. The ICN has made various recommendations regarding transparency in the merger review process, including that agencies should make information about the ‘current state of merger control law, policy, and practice readily available to the public’.176 The proliferation of merger guidelines goes a considerable way to promoting transparency of national merger regimes. However, the extent to which individual cases and reasons for decision-making are made available to the public varies considerably, with some agencies providing detailed reasons in all cases raising competition concerns and others, including the US, refraining from providing reasons in (almost) all cases.

12. COOPERATION, CONVERGENCE AND CONFLICT It is clear that the current multitude of PMN regimes impose heavy burdens on firms involved in transnational mergers. A key challenge confronting merger review in the future is to reduce cost and conflict associated with overlapping application of national merger control regimes. The cost and inefficiencies associated with multijurisdictional merger review have been widely recognized for decades. In addition to direct cost to parties and administrators, overlapping national regulation of mergers can generate conflict as a result of the potential for inconsistent outcomes. As a result, there have been various efforts to increase cooperation and convergence, ranging from comity considerations to international agreements on best practice in merger assessment177 and even proposals for supranational review.178

176

Ibid, recommendation VIII(C) (originally recommendation VI). ‘[T]he 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’: Business and Industry Advisory Committee to the OECD (BIAC) and International Chamber of Commerce (ICC), ‘Recommended Framework for Best Practices in International Merger Control Procedures’ (ICC and BIAC 2001) 2. 178 Outside the EU, these proposals have not met with any great success, for a variety of reasons. See further Sokol and Blumenthal (n 20) 321. 177

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 38 /

Date: 8/9

JOBNAME: Duns PAGE: 39 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 209

In one form or another, cooperation now takes place in relation to nearly all multinational mergers.179 Cooperation between agencies in relation to merger review and competition law has taken a variety of forms and has ranged from formal treaty arrangements to informal discussions. Numerically, at least, the majority of formal cooperation agreements have developed bilaterally180 and occur most commonly between countries which have close geographic or political ties, or are at similar stages in the development of their competition laws. For example, bilateral agreements such as those between the EU and the US,181 Canada and the US, Canada and the EU, and Australia and New Zealand, have proven particularly useful at promoting cooperation and understanding in merger review processes. Bilateral agreements between newer and established regimes are also becoming increasingly common. Bilateral competition agreements generally span the full breadth of competition law matters, rather than having any specific focus on the merger review process, although some targeted bilateral merger agreements do exist, most notably between the US and EU.182 As well as assisting coordination of timing and remedies and understanding of substantive issues, the agreements have promoted sharing of expertise and discussion of cases of mutual interests which helps increase understanding of each agency’s response and diffuse any tensions that might otherwise arrive where outcomes differ.183 Traditionally, concern about conflicting outcomes has arisen in the context of the US and EU and much work has been done to improve bilateral relations and cooperation with a view to avoiding conflict and to 179 A Neil Campbell and J William Rowley, ‘The Internationalization of Unilateral Conduct Laws – Conflict, Comity, Cooperation and/or Convergence?’ (2008–2009) 75 Antitrust Law Journal 267, 327. In many cases it should be noted that the ‘cooperation’ tends to be passive in that smaller jurisdictions often simply follow or build on outcomes determined in larger regimes. 180 A Douglas Melamed, ‘International Cooperation in Competition Law and Policy: What Can be Achieved at the Bilateral, Regional and Multilateral Levels’ (1999) Journal of International Economic Law 423, 425. 181 US-EU Merger Working Group, ‘Best Practices on Cooperation in Merger Investigations’ (US-EU Merger Working Group 2011). 182 Ibid. 183 See Sokol and Blumenthal referring to this as increasing trust which can in turn ‘defuse or minimise disagreements early’: Sokol and Blumenthal (n 20) 322. See also US-EU Merger Working Group (n 181) para 2: ‘[c]ooperation … increases the efficiency of the respective investigations, reduces the burden on merging parties and third parties, and increases the overall transparency of the merger review process.’

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 39 /

Date: 13/10

JOBNAME: Duns PAGE: 40 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

210 Comparative competition law

maintaining good relations when conflict does occur. For example, the Boeing/McDonnell Douglas merger (1997), which was cleared by the US and eventually approved by the EC, with concessions, and the GE/Honeywell merger (2001), which was opposed by the EU after being conditionally approved in the US, both sparked significant trans-Atlantic conflict.184 More recently, in 2011, different outcomes were reached in relation to the Intel/McAfee conglomerate merger, worth an estimated $7.7 billion, reviving concerns about protectionism. Despite these significant conflicts, the agencies continue to work closely to cooperate on mergers impacting both jurisdictions and have recently renewed and strengthened their bilateral ties.185 In the vast majority of cases consistent outcomes are reached. More recently, however, potential for conflict has increased with the introduction, most notably, of new PMN regimes in China, Brazil and India. While conflicts of great significance politically have been relatively rare in the past, given the volume of mergers reviewed annually, it is likely that, as more jurisdictions add PMN to their competition law repertoire, particularly where they lack the desired level of experience for investigating and analysing complex competition law issues, the risk of divergence will increase. In addition to bilateral agreements, regional agreements, most notably in Europe,186 have also emerged. Outside Europe, however, they remain limited in geographic scope and suffer many of the inefficiencies of bilateral agreements where a merger transcends regional boundaries. In addition to formal cooperation arrangements, it is clear that ‘informal cooperation occurs on a regular basis’.187 This may involve 184 In both cases claims of national protectionism were made, fuelled by political rhetoric. See for example Eleanor M Fox, ‘GE/Honeywell: The US Merger that Europe Stopped – A Story of the Politics of Convergence’ in Eleanor M Fox and Daniel A Crane (eds), Antitrust Stories (Foundation Press 2007) 331. 185 Department of Justice, ‘United States and European Union Antitrust Agencies Issue Revised Best Practices for Coordinating Merger Reviews: Agencies Celebrate 20th Anniversary of Cooperation Agreement’ (Department of Justice, 14 October 2011) . 186 For example, CARICOM (Caribbean regional merger control), Andean Community (South American merger control), Comesa and Cotonou (Africa) and Interstate Council on Antimonopoly Policy (ICAP) (Baltic States). 187 ABA Section of Antitrust Law, International Antitrust Cooperation Handbook (ABA 2004) 4, fn 8. See also Robert Pitofsky, ‘Competition Policy in a Global Economy – Today and Tomorrow’ (Speech delivered at the European

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 40 /

Date: 13/10

JOBNAME: Duns PAGE: 41 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 211

‘bilateral meeting and technical assistance visits’, informal discussions or staff-to-staff communications.188 Independently of inter-agency agreements, a number of multilateral agreements, recommendations and best practices have been aimed at encouraging cooperation and convergence in competition generally and in relation to multijurisdictional merger review specifically. Most successful have been developments within the OECD and, more recently, the ICN. Related work has been done by the International Chamber of Commerce (ICC),189 the ‘Merger Streamlining Group’,190 the Business and Industry Advisory Committee to the OECD (BIAC),191 the International League of Competition Law (LIDC)192 and by a number of prominent academics193 and government officials,194 which have influenced and informed the work of these organizations and of national laws generally. Institute’s Eighth Annual Transatlantic Seminar on Trade and Investment, Washington DC, 4 November 1998). See also US-EU Merger Working Group noting that a ‘number of these best practices already are routinely employed informally between the US and EU’: US-EU Merger Working Group (n 181). See also Brendan J Sweeney, The Internationalisation of Competition Rules (Routledge 2010) 322. 188 Christine A Varney, ‘Our Progress Towards International Convergence’ (Paper presented at the 36th Annual Fordham Competition Law Institute Conference on International Antitrust and Policy, New York, 24 September 2009) 6. 189 See for example BIAC and ICC (n 177). See also Galloway (n 55) 182. 190 See William Rowley and Omar K Wakil, ‘International Mergers: The Problem of Proliferation’ (Paper presented at the 33rd Annual Conference on International Antitrust Law and Policy, New York, September 2006) 11–12 . See also Galloway (n 55) 182, fn 18, noting that this group ‘was funded by leading international companies and consisted of leading antitrust practitioners’. 191 See for example BIAC and ICC (n 177). See also Galloway (n 55) 182. 192 Formed in 1930 as a Swiss independent scientific association, it consists of national and regional groups as well as individual members. It has previously called for ‘harmonization of the forms and documents required for notification of mergers and acquisitions according to a common model form’: see Richard Whish and Diane Wood, Merger Cases in the Real World – A Study of Merger Control Procedures (OECD 1994) 12, 117–199. 193 See for example Eleanor M Fox and Janusz A Ordover, ‘Internationalising Competition Law to Limit Parochial State and Private Action: Moving Towards the Vision of World Welfare’ (1996) 24 International Business Lawyer 458; Eleanor M Fox, ‘International Antitrust: Against Minimum Rules; for Cosmopolitan Principles’ (1998) XLIII Antitrust Bulletin 5; Eleanor Fox, ‘Antitrust Regulation across National Borders: The United States of Boeing versus the

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 41 /

Date: 13/10

JOBNAME: Duns PAGE: 42 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

212 Comparative competition law

Most of the efforts of the OECD and the ICN in relation to transnational mergers have been in relation to the promotion of procedural convergence of PMN, rather than substantive convergence.195 However, in recent years, both have begun to identify and promote best practice in relation to substantive merger issues. In addition to generic ‘soft’ law recommendations and best practices, the ICN and OECD both provide important opportunities for information exchange and a forum for the discussion of policy issues surrounding merger control. The OECD’s regular roundtables on competition law issues also have important educational value. In this respect, they have significantly influenced the creation and content of most bilateral agreements between OECD countries in particular.196 Unlike the OECD, the ICN comprises a network of competition agencies, rather than governments, and it is not limited in its membership to particular economic or political groupings. It was established in 2001 with 16 initial members and now enjoys an agency membership exceeding 100.197 This has both advantages, such as its inclusiveness of new and developing regimes, and limitations, such as the fact that its principles198 and recommendations199 carry no legal force, have a built-in bias towards European Union of Airbus’ (1998) 16 Brookings Review 30; Eleanor M Fox, ‘Linked-In: Antitrust and the Virtues of a Virtual Network’ (2009) 43 International Lawyer 151; Debra A Valentine, ‘Building a Cooperative Framework for Oversight in Mergers – The Answer to Extraterritorial Issues in Merger Review’ (1998) 6 George Mason Law Review 525; ICPAC (n 2). 194 For example, ‘Sir Leon Brittan, the former European Commissioner for Competition, Dr Wolfgang Kartte, the former Director of the German Federal Cartel Office (BKartA), and James Rill, the former Assistant Attorney General for Antitrust in the US, have all proposed various mechanisms for the increasing harmonization of antitrust procedures’: Whish and Wood (n 192). 195 Galloway (n 55) 182. 196 See for example Antitrust Modernization Commission, ‘Hearing on International Issues, Washington DC, 15 February 2006 (revised 2 March 2006)’ (Antitrust Modernization Commission 2006) 3 (Evidence of Eleanor M Fox); Pitofsky (n 187). 197 It is notable, however, that China’s Ministry of Commerce (MOFCOM) remains one of the very few significant competition agencies which is not a member of the ICN. 198 ICN, ‘Guiding Principles for Merger Notification and Review’ (Merger Working Group, 2002). 199 ICN, ‘Recommended Practices for Merger Notification and Review Procedures’ (first adopted in 2002 and subsequently revised, Merger Working Group 2002) and ICN, ‘Recommended Practices for Merger Analysis’ (first adopted in 2008 and subsequently revised, Merger Working Group 2008).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 42 /

Date: 13/10

JOBNAME: Duns PAGE: 43 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 213

agency agendas (rather than those of government or business)200 and the size of its growing membership may inhibit efforts towards meaningful agreement into the future. To date, however, it has proved highly successful, with clear evidence that the ICN’s recommendations on merger processes and merger analysis have influenced changes to existing regimes,201 in turn promoting ‘soft law’ harmonization towards recognized best practice. Part of the explanation for this success stems from the fact that its exclusive focus is on exploring mechanisms for cooperation and convergence in competition law matters.202 Part of this success may also be attributable to fortunate timing following the greater proliferation of merger regimes in the 1990s and to advances in economic understanding of competition law issues, or more generally to the composition of its membership. Other forums have, to date, proven less successful at enhancing cooperation and convergence in merger law and procedure, thereby eliminating or reducing cost burdens. Several attempts to add international competition issues to the WTO agenda have failed. The United Nations, through the United Nations Conference on Trade and Development (UNCTAD), has demonstrated a greater willingness to involve itself in competition law matters, culminating in the adoption of the Model Law on Competition in 2004, revised most recently in 2010.203 This Model Law is, however, directed towards developing nations and provides limited scope for the convergence of existing merger laws or reductions in transaction and review costs. Nevertheless, they provide a useful starting point and guide, particularly for developing and emerging 200

The ICN does, however, invite participation by non-governmental advis-

ers. 201 See generally Merger Working Group, ‘Merger Working Group Comprehensive Assessment 2010–2011’ (10th Annual Conference of the ICN, The Hague, Netherlands, May 2011). 202 Antitrust Modernization Commission, ‘Hearing on International Issues, Washington DC, 15 February 2006 (revised 2 March 2006)’ (Antitrust Modernization Commission 2006) 3 (evidence of Eleanor M Fox): observing that the ‘ICN is a ground-up network that focuses on the low-hanging fruit; it pursues tasks that are capable of achievement and promises to make a difference.’ Galloway notes that the ICN ‘appears to have consolidated and clarified the previous best practice recommendations’: Galloway (n 55) 183. 203 UNCTAD, ‘Model Law on Competition’ (TD/RBP/Conf.5/7/Rev.2, UNCTAD 2004). This has since been updated: UNCTAD, ‘Model Law on Competition: Substantive Elements for a Competition Law Including Commentaries and Alternative Approaches’ (TD/RBP/CONF.5/7/Rev.3, UNCTAD 2007) and ((TD/RBP/CONF.7/L.1) UNCTAD 2010).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 43 /

Date: 8/9

JOBNAME: Duns PAGE: 44 SESS: 6 OUTPUT: Wed Oct 21 12:16:51 2015

214 Comparative competition law

economies. Asia-Pacific Economic Cooperation (APEC) has also provided a valuable forum for the evaluation of member regimes and is a useful tool in promoting discussion and transparency of competition law regimes,204 but does not make any recommendations or conclude any related agreements for adherence by member states.

13. WHERE TO FROM HERE? Despite improvements in cooperative endeavours and increasing convergence towards international best practice, significant differences remain in all areas of merger law and procedure. In many respects this is not problematic; indeed, differences in law and process can be advantageous – at least to a point. Flexibility in the development and application of law and procedures enables them to be tailored to most accurately reflect different national economic, social and legal conditions and philosophies. In this respect, superimposing the law and processes of developed economies and experienced regimes on developing economies which lack the necessary expertise and experience to apply them effectively is undesirable even if it does produce greater harmonization of law and process. Perhaps more importantly, divergence enables comparison of the effectiveness of different approaches and promotes national efforts to improve the legal and economic analysis and procedural efficiencies, which might in turn help to drive a better global standard.205 There are, however, clear benefits to be achieved through greater harmonization of laws and procedures, combined with international cooperative efforts,206 although it is equally clear that there is no simple mechanism by which this might be achieved. In the past two decades various proposals have been put forward for both the procedural and substantive harmonization of multijurisdictional merger review.207 The focus, at least at government level, has been on 204 See A Douglas Melamed, ‘International Cooperation in Competition Law and Policy: What Can Be Achieved at the Bilateral, Regional, and Multilateral Levels’ (1999) Journal of International Economic Law 423, 429. 205 See Budzinski observing that ‘[m]utual learning through decentralized experimentation is particularly fruitful if there is no safe academic consensus on the “right” solution – like it is the case with competition economics’ Budzinski (n 14) 7–8. 206 Fox, ‘Linked-In’ (n 193) 151, 154. 207 For a discussion of some of the earlier proposals see Whish and Wood (n 192) 12. ICPAC (n 2) 5.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 44 /

Date: 14/10

JOBNAME: Duns PAGE: 45 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 215

limited ‘soft’ (non-binding) harmonization through best-practice recommendations or increased cooperation between nations.208 Calls for more substantive action, such as establishing ‘common’ filing forms for transnational mergers or the negotiation of a ‘hard law’ multilateral treaty governing the regulation of such mergers have been ignored or rejected, with good reason. A comprehensive or even limited international code for merger review, while intuitively appealing, would be impractical given the current divergence in law and institutions and predilection of states to give preference to national welfare interests affected by a proposed merger over global welfare interests. There is also no suitable international body equipped to assess mergers, even if nations would be prepared to defer their analysis to a central body. The idea of a common filing form has been around for some time209 and, while intuitively appealing, it would experience a number of problems. In particular, in an effort to reach agreement on the content of a form, any form produced may reflect a ‘highest common denominator’, alleviating many of the intended gains and perhaps increasing compliance costs for some mergers. An internationally agreed standard would also be more difficult to review and update to reflect emerging best practice. Various other ‘intermediate’ proposals have also been put forward, such as formalized ‘lead jurisdiction’ mechanisms,210 but none have met with any real success, nor are they likely to in the near future.211 As a consequence, outside the EU, whose unique circumstances have allowed it to achieve, though somewhat painstakingly, nearly full harmonization in relation to transnational mergers having an EU dimension, 208

OECD, ‘Policy Roundtables: Cross-Border Merger Control: Challenges for Developing and Emerging Economies’ (DAF/COMP/GF(2011)13, Roundtable No 122, OECD 2011) 11. 209 Stephen G Corones, ‘The Treatment of Global Mergers: An Australian Perspective’ (2000) 20 Northwestern Journal of International Law and Business 255, 284. See also J William Rowley and A Neil Campbell, ‘Multi-Jurisdictional Merger Review – Is it Time for a Common Form Filing Treaty?’ in ‘Policy Directions for Global Merger Review, a Special Report by the Global Forum for Competition and Trade Policy’ (Global Forum for Competition and Trade Policy 1999) 9. Recommending the creation of ‘one or two model filing forms, which request common information in a single format, and which use different country annexes as appropriate’: Whish and Wood (n 192) 108, Recommendation 6; Robert D Paul, ‘The Increasing Maze of International Pre-Acquisition Notification’ (2000) 11 International Company and Commercial Law Review 123. 210 See for example ICPAC (n 2) 76. 211 See further Clarke (n 150) ch 9.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 45 /

Date: 13/10

JOBNAME: Duns PAGE: 46 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

216 Comparative competition law

supranational options for transnational merger review seem unlikely and in most cases undesirable. Taking into account national interest considerations, varying substantive laws, administrative structures, expertise and resources, the most likely – and arguably most effective – way forward is through natural convergence towards international best practice, appropriately tailored to the relevant national economy. Evaluation and debate about best practice in this field should continue to be facilitated by bodies such as the ICN and OECD in their conferences, workshops and roundtables.212 As a preliminary measure, countries should work towards clearly articulating policy goals for merger review. The credibility of the system depends upon clear articulation of its purpose and it also represents the standard against which the success or otherwise of the regulation can be measured. Substantively, while potentially significant differences remain in the analytical approach to competition assessment, this also reflects different economic thinking and expertise and cannot be readily solved by reference to an agreed ‘best practice’ approach. While agencies should be equipped to undertake rigorous economic analysis as part of their merger assessment, it is not anticipated that there will be full convergence around any single superior analytical approach or that application of a common approach will necessarily yield identical results in different countries. The most immediate and arguably the least controversial issue to be addressed is transparency. While this has improved considerably in recent years, transparency, particularly in relation to notification requirements, is a fundamental element of any sound merger regime and merits ongoing attention. In addition to assisting parties to understand when notification is required, transparency in relation to merger outcomes (to the extent possible without breaching confidence) can assist other agencies to understand analytical processes in practice and enhance business understanding and predictability in relation to merger review. Significant efforts must also continue in relation to the level of information that should appropriately be required of parties at different stages in the merger review process. The requirements must be clear and should not extend beyond that necessary to enable the agency to make, first, an initial assessment of the prospect of competition concerns and, should the matter proceed, a meaningful assessment of the likelihood of such an outcome. The desire for information on the part of the agency 212 Larry Fullerton and Megan Alvarez, ‘Convergence in International Merger Control’ (2012) 26(2) Antitrust 20.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 46 /

Date: 8/9

JOBNAME: Duns PAGE: 47 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Current issues in merger law 217

should be balanced against the significant cost that onerous information requests may impose on merging parties. The final area of immediate concern relates to the time taken for review. Review times necessarily vary in practice depending on size and complexity, but the capacity for merger review to hold up or even jeopardize efficient mergers demands that efforts be made to limit the time taken for review to the extent practical to enable thorough analysis to be conducted. The ability to maintain reasonable time frames will necessarily depend on a combination of agency resources, experience and expertise, and work must be done to ensure that agencies are adequately funded to enable them to undertake their analysis in a timely manner. The increasing number of transnational mergers and merger regimes also necessitates effective cooperation between reviewing countries.213 To a degree, this already occurs. Significant discussion between many countries in relation to concerns over potentially anti-competitive transnational mergers is common.214 Less frequently, work sharing may occur on a case-by-case basis.215 This can assist generally in promoting best practice and can assist in improving efficient and coordinated outcomes in particular cases. The importance of all forms of agency cooperation will grow with the emergence of more new and significant merger regimes and energy must be devoted towards extending cooperation efforts to these agencies.

14. CONCLUSION The desirability of preventing anti-competitive mergers justifies the regulatory intrusion into efficient as well as anti-competitive mergers that is occasioned by increasingly prevalent PMN regimes. Despite conflicting philosophies and preferences in relation to the scope and application of competition law generally, and merger law specifically, there remains a high level of convergence on substantive aspects of merger law and relatively small levels of jurisdictional conflict. Work is, however, needed to improve some aspects of procedure which impose unnecessary costs on merging parties. Not all conflict or inefficiency associated with international merger regulation can be overcome so long as a system of sovereign states 213 See Lise Davey and John K Barker, ‘Merger Review Benchmarking Report’ (Competition Bureau (Canada) 2001) 7. 214 Ibid 24. 215 ICPAC (n 2) 7–8.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 47 /

Date: 8/9

JOBNAME: Duns PAGE: 48 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

218 Comparative competition law

remains. However, continued efforts towards developing and promoting best practice and improved dialogue and cooperation between agencies can assist in improving agency efficiency and minimizing conflicting outcomes and it is to that end that future work in this area should be directed.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 11_Chapter8

/Pg. Position: 48 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

9. Vertical conduct: Non-price restraints John Duns

1. INTRODUCTION This chapter considers the challenges to competition policy posed by non-price vertical restraints (‘NPVRs’) and the responses of leading jurisdictions to these challenges.1 Jurisdictions certainly vary in their treatment of NPVRs, and sometimes they do so in important respects. Nevertheless it will be suggested that there appears to be a move towards a greater consistency in approach and that the differences may ultimately be confined to the margins rather than reflecting more fundamental divergences. The chapter begins by dealing with some issues of definition and categorization. One reason for this is simply to deal, for the purposes of this chapter, with the confusing variety of labels applied to the conduct being examined. But the issue of categorization, it will be seen, also runs deeper than this. Section 3 then provides an overview of the economics of NPVRs, with a particular emphasis on the circumstances under which they are likely to be anti-competitive. This is because one of the key issues for competition law in this area is how it should respond to the vast economic literature on NPVRs. Section 4 provides an overview of how NPVRs are dealt with in the leading jurisdictions, followed by a brief comparison with the approach in some other jurisdictions. It is important to stress that the aim of the chapter is not to provide an exhaustive statement of the United States (US) and European Union (EU) law (in particular) but to identify common issues in the treatment of NPVRs and to examine how these are being dealt with. Although the issue of terminology is taken up in the next section, I would like to make two preliminary points at the outset. The first is with regard to the term ‘vertical’ in this context. For the most part this is

1

Vertical price restraints are typically regulated separately from non-price issues and this approach has been followed in this book. Accordingly, vertical price fixing (resale price maintenance) is considered separately in Chapter 10.

219

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 2 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

220 Comparative competition law

uncontroversial2 and is commonly understood to refer to dealings between firms operating at different functional levels, such as dealings between a manufacturer and retailer. The EU Commission (EC), for example, uses the term ‘vertical’, in the context of vertical agreements, as agreements ‘entered into between two or more undertakings each of which operates, for the purposes of the agreement or the concerted practice, at a different level of the production or distribution chain …’.3 In practice, vertical agreements or conduct are often understood as being confined to goods or services in the same product market but, as can be seen from the EU Commission’s definition, this is not always the case. The second point relates to the common requirement that the relevant restraint be contained in an ‘agreement’. As is seen in the next chapter on resale price maintenance (and as is discussed further below), much of the US and EU law on vertical conduct requires an agreement.4 But, as will also be seen, this is not inevitable.

2. DEFINITIONS AND A TAXONOMY As noted above, there is an unfortunate array of terminology employed when referring to NPVRs. Different jurisdictions commonly use different terms for the same conduct. To give just one example, what in the EU now tends to be called a ‘non-compete’ clause is referred to in the US as ‘exclusive dealing’, while the Australian legislation confusingly uses the term ‘exclusive dealing’ to cover all NPVRs. If for no other reason than to provide clarity for the purposes of this chapter, there is a need to provide some consistency in terms used. There is also a more fundamental reason to categorize conduct that might constitute a NPVR. This is because some jurisdictions treat different categories of NPVRs differently. For example, it will be seen that US law treats ‘tying’ differently from other NPVRs, the EU does likewise for certain territorial or customer restrictions and Australia treats ‘third-line forcing’ differently 2 This is not to deny, however, that in practice the characterization of conduct as vertical rather than horizontal may well be difficult: M Lemley and C Leslie, ‘Categorical Analysis in Antitrust Jurisprudence’ (2008) 83 Iowa Law Review 1207, 1219–1220, 1232–1240. 3 Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1, art 1(a) (hereafter ‘vertical BER’): generally referred to as the vertical ‘Block Exemption Regulation’ or ‘BER’. 4 See Chapter 4 for a discussion of the meaning of ‘agreement’.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 2 /

Date: 13/10

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 221

from all other NPVRs. To make sense of these different approaches and to understand the nature of the conduct being regulated, this section provides a brief taxonomy. A useful starting point is to distinguish the parties who use the NPVR. I will refer to the ‘supplier’ as the party higher in the vertical chain (eg the manufacturer or wholesaler), with the other party being referred to as the ‘distributor’. These terms are taken from the EU Guidelines but seem consistent across jurisdictions. The first two categories below deal with the situation where the supplier in some way restricts the distributor in the distribution of the supplier’s product. This may also work in reverse, with the distributor restricting the supplier’s activities and this is taken up in the third category. 2.1 Restrictions on Acquisition The first category of NPVRs is that which involves the supplier restricting the products (goods or services) that the distributor may acquire. There are typically four kinds of restrictions that fall within this category. These are as follows: (a)

The supplier’s requirement that the distributor purchase only the supplier’s products. This is commonly referred to as ‘exclusive dealing’ in the US and as a ‘non-compete’ clause or ‘single branding’ in the EU.5 In this chapter, this conduct is referred to as ‘exclusive dealing’. EU terminology can be confusing here. Sometimes this ‘single branding’ conduct is also referred to as an ‘exclusive purchasing’ arrangement, which is the type being referred to here. The EU also uses the term ‘exclusive distribution’ but this generally refers to a different practice, namely to an arrangement that provides for distributors to be the sole distributors,

5 The EU terminology can be found in the Vertical BER (n 3) and in the European Commission ‘Commission Notice: Guidelines of Vertical Restraints’ (C(2010) 2365 European Commission 2010) (hereafter ‘EU Vertical Guidelines’) and the Commission Regulation (EC) Guidance on the Commission’s enforcement priorities in applying Article 82 of the EC Treaty [now Article 102 of the Treaty on the Functioning of the European Union] [2009] OJ C45/02 (hereafter ‘EU Article 82/102 Guidelines’).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

222 Comparative competition law

at least in specified territories. Such a territorial restriction is considered in category (b) below.6 The supplier’s requirement that the distributor purchase the supplier’s products in minimum quantities. This may have the same effect as an ‘exclusive dealing’ restraint referred to above, that is, if the minimum quantities prescribed effectively preclude the distributor from sourcing from a competitor of the supplier. This is referred to as a ‘minimum quantities’ condition. The supplier’s requirement that the distributor purchase other products along with the product being supplied. This is a ‘tie’. It involves a ‘tying product’ to which another product, the ‘tied product’, is also required to be purchased by the distributor. Ties may be of two types: the tied product may be supplied by the supplier or it may be supplied by a third party. The former is sometimes referred to as ‘full-line forcing’ and the latter as ‘thirdline forcing’. Closely related to tying, and sometimes referred to interchangeably with it, is ‘bundling’. This refers to the practice of supplying different products as a single package (that is, at one price). This is sometimes further classified into ‘pure’ and ‘mixed’ bundling. ‘Pure’ bundling refers to the situation where the bundled products are not supplied separately and ‘mixed’ bundling is where the products are available separately but are provided at a discount if both are purchased. There is sometimes also a distinction drawn between ‘dynamic’ bundling and other types of bundling, with ‘dynamic’ bundling referring to a situation where the tied product is sold in variable, as opposed to fixed, proportions.7

(b)

(c)

(d)

6 The Commission also distinguishes between ‘exclusive’ and ‘selective’ distribution. Exclusive distribution involves the supply of the product to a limited number of distributors who exclusively handle the product. A ‘selective’ distribution requires distributors to meet certain criteria before they will be supplied and is commonly found in the distribution of luxury goods. 7 R Van den Bergh and P Camesasca, European Competition Law and Economics: A Comparative Perspective (2nd edn, Thomson Sweet & Maxwell 2006) 264–265; J Langer, Tying and Bundling as a Leveraging Concern under EC Competition Law (Kluwer Law International 2007) 4–6; EU Article 82/102 Guidelines (n 5) para 48.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 223

2.2 Restrictions on Resupply A supplier may also restrict the resupply of its product. In practice, this is far more likely to involve arrangements for the supply of goods rather than services, as the concept of resupplying a service is an awkward one.8 The two most common illustrations of this are:9 (a)

(b)

a ‘territorial’ (or ‘location’) restraint, where the distributor is restricted in the locations in which it can resupply the product. This is commonly the case, for example, with franchise agreements, where the supplier decides where the products are to be sold by the distributor. Territorial restraints have been prominent in the EU where restrictions on ‘exporting’ outside a Member State have been common; and a ‘customer’ restraint, where the distributor is restricted in the customers to whom it can supply the product. Typically this restraint will be prompted by the supplier’s desire to retain these customers for itself.

2.3 Restrictions by Acquirers As noted, the restrictions referred to above need not necessarily emanate from the supplier. For example, it may be the distributor that insists that the supplier only supply in minimum quantities or that other products of the distributor are to be bundled or tied with its purchases. Further, the restrictions may also work in reverse. That is, it may be the distributor that wishes to restrict the supplier. For example, the distributor may acquire the supplier’s product on the condition that the supplier is not to supply the distributor’s competitor. Before considering the economic assessment of these various restraints, some general comments on the above categories are relevant. The first is that the categories are not, of course, mutually exclusive. The 8 It is conceivable that services could be resupplied. For example, the Australian Act defines a resupply of services to include a supply of services ‘that are substantially similar to the original services, and could not have been supplied if the original services had not been acquired by the person who acquired them from the original supplier’: Competition and Consumer Act 2010, s 4C(f). This definition has not yet been tested in the courts however. 9 Not surprisingly there are a number of common variations on this theme. See, for example, F Alese, Federal Antitrust and EC Competition Law Analysis (Ashgate 2008) 146.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

224 Comparative competition law

one agreement will commonly contain more than one restraint. A typical franchise agreement would normally contain at least exclusive dealing, ties and territorial restraints. Secondly, the restraints referred to may be part of an agreement between the supplier and distributor or may be the result of unilateral conduct. This, it will be seen, is important if the relevant provisions are confined to agreements, which is the case in some jurisdictions. Finally, the restrictions are commonly imposed (whether through agreement or unilaterally) directly when the supplier and distributor deal with each other, but may also be enforced indirectly, as for example where the supplier refuses to deal with a distributor who fails to comply with one of the restrictions imposed.

3. THE ECONOMICS OF NPVRS Perhaps more than any other area of competition law, the economics analysis of NPVRs has been subject to dramatic shifts.10 After the US courts initially treated vertical restraints with the same scepticism that they held for horizontal restraints, the Chicago School persuasively demonstrated that vertical restraints could be pro-competitive. Indeed, they went further, and argued that vertical restraints were inevitably pro-competitive. There was, perhaps predictably, a swing back from the extremity of this position and it is now recognized that NPVRs may also, in some circumstances, be anti-competitive. The conditions under which NPVRs may be anti-competitive are considered below. One clear outcome of the voluminous economic analysis is that vertical conduct is ambiguous in its anti-competitive effect. That is, the conduct itself is not inherently anti-competitive or pro-competitive but requires examination in its particular context in order to determine whether it is in fact pro- or anti-competitive (or competitively neutral). This is important for competition policy as it suggests that a per se prohibition of NPVRs is unlikely to be justified. One of the key insights of the Chicago School was the recognition that, unlike horizontal conduct, an NPVR imposed by a firm does not (at least in an anti-competitive sense) have an impact on the conduct of the firm’s competitors. The restraint is confined to the firm’s product (an ‘intrabrand’ restriction, as opposed to a restriction involving competitors, which is an ‘inter-brand’ restriction). Further, the Chicago School was able to explain why firms might impose NPVRs and how such restraints 10

Columns Design XML Ltd

See generally Van den Bergh and Camesasca (n 7) ch 6.

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 225

can be pro-competitive. As a result of this analysis, NPVRs are now seen as often prompted by pro-competitive concerns – to make the firm imposing them more competitive by making it more efficient. The restraint is seen as an efficient alternative to a decision to integrate vertically. This decision, between integration and relying on a distributor, is influenced by a comparison of the transaction costs of these alternatives and this in turn will be affected by a range of factors, such as the nature of the product and taxation implications.11 The pro-competitive reasons for NPVRs are now well accepted. The EU Guidelines on Vertical Restraints, without claiming to be exhaustive, identify nine justifications that might apply to a particular restraint.12 It is not necessary to set out all of these here, but some are given as an illustration of how NPVRs can be pro-competitive: (a)

NPVRs may prevent competitors of the distributor from taking a ‘free ride’ on services provided by the distributor. A supplier may rely on the distributor providing services (pre-sale or post-sale) to customers as part of selling the product. Failure to provide these services may result in less than optimal sales. For this reason, the supplier may be anxious to ensure that the distributor co-operates by providing the relevant services. The risk for the distributor, however, is that it will incur the costs of these services but will not necessarily obtain the sale. That is, the customer may take the services from the distributor in order to obtain information about the product but then turn to another retailer who, because they have not had to go to the expense of providing the service, can take a ‘free ride’ on the distributor’s services and offer the product at a lower price. This, of course, has become a common issue as a result of consumers being able to buy over the internet. Ultimately the distributor will be deterred from providing the services and the supplier’s sales will suffer as a result. The present point is that vertical restraints, such as exclusive dealing for example, may

11 See generally F Lafontaine and M Slade, ‘Transaction Cost Economics and Vertical Market Restrictions – Evidence’ (2010) 55 Antitrust Bulletin 587; H Hovenkamp, ‘Harvard, Chicago, and Transaction Cost Economics in Antitrust Analysis’ (2010) 55 Antitrust Bulletin 613; E Gellhorn, W Kovacic and S Calkins, Antitrust Law and Economics in a Nutshell (5th edn, Thomson 2004) 338–339. 12 EU Vertical Guidelines (n 5) para 107. See also, for example, R Whish, Competition Law (6th edn, Oxford University Press 2009) 616–618; Van den Bergh and Camesasca (n 7) 206–219; Langer (n 7) 17.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

226 Comparative competition law

prevent this freeriding and so optimize sales for the supplier (a pro-competitive outcome).13 NPVRs may act as a ‘quality certification’. This may apply, in particular, to a new supplier which is looking to break into a market. If it can distribute its product through a well-known distributor, this may be seen by customers as an indication of the quality of the product. To achieve this, the supplier may need to deal exclusively with the distributor. NPVRs may encourage distributors to invest time and money into marketing the supplier’s product. Again, an exclusive dealing clause may protect the distributor who may otherwise be reluctant to make such an investment. As stated by Lafontaine and Slade, ‘[m]any of the efficiency enhancing motives for using VR [vertical restraints] are based on the idea of aligning incentives between manufacturer and retailer. Indeed, when those two links in the vertical chain are independent firms, each has its own objectives, and those objectives can diverge’.14 NPVRs may generate economies of scale in distribution. The supplier may impose a minimum quantities requirement, an exclusive dealing requirement, a territorial restriction or bundle products in order to reduce delivery or other distribution costs. NPVRs may ensure quality control. A supplier may impose a tie, for example, as a means of ensuring that the tying product works most effectively.

(b)

(c)

(d)

(e)

These illustrations suggest that there will be at least some occasions where vertical restraints will be pro-competitive. Indeed, empirical studies suggest that NPVRs are in fact more likely than not to be procompetitive.15 However, as noted above, the Chicago School argument went further and suggested that vertical restraints were invariably procompetitive. The argument here is that, once the supplier’s product is in the hands of the distributor, the supplier’s only interest is to have the distributor maximize sales of the supplier’s product. Therefore, the only 13

For a broader view of the problem of freeriding and the role of NPVRs in combating it, see B Klein and A Lerner, ‘The Expanded Economics of Freeriding: How Exclusive Dealing Prevents Free-riding and Creates Undivided Loyalty’ (2007) 74 Antitrust Law Journal 473. 14 Lafontaine and Slade (n 11) 599. 15 Lafontaine and Slade (n 11); J Cooper, L Froeb, D O’Brien and M Vita, ‘Vertical Antitrust Policy as a Problem of Inference’ (2005) 23 International Journal of Industrial Organization 639.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 227

rational explanation for a vertical restraint is that it must be designed to increase sales by the distributor. The manner in which it might do so has been outlined above. However, it subsequently became clear that this reasoning was flawed, at least as a complete explanation of NPVRs. One reason for this is that vertical restraints may in fact be employed as a means of enforcing an anti-competitive horizontal restraint. For example, if suppliers enter into a market-sharing cartel, one means of implementing this might be for them to enter into an exclusive dealing arrangement with distributors in their allocated areas.16 Likewise, if distributors enter into an anticompetitive agreement, one way of implementing this might be to turn to the supplier and request that it impose a vertical restraint; the supplier might effectively divide the retail market through territorial restraints, for instance.17 Clearly the main concern for competition policy here is the horizontal arrangement, but the present point is that the vertical restraint in such a case is not being used for the pro-competitive purposes that the above Chicago School argument would otherwise suggest. Leaving aside horizontal concerns, the argument that NPVRs are pro-competitive generally holds true where the supplier lacks market power.18 In such a case the supplier faces the risk of the customer (and so distributor) turning to a competitor if the vertical restraint is anticompetitive. The vertical restraint, as the Chicago School argued, in effect provides a means for the supplier to compete effectively. However, the situation potentially changes if the supplier has market power. In this case, the possibility emerges for vertical restraints to be used for anti-competitive purposes. What economic analysis has shown is that a supplier with market power may be able to use vertical restraints to maintain or increase that power. It does this by using the vertical restraint to create a barrier to entry, either by ‘foreclosure’ or ‘leveraging’.19

16

I Paldor, ‘The Vertical Restraints Paradox: Justifying the Different Legal Treatment of Price and Non-Price Vertical Restraints’ (2008) 58 University of Toronto Law Journal 317, 329. 17 Ibid 327. 18 On the other hand, intra-brand competition may be necessary if there is little inter-brand competition: Whish (n 12) 613. On intra-brand competition generally in this context, see C Rodes, ‘Giving Teeth to Sherman Act Enforcement in the Intrabrand Context: Weaning Courts Off Their Interbrand Addiction Post-Sylvania’ (2009) 84 Notre Dame Law Review 957. 19 See generally K Hylton, Antitrust Law; Economic Theory & Common Law Evolution (Cambridge University Press 2003) ch 14.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

228 Comparative competition law

3.1 Foreclosure One of the key concerns with NPVRs, and particularly exclusive dealing, is that the restraints can be used by a firm to foreclose the market.20 The argument is that an exclusive dealing agreement between a supplier and distributor will deny competitive suppliers an outlet for their products. By increasing the competitor’s costs of supply in this way (what economists refer to as ‘raising rivals’ costs’), the supplier with the exclusive dealing agreement has erected a barrier to entry.21 However, a firm will only be able to raise barriers effectively in certain situations. An important requirement is that the firm has market power. An attempt to impose an exclusive dealing arrangement on a distributor that would otherwise resist it will be ineffective if the supplier lacks market power. The distributor will simply turn to a competitor that does not require such a restriction. It is only where the distributor lacks this option, because of the supplier’s market power, that foreclosure becomes viable. Even then, the length of the exclusive dealing agreement may deny effective foreclosure. If the agreement is for a short period only, this will allow other distributors to compete for the supplier’s customers when the period expires. It is thus longer-term restraints by suppliers with market power that are of real concern in this context. 3.2 Leveraging One of the earliest concerns with tying was its perceived effect on the market for the ‘tied’ product. A firm with market power in the ‘tying’ product market was thought to be using its power in that market as a ‘lever’ to gain market power in the tied product market. As noted above, such ties are often now seen as merely an efficient bundling of products into a package more desirable to consumers than a supply of the products separately. This pro-competitive use of tying was explained by the Chicago School as being the only rational explanation for tying – a firm could only extract monopoly profits once and so after it had done so in the tying product market, any attempt to increase prices in the tied product market would be counterproductive as it would only result in a

20 Much has been written on this. For a summary, see Van den Bergh and Camesasca (n 7) 215–217. 21 Strictly, this will be a ‘mobility’ barrier rather than an entry barrier if the other supplier has already entered the market.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 229

reduced demand for the two products as a whole.22 However, it is now recognized that in certain conditions tying may in fact be anticompetitive.23 As with exclusive dealing discussed above, one of these conditions is that the firm must have market power in the tying product market.24 Without this, any attempt to impose an unwelcome tie by a supplier will be rejected in favour of a competitive supplier who does not insist upon such a tie. The point of the above overview is not to attempt to draw a conclusion as to whether NPVRs are pro- or anti-competitive.25 It is to demonstrate that NPVRs are competitively ambiguous. This would suggest that competition law should subject NPVRs to some sort of ‘lessening of competition/rule of reason’ analysis. In any such assessment market power will be critical. One issue on which there is agreement is that if the supplier lacks market power then the NPVR is very unlikely to be of interest to competition policy. Another challenge, which the leverage discussion above raises, is that if tying and bundling are to receive particular treatment, how are the courts to distinguish between a supply of one product that has different components (such as shoes and shoelaces, jars and lids, etc) from a tie of two otherwise separate products? With these points in mind, we turn to a brief overview of the treatment of NPVRs in different jurisdictions.

4. US APPROACH There are three sections of the US antitrust laws that may apply to NPVRs. These are ss 1 and 2 of the Sherman Act 1890 and s 3 of the 22 See, for example, R Bork, The Antitrust Paradox: a Policy at War with Itself (Basic Books 1978) 380–381. 23 See, for example, L Kaplow, ‘Extension of Monopoly Power Through Leverage’ (1985) 85 Columbia Law Review 515; M Whinston, ‘Tying, Foreclosure and Exclusion’ (1990) 80 American Economic Review 837. For a good summary, see E Hortle, ‘Third-line Forcing and the Notification Process: Yet Another Reason to Abolish the Per Se Prohibition’ (2010) 18 Competition and Consumer Law Journal 39. 24 See generally Whinston (n 23); R Joseph, ‘Antitrust Law, Franchising and Vertical Restraints’ (2011) 31 Franchise Law Journal 3; Langer (n 7) 17–28. 25 Indeed, according to one economist, ‘The economic analysis of bundling and tying is a mess’: M Walker, ‘Bundling: Are US and European Views Converging?’ (2008) 4 European Competition Journal 275, 280.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 11 /

Date: 13/10

JOBNAME: Duns PAGE: 12 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

230 Comparative competition law

Clayton Act 1914. Section 1 of the Sherman Act prohibits anticompetitive agreements (‘a combination … in restraint of trade’) and s 2 prohibits ‘monopolization’ or ‘attempted monopolization’. The provisions of both sections are considered in detail in other chapters26 but their application in the context of NPVRs is considered here. Section 3 of the Clayton Act is the only section to make explicit provision for NPVRs. It provides that it shall be unlawful for any person engaged in commerce … to lease or make a sale or contract for the sale of goods, wares, merchandise, machinery, supplies or other commodities … or fix a price charged therfor, or discount from, or rebate upon, such price, on the condition, agreement or understanding that the lessee or purchaser thereof shall not use or deal in goods [etc] of a competitor or competitors of the lessor or seller, where the effect of such lease, sale or contract for sale or such condition, agreement or understanding may be to substantially lessen competition or tend to create a monopoly in any line of commerce.

In comparison with s 1 of the Sherman Act, considered below, s 3 of the Clayton Act is in one respect broader in scope but also, in another respect, narrower. It is broader in that, unlike s 1, the NPVR need not necessarily be part of an agreement. The meaning of agreement is examined in detail in Chapter 4 of this volume. As seen there, despite the relatively liberal interpretation of agreement in s 1, the requirement that the supplier and distributor have ‘combined’ may nevertheless allow some NPVRs to escape examination under the section. Section 3 of the Clayton Act, in contrast, extends to a ‘lease or … sale’ on a ‘condition’ and so a ‘take it or leave it’ NPVR may be caught under s 3 of the Clayton Act even though it may avoid s 1 of the Sherman Act. On the other hand, s 3 is narrower than s 1 in that it applies only to ‘goods, wares, merchandise, machinery, supplies or other commodities’, and so does not apply to NPVRs that are imposed in connection with the supply of services. Section 1 of the Sherman Act has no such restriction. Despite some differences in scope between the two provisions, the approach taken by the courts to NPVRs has been fundamentally similar under both Acts. As will be seen below, the most significant issue concerning the application of these provisions has been whether NPVRs should be assessed according to the rule of reason or be subject to a per se (or modified per se) treatment. It will be seen that the general trend in the decisions has been towards subjecting NPVRs to a rule-of-reason analysis. Having said this, the absence of recent Supreme Court decisions 26

Columns Design XML Ltd

See Chapters 4, 5 and 6.

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 12 /

Date: 8/9

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 231

in the area and the inconsistency in treatment of the different types of NPVRs, particularly tying and bundling, creates some uncertainty. 4.1 Sherman Act s 1 and Clayton Act s 3 4.1.1 Territorial and customer restraints The most settled treatment of NPVRs has been with regard to territorial and customer restraints. In Schwinn, US v General Motors Corp27 the Supreme Court had held that a territorial restraint was prohibited per se.28 In a significant pendulum swing the Supreme Court in Continental TV, Inc v GTE Sylvania29 overruled Schwinn and, in a judgment drawing strongly on economic reasoning, held that such NPVRs should be subject to a rule-of-reason analysis.30 Despite all the uncertainties that a rule of reason brings with it, the virtue of a full rule-of-reason approach is that it allows a complete analysis of the NPVR and, as seen in section 3 of this chapter, this is consistent with the competitive ambiguity of NPVRs. 4.1.2 Exclusive dealing There is less clarity around exclusive dealing in comparison with the restraints considered above, primarily because of the lack of more recent, definitive Supreme Court decisions on this type of NPVR. Nevertheless, the existing decisions suggest a rule-of-reason analysis is to be applied to them. The early Standard Stations decision31 distinguished between tying and exclusive dealing, holding that the latter was less likely to be anti-competitive and so required evidence of substantial foreclosure of the market. The subsequent decision of Tampa Electric Co v Nashville Coal Co32 confirmed this approach and adopted a stronger emphasis on market power in doing so. This was in 1961. Five years later, in FTC v Brown Shoe Co,33 the Supreme Court muddied the waters somewhat in a decision made under s 5 of the Federal Trade Commission Act 1914, but

27

384 US 127 (1966). See also US v Topco Associates, Inc 405 US 596 (1972). 29 433 US 58 (1977). 30 On such ‘pendulum swings’ in enforcement of NPVRs, see S Colino, Vertical Agreements and Competition Law. A Comparative Study of the EU and US Regimes (Hart Publishing 2010) 90 and the articles cited therein. 31 Standard Oil Co of California v US 337 US 293 (1949). 32 364 US 320 (1961). 33 384 US 316 (1966). 28

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

232 Comparative competition law

this does not seem to have detracted from the influence of Tampa Electric.34 4.1.3 Tying and bundling In the early decision of Standard Oil Co v US35 the Supreme Court notoriously declared that tying clauses ‘serve hardly any purpose beyond the suppression of competition’ and prohibited them per se.36 This was soon stated to be the case for both s 1 of the Sherman Act and s 3 of the Clayton Act.37 Subsequent decisions made some inroads into this per se approach but have been unable to shrug it off entirely.38 In what has been described as ‘one of the more deliberate missed opportunities in modern antitrust decision-making’,39 the Supreme Court majority in Jefferson Parish Hospital District No 2 v Hyde40 (hereafter ‘Jefferson’) held that tying was not to be subjected to a full rule-ofreason analysis although it did significantly modify the earlier per se approach.41 Unlike the Court in Sylvania, the majority in Jefferson considered it ‘far too late in the history of our antitrust jurisprudence to question the proposition that certain tying arrangements pose an unacceptable risk of stifling competition and therefore are unreasonable “per se”’.42 Under the Jefferson approach, tying is prohibited per se only where the plaintiff is able to show that (1) the defendant has market power in the tying product;43 and (2) a ‘substantial volume of commerce’ 34

See, for example, Eastern Food Services Inc v Pontifical Catholic University Services Association 357 F 3d 1 (1st Cir 2004). 35 337 US 293, 305. 36 See also International Salt Co v US 332 US 392 (1947). 37 Northern Pacific Railway Co v US 356 US 1 (1958). See also US v Jerrold Electronics Corp 187 F Supp 545 (1960), affirmed 365 US 567 (1961). 38 See, for example, United States Steel Corp v Fortner Enterprises 429 US 610 (1977). 39 Gellhorn, Kovacic and Calkins (n 11) 389. 40 466 US 2 (1984). 41 See generally C Ahlborn, D Evans and A Padilla, ‘The Antitrust Economics of Tying: A Farwell to Per Se Illegality’ (2004) 48 Antitrust Bulletin 287. 42 466 US 2 (1984) 7; A ‘concurring’ majority, led by Justice O’Connor, on the other hand, declared that ‘The time has … come to abandon the “per se” label and refocus the inquiry on the adverse economic effects, and the potential economic benefits, that the tie may have’ (35). 43 In contrast with early decisions, which were prepared to presume market power from the holding of a patent, the courts now require proof of market power: see, for example, Illinois Tool Works International Inc 547 US 28 (2006). In that decision, the Court noted that ‘Over the years … this Court’s strong disapproval of tying arrangements has substantially diminished’ (33). For a

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 233

will be foreclosed by the tie. The Court of Appeals in the Microsoft case distinguished Jefferson on its facts, holding that at least on the facts of that case, ‘where the tying product is software whose major purpose is to serve as a platform for third-party applications and the tied product is complementary software functionality’, per se treatment was inappropriate.44 The situation with bundling is less settled. Some US courts have treated bundling in a similar manner to tying.45 Others have been more lenient and prohibited bundling only where at least one of the bundled products is sold below cost.46 This more timid approach in the case of tying and bundling conduct seems difficult to justify, at least in economic terms, and creates a doubtful distinction in the treatment of tying and bundling compared with other NPVRs. One significant consequence is that it requires a legal distinction to be drawn between tying and other conduct. Distinguishing tying, which assumes the existence of two products (the tying and the tied products), from exclusive dealing is not always easy. The court in Jefferson held that this is a question of whether there is a distinct consumer demand for each product.47 The Court of Appeals in the Microsoft case said that this approach generally worked as a proxy for efficient conduct – that is, the more distinct the demand for the two products, the less likely that a tie would be employed for efficiency reasons. However the Court in that case also stated that this test may fall

discussion of this decision see B Kobayashi, ‘Spilled Ink or Economic Progress? The Supreme Court’s Decision in Illinois Tool Works v Independent Ink’ (2008) 53 Antitrust Bulletin 5. See also R Joseph, ‘Antitrust Law, Franchising and Vertical Restraints’ (2011) 31 Franchise Law Journal 3. Compare the contentious approach to the identification of market power in Eastman Kodak v Image Technicolor Services 504 US 451 (1992), discussed in H Hovenkamp, ‘Market Power in Aftermarkets: Antitrust Policy and the Kodak Case’ (1993) 40 UCLA Law Review 1447. 44 US v Microsoft 253 F 3d 34, 95. 45 See, for example, 3M Co v LePage’s Inc 324 F 3d 141 (2003). 46 See, for example, Cascade Health Solutions v PeaceHealth 515 F 3 883. The cases are discussed in N Economides and L Lianos, ‘The Elusive Antitrust Standard on Bundling in Europe and in the United States in the Aftermath of the Microsoft Cases’ (2009) 76 Antitrust Law Journal 483. 47 As stated by the majority, ‘the answer to the question whether one or two products are involved turns not on the functional relation between them, but rather on the character of the demand for the two items’: 466 US 2 (1984) 19.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

234 Comparative competition law

down in the case of an innovative tie, that is, where previously untied products are tied for the first time.48 4.2 Sherman Act s 2 NPVRs may also constitute monopolization or attempted monopolization within the meaning of s 2 of the Sherman Act. Unlike the cases referred to above, s 2 clearly requires proof of market power in the traditional sense. Tying conduct has been held to fall within s 2 on the basis that the defendant has leveraged its market power in the tying product to monopolize or attempt to monopolize the tied market.49 Exclusive dealing50 and bundling51 have also been held to constitute exclusionary conduct, in certain limited circumstances, within s 2.

5. EU APPROACH The EU legislative approach towards NPVRs has some similarities with that of the US, although the different treatment of the two Microsoft cases also highlights important differences.52 Article 101 of the Treaty of the Functioning of the European Union (TFEU) prohibits anticompetitive agreements, along the lines of s 1 of the Sherman Act, and this provision has been applied by the courts to vertical agreements. Article 102, which prohibits dominant firms from abusing their dominance, and so has similarities with s 2 of the Sherman Act, has been held, as we saw with s 2, to apply to NPVRs. However, while a fundamental similarity between the two jurisdictions may be evolving,53 at present the 48

US v Microsoft 253 F 3d 34, 89. See, for example, Berkey Photo, Inc v Eastman Kodak Co 603 F.2d 263 (2nd Cir 1979); Spectrum Sports Inc v McQuillan 506 US 447 (1993); Verizon Communications Inc v Law Offices of Curtis v Trinko, LLP 124 S Ct 872 (2004). 50 See, for example, Lorain Journal Co v US 342 US 143 (1951). 51 Le Page’s Inc v 3M 324 F 3d 141 (3d Cir 2003). See B Klein and A Lerner, ‘The Law and Economics of Bundled Pricing: Le Page’s, Peace Health and the Evolving Antitrust Standard’ (2008) 53 Antitrust Bulletin 555; B Burchfield, ‘Emerging Consensus on Bundled Discounts Under Section 2 of the Sherman Act’ (2009) 54 Antitrust Bulletin 233. 52 For a comparison of the EU and US approaches to tying and bundling, see generally Economides and Lianos (n 46). 53 This has been suggested by a number of commentators. See, for example, Colino (n 30) 5. 49

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 235

differences in treatment of NPVRs are more striking than the similarities.54 Two of the key features of the EU treatment of NPVRs, which are absent in US law, are: (1) the possibility (under art 101(3)) of a firm being exempted from liability under art 101 if specified criteria are met; and (2) the objective of having an integrated EU market. Both aspects have had a significant impact on the treatment of NPVRs and have led to a more restrictive approach than in many jurisdictions, and certainly than is the case in the US. 5.1 Article 101(1) TFEU Article 101(1)55 prohibits ‘all agreements between undertakings … and concerted practices which may affect trade between Member States and which have as their object or effect the prevention, restriction or distortion of competition within the common market’. The meaning of ‘agreements’ and ‘concerted practices’ is discussed in Chapter 4.56 The key reference for current purposes is ‘object or effect’. This is considered generally in Chapter 5 so the main focus in this chapter will be on its application in the context of NPVRs. In the early case of Consten and Grundig57 an argument that a predecessor of art 101(1) did not apply to vertical agreements was rejected. It should be noted at this point that the courts have held that an agreement must have an ‘appreciable’ impact on competition for it to be in breach of art 101(1).58 This has been reinforced by the Commission, which has published a Notice in which NPVRs that involve a market share of 15 per cent or less by the parties to the agreement are said to be

54

See generally A McElroy and P Haleen, ‘A Widening Gap? An Overview of US and EU Antitrust Rules for Franchisors’ (2009) 29 Franchise Law Journal 23. 55 Article 101 came into effect in December 2009, when it replaced the prior art 81. The two articles are, for all relevant purposes, the same. 56 For the meaning of ‘effect on trade between Member States’, see Whish (n 12) 142–146. 57 Etablissements Consten SARL and Grundig-Verkaufs-GmbH v Commission [1966] ECR 299. 58 See, for example, Franz Volk v SPRL Ets J Vervaecke [1969] ECR 295 and Javico v Yves Saint Laurent [1998] ECR I-1983. These decisions suggest that even agreements prohibited by ‘object’, as discussed below, need to meet this ‘appreciability’ test, although commentators generally agree that this is unlikely to occur in practice.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 17 /

Date: 13/10

JOBNAME: Duns PAGE: 18 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

236 Comparative competition law

‘de minimis’ and so outside art 101(1).59 This ‘safe harbour’ does not apply to ‘hardcore restrictions’. Such hardcore restrictions include specified territorial or customer restrictions.60 Once it was accepted that vertical agreements were within art 101, the question then became whether NPVRs would be treated as prohibited ‘by object or effect’. An agreement that has an ‘object’ of preventing competition is one that by its very nature is anti-competitive and thus, as explained in Chapter 5, is prohibited per se.61 Agreements that do not have such an object are prohibited only if they can be shown to have an anti-competitive effect. It is here that we find the impact of the EU objective of encouraging an integrated market. In one of the clearest illustrations of how the objective of creating a common market has influenced EU competition law, if an NPVR creates a market division along Member State lines, it will be treated as a per se breach of art 101(1). Thus an ‘export ban’, that is, a territorial restraint that prohibits the distributor from exporting outside allocated territory is prohibited per se.62 Likewise, restricting or preventing parallel imports is anticompetitive by ‘object’.63 Attempts to achieve a similar outcome indirectly, as for example by denying the benefit of a guarantee to those in other states, is also prohibited per se.64 So conduct that might be treated as designed to prevent freeriding, and hence pro-competitive in many jurisdictions, is prohibited per se in the EU.65 NPVRs that do not fall within this ‘object’ category will be prohibited by art 101(1) if they have an anti-competitive ‘effect’. Unlike the position in the US, NPVRs in the EU are far more likely to be enforced by the EC than by private action and so the EC’s view on their anti-competitiveness 59

See Commission Notice (EC) 2001 on agreements of minor importance which do not appreciably restrict competition under Article 81(1) [now Article 101(1)] of the Treaty establishing the European Community (de minimis) [2001] OJ C368/7. 60 Ibid point 11. 61 See the recent Court of Justice decision in GlaxoSmithKline Services Unlimited v Commission [2010] 4 CMLR 2, especially at paras 55–58. 62 Etablissements Consten SARL v Grundig-Verkaufs GmbH v Commission [1966] ECR 299; Case 19/77 Miller International Schallplatten GmbH v Commission [1978] ECR 131, [1978] 2 CMLR 334; GlaxoSmithKline Services Unlimited v Commission [2010] 4 CMLR 2. 63 Case 468/06-478/06 Sot Lelos kai Sia EE v GlaxoSmithKline AEVE [2008] ECR I-7139; GlaxoSmithKline (n 61). 64 Case 31/85 ETA Fabriques d’Ebauches v SA DK Investment [1985] ECR 3933, [1986] 2 CMLR 674; GlaxoSmithKline (n 61). 65 See generally Van den Bergh and Camesasca (n 7) ch 6.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 18 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 237

takes on a particular significance. The EC’s approach can be seen in its Vertical Guidelines. The Guidelines largely adopt the traditional economic analysis of NPVRs, discussed above.66 For this reason it has been suggested that EU competition law ‘is beginning to take a new more US-style profile’.67 The decisions, with the significant exception of territorial restraints noted above, also acknowledge that NPVRs will often be pro-competitive and so are generally subjected to a full economic analysis, with the emphasis on whether inter-brand competition is adversely affected. In this context, franchise agreements have generally been viewed favourably,68 as have minimum quantity provisions69 and other restrictions contained in vertical agreements.70 5.2 Article 101(3) TFEU A distinctive feature of EU competition law is the possibility provided by art 101(3) for exemption from what would otherwise be a liability under art 101(1). As can be seen from the terms of paragraph (3) below, the agreement will be exempt if it either ‘improves the production or distribution of goods’ or it promotes ‘technical or economic progress’. Exemption is not available, however, if the agreement is otherwise anti-competitive or unless customers receive a fair share of the resulting benefit. The paragraph provides that: (3) The provisions of paragraph 1 may, however, be declared inapplicable in the case of: Any agreement or category of agreements between undertakings; Any decision of category of decisions by associations of undertakings; Any concerted practice or category of concerted practices;

+ + +

which contribute to improving the production or distribution of goods or to promoting technical or economic progress, while allowing customers a fair share of the resulting benefit, and which does not: 66

EU Vertical Guidelines (n 5) 1.3.1 and 2. Colino (n 30) 5. 68 Case 161/84 Pronuptia de Paris GmbH v Pronuptia de Paris Irmgaard Schillgalis [1986] ECR 353, [1986] 1 CMLR 414. 69 Delimitis v Henninger Brau AG [1991] ECR-I-935. 70 See, for example, Metro-SB-Grossmarkte GmbH v Commission [1977] ECR 1875; Neste Markkinointi Oy v Yotuuli [2000] ECR I-11121. There have been few cases on tying under art 101: see Whish (n 12) 636, although some decisions under art 102 are considered below. 67

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 19 /

Date: 13/10

JOBNAME: Duns PAGE: 20 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

238 Comparative competition law (a) impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives; (b) afford such undertakings the possibility of eliminating competition in respect of a substantial part of the products in question.

This exemption provision has been particularly significant in the context of NPVRs. It has in fact been suggested that in this context the exception has become the rule71 and that it has created an equivalent to the US rule-of-reason approach.72 Since the introduction of the so called ‘modernization package’, national courts and regulators, as well as the EC, are able to determine whether art 101(3) applies.73 Nevertheless, in practice the application of the exemption is largely a matter for the Commission and it has passed a regulation which establishes categories of exemption (a ‘block exemption’) within art 101(3) for NPVRs.74 The current block exemption regulation (‘BER’) came into effect in April 2010.75 The Preamble to the BER acknowledges that vertical restraints can be pro-competitive. It recites that ‘[c]ertain types of vertical agreements can improve economic efficiency … by facilitating better coordination between the participating undertakings. In particular, they can lead to a reduction in the transaction and distribution costs of the parties and to an optimisation of their sales and investment levels.’ However the Preamble then goes on to state that ‘[t]he likelihood that such efficiency-enhancing effects will outweigh any anti-competitive effects due to restrictions 71

Colino (n 30) 96–97. But see Whish (n 12) 618 on why it may still be important to know if an agreement falls within para (1). Colino argues that there is in fact too much emphasis on the operation of the exception in para (3) and that a more appropriate focus would be on para (1): Colino (n 30) 98–99. See also B Hawk, ‘System Failure: Vertical Restraints and EC Competition Law’ (1995) 32 Common Market Law Review 973. 72 Colino (n 30) 93, citing Metropole Television v Commission [1996] ECR II-649 para 76. 73 Reg 1/2003 art 1, which came into effect in May 2004, provides for this. 74 There is also a Commission regulation (Regulation 1400/2002) dealing specifically with vertical agreements for the purchase, sale or resale of new motor vehicles. This regulation remains in force until 31 May 2013. See generally Colino (n 30) 111–129. 75 Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1. This replaced Commission Regulation (EC) 2790/1999 on the Application of Article 81(3) of the Treaty to Categories of Vertical Agreements [1999] OJ L336, which expired in May 2010. For current purposes, the two regulations are largely the same.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 20 /

Date: 13/10

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 239

contained in vertical agreements depends on the degree of market power of the parties to the agreement.’ Market share is used as a proxy for market power in this context. The Preamble states that: [i]t can be presumed that, where the market share held by each of the undertakings party to the agreement on the relevant market does not exceed 30%, vertical agreements which do not contain certain types of severe restrictions of competition generally lead to an improvement in production or distribution and allow consumers a fair share of the resulting benefits.

The ‘severe restrictions’ referred to are ‘certain types of severe restrictions of competition such as minimum and fixed resale prices, as well as certain types of territorial protection’ and the Preamble states that these provisions ‘should be excluded from the benefit of the block exemption’. Other NPVRs, even though not ‘severe’, are entitled to exemption only if they meet certain conditions. The Preamble states that the ‘exemption of non-compete [exclusive dealing] obligations should be limited to obligations which do not exceed a defined duration’ and that ‘any direct or indirect obligation causing the members of a selective distribution system not to sell the brands of particular competing suppliers should be excluded from the benefit of this Regulation’. The Articles of the BER then flesh out these general principles. The starting point for the treatment of vertical restraints is provided for by art 2. It states that ‘[p]ursuant to Article 101(3) of the Treaty and subject to the provisions of this Regulation, it is hereby declared that Article 101(1) of the Treaty shall not apply to vertical agreements.’76 However the BER goes on to qualify this general exemption.77 One relatively straightforward limitation is that the block exemption applies only where the market share of the supplier or the buyer does not exceed 30 per cent.78 However, other, more complex limitations are contained in a scheme that denies a block exemption to what the BER refers to as ‘hardcore’ 76

The definition of ‘vertical agreement’ is contained in art 1 and is set out in the Introduction to this Chapter. This definition is elaborated in the EU Vertical Guidelines (n 5) para 25. 77 It does not follow from a failure to meet the requirements of the exemption that there has been a breach of TFEU art 101. 78 See art 3 of the regulation. Calculation of this market share is elaborated in the Vertical Guidelines (n 5). One of the changes made by the 2010 Regulation was the introduction of a buyer’s market share threshold of 30%. Previously only the supplier’s market share was taken into account. The Commission stated that its intention is to protect small and medium competitors of powerful buyers or buyers facing strong countervailing market power of a buyer.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

240 Comparative competition law

and ‘excluded’ restrictions.79 Both these sets of exclusions are essentially intra-brand restrictions and so this harks back to concerns over market integration. If an agreement contains a ‘hardcore restriction’ the whole agreement is excluded from the exemption.80 If the agreement includes an ‘excluded restriction’, that particular restriction is denied exemption.81 The hardcore restrictions are essentially customer and territorial restraints. However they contain a complex web of exceptions that deny brief summary here. Excluded restrictions are broadly concerned with exclusive dealing provisions.82 At the risk of oversimplifying, it can be seen from the above discussion that art 101, with the notable exception of territorial and customer restrictions, is applied by both the Commission and courts in a manner that is generally tolerant towards NPVRs. Where the parties to the agreement have a market share of less than 15 per cent the NPVR will generally be safe from art 101(1) under the Commission’s ‘de minimis’ Notice. If the market share of supplier or buyer is under 30 per cent, the block exemption will normally mean that the exemption provision of art 101(3) will come into play. And, as indicated in the decisions and the Commission’s Vertical Guidelines, even if the block exemption does not apply, the NPVR may nevertheless either fall outside art 101(1) or qualify for exemption if it can be shown that it meets the requirements of art 101(3). 5.3 Article 102 TFEU Article 102 prohibits an ‘abuse by one or more undertakings of dominant position’. Thus, like s 2 of the Sherman Act, it does not require an agreement but extends to unilateral conduct.83 An NPVR may constitute an abuse by a dominant firm and hence be prohibited by art 102. Misuse of market power is analysed in detail elsewhere in this volume84 and so this chapter will consider only its possible application to NPVRs. The 79

See generally Vertical BER (n 3) pt III. The hardcore restrictions are set out in Vertical BER (n 3) art 4. 81 Vertical BER (n 3) art 5. On the severability of excluded restrictions, see paras 70–71. 82 For a convenient summary, see A Jones and B Sufrin, EC Competition Law; Text, Cases and Materials (3rd edn, Oxford University Press 2008) 652–662. 83 Economides and Lianos (n 46) 497–499, argue that art 102 is broader in scope than s 2 of the Sherman Act, with the former more concerned with ‘regulating’ dominant firms. 84 See Chapters 6 and 7. 80

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 241

requirement of the article that an undertaking be ‘dominant’ creates a significant threshold limitation to the application of the article. However, if this threshold can be passed, the case law and the relevant Commission Guidelines suggest that this article is much more likely to be used against NPVRs than is the case with s 2 of the Sherman Act. Indeed, the application of art 102 to NPVRs is one of the more controversial uses of that article. In addition to the concern that NPVRs may threaten market integration,85 there is also a suggestion that art 102 may be concerned not just with the promotion of competition, but also with the protection of competitors.86 Two NPVRs that have created the most controversy have been exclusive dealing and tying/bundling. In relation to the former, the courts have taken a particularly hard line, especially as concerns loyalty rebates or discounts. In Hoffmann-La Roche v Commission87 the European Court of Justice stated that: Obligations of this kind to obtain supplies exclusively from a particular undertaking … are incompatible with the objective of undistorted competition within the Common Market, because … [they] are designed to deprive the purchaser of or restrict his possible choice of sources of supply and to deny other producers access to the market.

This approach has been adopted in subsequent decisions.88 It has been pointed out by commentators that this is close to treating such loyalty rebates as being prohibited per se.89 The greater emphasis on economic reasoning evident in the art 101 decisions discussed above appears not to be matched, at least in this context, in regard to Article 102.

85

Whish (n 12) 702–703, refers to United Brands v Commission [1978] ECR 207; Football World Cup [2000] 4 CMLR 963 and Amminstratizione Autonoma dei Monopoli dello Stato v Commission [1998] 5 CMLR 168 (on appeal [2001] ECR II-3413) as illustrations. 86 Whish (n 12) 191–193. 87 [1979] 3 CMLR 211 para 90. 88 BPB Industries and British Gypsum Ltd v Commission [1993] ECR II-389; Nederlandsche Banden-Industrie Michelin v Commission [1983] ECR 3461; Manufacture Francaise Des Pneumatiques Michelin v Commission [2004] 4 CMLR 923; British Airways v Commission [2003] ECR II-5917; Van den Bergh Foods v Commission [2003] ECR II-4653. 89 D Waelbroeck, ‘Michelin II: A Per Se Rule Against Rebates by Dominant Companies?’ (2005) 1 Journal of Competition Law and Economics 149; Whish (n 12) 195–197.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 23 /

Date: 13/10

JOBNAME: Duns PAGE: 24 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

242 Comparative competition law

In relation to tying and bundling, such case law as there is also suggests a similarly hard-line approach.90 Article 102(2)(d) explicitly refers to ‘making the conclusion of contracts subject to acceptance by the other parties of supplementary obligations which, by their nature or according to commercial usage, have no connection with the subject matter of such contracts’. A leading decision on tying is the Microsoft decision.91 As seen in that case, whether there are two distinct products that can form the basis of a tie is treated as a matter of whether there is a separate consumer demand for the products.92 As with art 101, the Commission is the most likely plaintiff in an art 102 case and for this reason its Guidelines on art 10293 are highly significant. These Guidelines were updated in 2009 and this most recent iteration appears to indicate a shift towards a more economics-focused approach to art 102. The Guidelines place the emphasis on competition rather than on competitors: The emphasis of the Commission’s enforcement activity in relation to exclusionary conduct is on safeguarding the competitive process in the internal market and ensuring that undertakings which hold a dominant position do not exclude their competitors by other means than competing on the merits of the products or services they provide. In doing so the Commission is mindful that what really matters is protecting an effective competitive process and not simply protecting competitors.94

The Guidelines nevertheless go on to state that ‘behaviour that undermines the efforts to achieve an integrated internal market is … liable to infringe’ art 102.95 Significantly, the Guidelines acknowledge that efficiencies may justify conduct and take it outside art 102.96 Specific guidance is provided on how NPVRs are to be assessed.97 90 See, for example, V Korah, ‘The Paucity of Economic Analysis in the EEC Decisions on Competition: Tetra Pak II’ [1993] Current Legal Problems 150. 91 Case T-201/04 Microsoft v Commission (2007). 92 See also Hilti AG v Commission [1990] ECR II-163, upheld on appeal Case 53/92 [1994] ECR I-667; Tetra Pak International SA v Commission [1996] ECR I-5951. For a discussion of the separate products test and a comparison of the US and EU positions on this, see Economides and Lianos (n 46) 518–528. 93 EU Article 82/102 Guidelines (n 5). 94 Ibid 6. 95 Ibid 7. 96 Ibid 28. 97 Ibid paras 32–46 (exclusive dealing) and 47–58 (tying and bundling).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 24 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Non-price restraints 243

6. OTHER JURISDICTIONS Jurisdictions have inevitably been influenced by the US and EU approaches to NPVRs. In some cases, these models have been adopted directly. This is the case, for example, with Malaysia and Singapore. These jurisdictions have modelled their competition provisions squarely on those of the EU and so have anti-competitive agreement and abuse of dominance provisions that, in form at least, will apply to NPVRs as a mirror image of the EU approach.98 Others are more loosely based on these approaches and have introduced local variations. An illustration is the Australian Competition and Consumer Act 2010. The Australian provisions are broadly based on the US antitrust provisions but with significant variations. For example, NPVRs caught under this Act are not confined to ‘agreements’ but apply where there has been a supply (or acquisition) ‘on a condition’ that includes an NPVR.99 They are generally subjected to a rule-of-reason treatment.100 There is also a possibility of exemption from the prohibition, loosely similar to the EU exemption process, on the broad ground that the NPVR creates a ‘public benefit’.101

7. CONCLUSION As this chapter has shown, it is now generally accepted that NPVRs are competitively ambiguous. This implies that, as a matter of competition policy, that NPVRs are appropriately subjected to a lessening of competition test. It is only such a test that will allow the court to conclude whether the restraints are pro- or anti-competitive. While the jurisdictions surveyed in this chapter have generally adopted this approach, it was seen that the treatment of NPVRs is far from consistent. The US has certainly 98 For the anti-competitive agreement provisions and abuse of dominance provisions respectively, see Competition Act 2010 (Malaysia) ss 4, 10 and Competition Act 2006 (Singapore) ss 34, 47. In a curious variation, s 4 of the Malaysian Act refers explicitly to ‘horizontal’ and ‘vertical’ agreements and concerted practices, terms which are defined in s 3. 99 Competition and Consumer Act 2010 (Cth) s 47. A similar approach is taken in s 77 of the Canadian Competition Act RSC, 1985, c C-34. 100 The one anomalous exception is third-line forcing (ss 47(6) and 47(7)). 101 This process is created in Competition and Consumer Act 2010 (Cth) Parts VII and IX of the Act. The concept of ‘public benefit’ is a broad one but efficiencies are one of the most commonly argued benefits: see J Duns, ‘Competition Law and Public Benefits’ (1994) 16 Adelaide Law Review 245.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

244 Comparative competition law

moved away from per se treatment but still generally retains a modified per se treatment for tying conduct. The EU approach towards the treatment of NPVRs has traditionally given priority to the objective of market integration, and the perceived threat that NPVRs might pose to this, over economic analysis. This has resulted in a more complex, hard-line treatment of NPVRs than can be justified in economic terms. Having said that, it does appear, however, that the EU is moving towards a more economics-based treatment, although the pace at which this is occurring has been the subject of criticism by commentators. It is certainly the case that the failure to treat NPVRs consistently, as is currently the case in both jurisdictions, is difficult to justify as a matter of policy. As discussed in Chapter 5 of this volume, there are certainly problems associated with a full rule-of-reason analysis. It can create uncertainty and unpredictability, two traits which are undesirable in commercial transactions. In this regard, the combination of a ‘rule-of-reason’ approach and ‘safe harbours’, such as those created by the EU ‘de minimis’ notice and BER, are arguably the most effective way of tackling NPVRs.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 12_Chapter9

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

10. Vertical conduct: Resale price maintenance Eugène Buttigieg

1. INTRODUCTION There are various forms of resale price maintenance. Once an independent distributor (a wholesaler or a retailer), as opposed to a commercial agent selling on behalf of his principal, purchases a product from a manufacturer for the purpose of resale, as the owner of the product, he is generally free to resell the product to his customers at the price that he deems fit. But a producer may in his contractual relationships with his distributors insert clauses or impose obligations by which he exerts a degree of control over the prices charged by the resellers for these products. This is known as resale price maintenance (RPM), some forms of which give rise to serious competition concerns. One should distinguish between ‘fixed’ RPM and ‘recommended’ RPM. Fixed RPM occurs where the producer actually sets the price that the distributor should charge when reselling his product and the distributor is obliged to comply;1 while recommended RPM occurs where the producer merely recommends the price but the distributor remains free to charge a lower or higher price. Two other forms of RPM are ‘minimum’ RPM and ‘maximum’ RPM where, in the former case, the producer obliges the distributor not to sell below a certain price and, in the latter case, not to 1 Fixed RPM may be accomplished by direct or indirect means: ‘Direct’ RPM is achieved by an agreement or other practice by which the parties agree to fix the retail price of a good or service; ‘indirect’ RPM can be accomplished by other contractual terms that create a disincentive for the retailer to set its own prices, for example by fixing the distribution margin or prohibiting discounts or fixing the maximum permitted level of discount or making promotional support conditional on respecting a specified price or using price reporting and monitoring systems or prohibiting the advertising of discounted prices on the internet. Indirect RPM can also be achieved by suppliers putting pressure on retailers to deter discounting, including by threatening to delist discounters and actually delisting them. For other examples, see European Commission Guidelines on Vertical Restraints [2010] OJ C130/1 para 48.

245

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 2 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

246 Comparative competition law

sell above a certain price so that the distributor retains the freedom to sell above or below the price set by the producer, as the case may be. For years it has generally been perceived, based on scholarly economic studies that recommended (when it is truly recommended and not de facto fixed through the pressure exerted by the manufacturer on the distributors to follow the ‘recommended’ price) and maximum RPM do not in themselves give rise to competition concerns, and that their legality depends entirely on a proper assessment, on a case-by-case basis, of the extent to which the pro-competitive benefits balance out any negative effects on competition. Conversely, for fixed or minimum RPM, any existence of efficiencies deriving from these forms of RPM has been strongly denied so that the standard approach in several jurisdictions has been to equate such RPM with horizontal price fixing and thereby to condemn such agreements as automatically (or ‘per se’) illegal. However, more recent economic thinking has mooted the possibility that such RPM may indeed generate certain efficiencies that have been overlooked to the detriment of consumer welfare. This has led the US Supreme Court to overturn an almost century-old precedent that minimum RPM is per se illegal. This, in turn, has led to antitrust authorities elsewhere seriously considering reviewing their strict approach to RPM. Concurrently, however, other scholars and enforcers maintain that no empirical evidence exists of such beneficial effects and so any relaxing of the strict prohibition against minimum RPM would be ill-advised. This has reignited a debate on the pros and cons of RPM with contrasting positions being taken across Europe and the United States. What the antitrust approach to fixed or minimum RPM should be remains as uncertain and controversial as ever.

2. ANTI- AND PRO-COMPETITIVE EFFECTS2 The two main anti-competitive effects in relation to RPM that are traditionally cited are the reduction of intra-brand price competition, and the resulting risk of a reduction in inter-brand competition deriving from 2

For a detailed critique of the recognised anti-competitive harms of RPM and their potential pro-competitive justifications see M Lao, ‘Resale Price Maintenance: A Reassessment of its Competitive Harms and Benefits’ in J Drexel, WS Grimes, CA Jones, RJR Peritz and ET Swaine (eds), More Common Ground for International Competition Law? (Edward Elgar Publishing 2011) 59, 66–82.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 2 /

Date: 13/10

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 247

increased price transparency that facilitates price collusion (tacit or explicit) at a horizontal level between manufacturers or distributors. Reduced intra-brand competition is said to occur because the distributors will no longer be able to compete on prices, leading to a total elimination of intra-brand price competition; consumers would thereby be deprived of the possibility of lower retail prices that might have occurred had there been price competition. It is presumed that the reduced downward pressure on prices could also have indirect negative effects on inter-brand competition. Moreover, RPM increases price transparency, which in itself could favour successful cartelisation because the monitoring and enforcement of a cartel can be simplified if all cartel members have price-fixing clauses in their distribution contracts. Furthermore, RPM also favours tacit collusion. At a downstream level, it also has the effect of a horizontal pricing agreement between the distributors and this could act as a barrier to entry for discount stores;3 at the upstream level, RPM could reduce any incentive to deviate from a cartel, as the greater market transparency would make a deviation immediately visible.4 When it entails a uniform resale price, RPM prevents efficient price discrimination by resellers; price discrimination can lead to an increase in output and a more efficient recovery of fixed costs. In the words of the European Commission: RPM may restrict competition in a number of ways. Firstly, RPM may facilitate collusion between suppliers by enhancing price transparency on the market, thereby making it easier to detect whether a supplier deviates from the collusive equilibrium by cutting its price. RPM also undermines the incentive for the supplier to cut its price to its distributors, as the fixed resale price will prevent it from benefiting from expanded sales. Such a negative effect is particularly plausible where the market is prone to collusive 3 But for RPM to be used effectively to sustain a retail cartel, the retailers must collectively have some power to induce manufacturers to enter into RPM agreements and to prevent discounted sales to potential rival retailers. They can use the imposition of multiple RPM agreements by an upstream firm, acting as a ‘common agent’, to facilitate downstream price coordination. In effect, RPM in certain circumstances can thus be an effective substitute for horizontal collusive arrangements between retailers (hub-and-spoke cartel). See C Ewald, ‘The Economics of Resale Price Maintenance – Why Europe is Right not to Follow the USA on the Slippery Slope of Leegin’ (2012) 3(3) Journal of European Competition Law & Practice 300. 4 In addition, under RPM retailers can act as ‘common agents’ and enforce upstream collusion by informing manufacturers of breaches of the cartel agreement. Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

248 Comparative competition law outcomes, for instance if the manufacturers form a tight oligopoly, and a significant part of the market is covered by RPM agreements. Second, by eliminating intra-brand price competition, RPM may also facilitate collusion between the buyers, that is, at the distribution level. Strong or well organised distributors may be able to force or convince one or more suppliers to fix their resale price above the competitive level and thereby help them to reach or stabilise a collusive equilibrium. The resulting loss of price competition seems especially problematic when the RPM is inspired by the buyers, whose collective horizontal interests can be expected to work out negatively for consumers. Third, RPM may more generally soften competition between manufacturers and/or between retailers, in particular when manufacturers use the same distributors to distribute their products and RPM is applied by all or many of them. Fourth, the immediate effect of RPM will be that all or certain distributors are prevented from lowering their sales price for that particular brand. In other words, the direct effect of RPM is a price increase. Fifth, RPM may lower the pressure on the margin of the manufacturer, in particular where the manufacturer has a commitment problem, that is, where it has an interest in lowering the price charged to subsequent distributors. In such a situation, the manufacturer may prefer to agree to RPM, so as to help it to commit not to lower the price for subsequent distributors and to reduce the pressure on its own margin. Sixth, RPM may be implemented by a manufacturer with market power to foreclose smaller rivals. The increased margin that RPM may offer distributors, may entice the latter to favour the particular brand over rival brands when advising customers, even where such advice is not in the interest of these customers, or not to sell these rival brands at all. Lastly, RPM may reduce dynamism and innovation at the distribution level. By preventing price competition between different distributors, RPM may prevent more efficient retailers from entering the market or acquiring sufficient scale with low prices. It also may prevent or hinder the entry and expansion of distribution formats based on low prices, such as price discounters.5

However, fixed and minimum RPM could also have certain positive competitive effects. RPM arguably allows suppliers to protect themselves against the risk that retailers lower prices at the expense of quality of service. A related argument is that such vertical price fixing could prevent freeriding by retail price discounters on the pre-sales services and reputation of full-price dealers. One distribution outlet might provide high-quality services (e.g. highly trained salesmen, showroom displays, certification etc)6 on which customers would rely when buying at a

5

Vertical Restraints (n 1) para 224. Certain retailers invest substantial resources in screening the products they carry, ensuring that only high-quality products appear on their shelves. 6

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 249

cheaper discounter that charges lower prices because it does not provide these services. A minimum price would take away the pricing advantage from the discounter and replace intra-brand price competition with competition on services. Thus, minimum RPM could sometimes be economically and commercially justifiable if certain conditions are present such as where the products concerned are new or technically complex and of a reasonably high value and where it is not feasible for the supplier to impose contractually on all the distributors effective service requirements concerning pre-sales services. It might not always be possible to solve the freeriding problem through less restrictive means such as through exclusivity clauses or selective distribution. Another consideration is that RPM protects the image of certain branded products such as luxury goods which would otherwise be less valued by customers. It is also argued that suppliers will often use RPM to convince new operators to join a distribution network; absent RPM new retailers may hesitate joining a distribution network for fear of facing aggressive price competition from incumbent retailers. Moreover, in the publishing sector, where one may often find RPM agreements, usually covering the entire territory of a national market, it is argued that these agreements are indispensable in order to secure the production and distribution of books, in particular non-commercial ones, as they enable publishers and booksellers to engage in cross-subsidy of non-commercial books with the revenue from relatively high prices of commercial ones. It is also claimed that such arrangements make possible a viable network of small specialised bookshops. It is contended that abolishing this type of agreement would have a negative effect on the availability of books in certain regions, on the number of different titles available and on the independence of publishers and book distributors. From this perspective, RPM arrangements are justified by the anthropological goal that the publishing and distribution of books, particularly those which have a special cultural value, should not be endangered for they are in the public interest.

3. THE UNITED STATES In the United States the debate on RPM has been raging on since 1911 when the Supreme Court in Dr Miles ruled that all forms of RPM are ‘injurious to the public interest and void’; thus RPM was held to be per se illegal and presumed not to have any redeeming virtues, meaning that

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

250 Comparative competition law

manufacturers and dealers were not able to justify any such practice.7 This rule was consistently followed by the courts but in the 1930s states started enacting so-called ‘fair trade’ laws to nullify the Dr Miles per se rule, with the intention of protecting small retailers against the pricecutting competition of large chain stores. By 1945, 45 out of the then 48 states allowed the practice of RPM.8 These fair trade laws were, however, all abolished in 19759 when Congress, through the Consumer Goods Pricing Act,10 repealed the Miller-Tydings Fair Trade Act11 and the McGuire Act,12 which permitted individual states to enact fair trade laws and thereby the per se prohibition was fully restored in all states. This, however, did not put an end to the academic debate, which seriously questioned the justification for a per se rule prohibiting RPM. One contribution to this debate was a pioneering article by Telser that appeared in 1960 and which argued that manufacturers impose RPM on their dealers not in order to give the latter monopoly profits, which would be an absurd motivation for a manufacturer, but to induce them to provide the optimum level of pre-sale services.13 Unable to compete on price, dealers would compete with each other in offering successively greater services until their marginal cost curve rose to intersect the price fixed by the manufacturer. Therefore, it was claimed that simply by altering the price, the manufacturer could automatically vary the level of dealer pre-sale services provided. However, this article did not dent the rigid approach taken by the courts. In fact, in the decade following its publication the per se rule was even extended to vertical non-price restraints when in Schwinn14 the Court ruled that once the manufacturer has parted with title and risk his 7

Dr Miles Medical Co v John D Park & Sons Co 220 US 373, 408 (1911). In justifying its decision, the Court focused on two problems: the existence of a restraint on alienation and how this restriction would reduce the freedom of distributors; concerns that later critics pointed out were not, strictly speaking, competition related. 8 Schwegmann Bros. v Calvert Distillers Corp., 341 US 384, 398 (1951). 9 In 1975, 36 states still had fair trade laws allowing minimum RPM so that it was outlawed only in 14 of the states – per Justice Breyer’s dissenting opinion in Leegin Creative Leather Products, Inc. v PSKS, Inc. 127 S Ct 2705 (2007). 10 89 Stat 801. 11 50 Stat 693. 12 66 Stat 631. 13 LG Telser, ‘Why Should Manufacturers Want Fair Trade?’ (1960) 3 Journal of Law & Economics 86. 14 United States v Arnold, Schwinn & Co 388 US 365, 87 S Ct 1856 (1967). In US v Bausch & Lomb Optical Co 321 US 707 (1944) the Court had already

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 251

effort thereafter to restrict territory or customers is a per se violation of s 1 of the Sherman Act.15 It was only 10 years later, following heavy criticism especially by economists and exponents of the Chicago School, who identified the potential and real benefits of vertical restraints and showed that the adverse effects of vertical restraints are more theoretical than real, that the Court overturned Schwinn when it ruled in Sylvania16 that vertical non-price restraints should be considered under a rule-of-reason approach rather than the rigid per se approach. In doing so, the Court recognised that non-price vertical restraints may very well generate offsetting efficiencies such as more efficient distribution, protection against freeriders and lower costs. The Court recognised that distribution restrictions could achieve efficiencies by promoting retailer and distributor investments in promotional activities and quality controls. It noted that such restrictions could be used by manufacturers to penetrate new markets, to protect full-priced dealers from freeriders who otherwise would destroy any incentives for dealer investment in promotion and services and to assure product quality and customer safety. The Court acknowledged that held that vertical territorial and customer restraints are per se illegal if they are an integral part of an agreement to fix prices. 15 Posner explains that though Telser had already published his article at the time of the Schwinn judgment, this article only started winning supporters after the Schwinn case and that is why it had no influence on that decision. Telser’s analysis was not immediately influential because many economists at the time viewed the pre-sale services encouraged by RPM as of dubious value to consumers. These services were thought to be forms of advertising and economists in this period regarded advertising as a socially wasteful activity that served only to reduce the elasticity of demand for advertised products. In particular, it was thought that heavy advertising convinced the consuming public, without any basis in real differences among brands of the same product, that less heavily advertised brands were not good substitutes for more heavily advertised ones: RA Posner, ‘The Rule of Reason and the Economic Approach: Reflections on the Sylvania Decision’ (1977) 45 University of Chicago Law Review 1. Telser himself tried to rebut this perception of advertising in another article by observing that advertising is a weapon of competition rather than of monopoly as studies have shown that heavy advertising is associated with unstable brand preferences; intensive advertising does not cement consumer loyalty to the advertised brand but on the contrary is the recourse of sellers who have difficulty holding on to their customers: L Telser, ‘Advertising and Competition’ (1964) 72 Journal of Political Economy 537. 16 Sylvania 433 US 36, 97 S Ct 2549 (1977). Posner observes that by this time economists had started viewing advertising as a source of genuine consumer information: Posner (n 15).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 7 /

Date: 13/10

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

252 Comparative competition law

depending on the scope of the restrictions, intra-brand competition would be stifled but that this could, however, be outweighed by benefits to economic efficiency and inter-brand competition. In particular, it ruled that a mere showing of an adverse intra-brand impact, which is a necessary result of every successful vertical restraint, is not enough. Under the Sylvania approach, it is necessary to consider whether the arrangement is likely to have a ‘pernicious effect’ on inter-brand competition and whether the restraint has ‘redeeming virtues’. Such a pernicious effect is manifest if the vertical restraint reduces output and raises the price of the product significantly above the competitive level in the industry as a whole. However, the per se prohibition against RPM remained because the Supreme Court in Sylvania reiterated that resale price agreements remained void under the rule of Dr Miles. It was deemed that in order to be consistent with the per se approach taken to price fixing at the horizontal level, vertical RPM also had to fall under the per se category. Three years later, in another case, the Court confirmed the per se illegality of RPM.17 Gellhorn and Tatham were critical of this approach, asserting that according to Liebler all vertical restraints, whether price or non-price, have similar price effects and cannot be distinguished on economic or other analytical grounds. They maintained that the Court confused the issue by condemning all direct price fixing. Indeed they claimed that while courts following Sylvania maintained the per se prohibition of vertical price fixing, they sometimes employed artificial distinctions to avoid per se liability when the challenged practice appeared to be clearly desirable.18 Indeed in Monsanto the Supreme Court required substantial proof before the per se rule could be applied to an RPM scheme.19 Posner was likewise critical of this approach and contended that Sylvania should have applied the same reasoning regarding the freerider problem to RPM, which, as shown above, could be used for exactly the same purpose as non-price restraints. It might be that the manufacturer, because of the nature of the product, cannot impose territorial restrictions to give the optimal level of pre-sale service. Price restrictions may be the only (or the most efficient) way of stimulating provision of such service. Indeed, Telser’s original argument, which helped trigger an awareness of 17 California Retail Liquor Dealers Association v Midcal Aluminium 445 US 97, 100 S Ct 937 (1980). 18 E Gellhorn and T Tatham, ‘Making Sense out of the Rule of Reason’ (1984–85) 35 Case Western Reserve Law Review 155. 19 Monsanto v Spray Rite Serv Corp 465 US 752 (1984).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 253

the potential efficiency of certain vertical restraints, was based precisely on RPM.20 One of the reasons that the Court offered for distinguishing between price and non-price restraints was that RPM can be used to bolster collusion at the manufacturer level: if colluding manufacturers have fixed the retail as well as the wholesale price, then one of the manufacturers who is cheating his associates by undercutting the cartel price cannot pretend that his market share is increasing only because the retailers who carry his product have decided for their own reasons to accept a lower mark-up over the wholesale price of his product. However, Posner rejects this argument by saying that this collusion-enhancing effect of RPM depends on each retailer’s handling only one manufacturer’s brand of the product in question, which is exceptional rather than the normal RPM situation. The retailer who handles the brands of several manufacturers will promote the brand of the manufacturer who offers him the lowest wholesale price – the cartelist who is cheating. The presence of RPM will not make this type of cheating any easier to detect. Furthermore, even non-price restrictions such as exclusive territories could be potential methods of shoring up a manufacturers’ cartel, so there is nothing to distinguish the two. Non-price restraints could prevent a cheater from arguing that any increase in his market share above the quota assigned to him by the cartel was due not to his price cutting but to competition among his dealers. Indeed, like RPM, they would be effective in shoring up cartels only in exclusive dealing situations and like RPM their effectiveness in shoring up cartels would be similarly limited. So, in short, the possible role of restricted distribution in facilitating collusion at the manufacturer’s level does not distinguish price from non-price restrictions. Posner also argues that it is nonsensical to look favourably at vertically imposed territorial restraints that affect both price and service competition and then continue to consider as per se illegal RPM that only affects price competition.21 Forbidding a dealer or distributor from selling outside of its territory when it is the only distributor or dealer of the manufacturer’s brand in the territory has a greater adverse effect on intra-brand competition than fixing the price at which it may resell the product. Market division and price fixing have always been treated 20 Posner (n 15) and id, ‘The Next Step in the Antitrust Treatment of Restricted Distribution: Per Se Legality’ (1981) 48 University of Chicago Law Review 6. 21 Posner, ‘The Next Step in the Antitrust Treatment of Restricted Distribution’ (n 20).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

254 Comparative competition law

symmetrically in the horizontal context and so they should be so treated in the vertical context as well. Posner therefore concludes that Sylvania had no reason to retain Dr Miles as far as RPM is concerned but should have laid down the rule for all vertical restraints, be they price or non-price restraints. It was only in Khan,22 20 years later, that the Court started changing its approach towards RPM by permitting manufacturers to agree with retailers on the maximum prices that retailers may charge, thereby reversing Kiefer-Stewart Co v Joseph E Seagram & Sons23 and Albrecht v Herald Co24 where it had held that maximum RPM was per se illegal. Stressing the importance of inter-brand competition and low prices to consumers it held: ‘[We] find it difficult to maintain that verticallyimposed maximum prices could harm consumers or competition to the extent necessary to justify their per se invalidation’. Henceforth, the per se rule at vertical level was reserved solely to minimum RPM, though it remained permissible for suppliers, under the so-called Colgate doctrine, to recommend prices and to unilaterally refuse to deal with those who do not adhere to the recommended prices25 and of course to impose a minimum price where the retailers are their agents.26 The change in approach was completed eight years ago in 2007 when in Leegin27 the Supreme Court with a narrow majority of five to four finally completed what Monsanto and Khan had started; it entirely overruled Dr Miles, holding that all vertical price restraints are to be judged according to the rule of reason because resale price agreements are not so clearly anti-competitive that they should be deemed per se illegal. The Court noted that the ‘economics literature is replete with 22

State Oil v Khan 522 US 3 (1997). For a detailed commentary on this case see GE Bamberger, ‘Revisiting Maximum Resale Price Maintenance: State Oil v Khan (1997)’ in JE Kwoka and LJ White (eds), The Antitrust Revolution: Economics, Competition, and Policy (4th edn, Oxford University Press 2004). 23 340 US 211, 71 S Ct 259 (1951). 24 390 US 145, 88 S Ct 869 (1968). 25 US v Colgate & Co 250 US 300 (1919) (hereafter ‘Colgate’), which was reaffirmed by the Supreme Court in Monsanto v Spray Rite Serv Corp where the Court held that ‘independent action is not proscribed. A manufacturer, of course, generally has a right to deal, or refuse to deal, with whomever it likes, as long as it does so independently’: 465 US 752 (1984). Under Colgate, the manufacturer can announce its resale prices in advance and refuse to deal with those who fail to comply. And a distributor is free to acquiesce in the manufacturer’s demand in order to avoid termination’: Colgate 250 US 300 (1919) 761. 26 US v General Electric Co 272 US 476 (1926). 27 Leegin v PSKS (n 9).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 255

pro-competitive justifications for a manufacturer’s use of resale price maintenance’,28 noting that ‘the justifications for vertical price restraints are similar to those for other vertical restraints’,29 though also warning that ‘potential anti-competitive consequences of vertical price restraints must not be ignored or underestimated’.30 However, this has not put an end to the debate on the legality of RPM. On the one hand, the Federal Trade Commission (FTC) has followed Leegin and brought federal enforcement of RPM into line with the ruling.31 In Nine West the FTC,32 applying the rule of reason, decided to allow agreements under which a manufacturer imposed prices on retailers. This was because the manufacturer lacked market power so that any attempt on its part to charge supra-competitive prices over a sustained period of time would be rendered unsuccessful by inter-brand competition. Secondly, because the manufacturer itself was the driving force behind the RPM scheme, rather than a dominant retailer or a retailer cartel, the FTC accepted that the scheme was intended to provide an incentive to retailers to offer more or better services to consumers. Moreover, the FTC stated that even if a manufacturer has market power or the RPM policy originated with its retailers, the manufacturer may still justify its RPM policy ‘by presenting evidence that while the practice might increase resale prices for its products over what they would otherwise be, it enhances output. That might suggest that consumers place a higher value on non-price factors (such as service) than they do on price, so that the practice may be viewed as efficiency-enhancing.’33 On the other hand, state attorneys general remain largely opposed to the Leegin ruling and wish to maintain the per se prohibition under state law.34 Indeed, in 2009 the state of Maryland amended its Maryland 28

Ibid. Ibid. 30 Ibid [2]. 31 The FTC had not been unanimous in its support for the DOJ (Department of Justice) brief urging the overruling of the per se precedent in Leegin as two of the FTC Commissioners had dissented from the FTC decision to support the DOJ brief: see open letter of Commissioner Pamela Jones Harbour to the Supreme Court (26 February 2007) . 32 In the Matter of Nine West Group, Inc. File No. 981 0386, Docket No C-3937 (2008) (hereafter ‘Nine West’). 33 Ibid. 34 Ibid: RD Blair and JS Haynes, ‘Leegin, The Political Backlash’ (2010) 2 The CPI Antitrust Journal 2, 4. 29

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

256 Comparative competition law

Antitrust Act to confirm that, despite Leegin, minimum RPM remains per se illegal under Maryland law.35 Although, so far, Maryland remains the only state to have legislated to expressly exclude the application of Leegin to state law, it is claimed that ‘New York and California are notably Leegin-unfriendly, and 13 states reportedly still consider RPM a per se violation’.36 This state of ambivalence is also reflected in the state court rulings post-Leegin. Two lower courts applying state antitrust laws have ruled that Leegin applies also to state law but one of the lower courts was then overruled by the state Supreme Court. In Spahr v Leegin Creative Leather Products,37 the Federal District Court for the Eastern District of Tennessee considered that since there had not been any pre-Leegin state court rulings confirming that the Tennessee Trade Practice Act laid down a per se prohibition for minimum RPM, there was no good reason to believe that Tennessee courts should not be able to follow Leegin. For this reason the Court applied a rule-of-reason analysis to the claim and held that the plaintiff had not adequately shown anti-competitive effects. The appeal was dismissed. A similar finding was made at trial in O’Brien v Leegin Creative Leather Products Inc.38 However the decision was overturned in 2012 on appeal to the Kansas Supreme Court.39 After reiterating previous holdings confirming the position that Kansas is not required to follow federal antitrust precedent, the Court, relying on a ‘clear statutory language’ interpretation of the Kansas Restraint of Trade Act, held that vertical price fixing is per se illegal in the State of Kansas. This made Kansas the first state whose highest court came down against the rule-of-reason

35 A provision that states that ‘a contract, combination, or conspiracy that establishes a minimum price below which a retailer, wholesaler, or distributor may not sell a commodity or service is an unreasonable restraint of trade or commerce’ was introduced: Md Code Ann, Com Law Section 11-204 (new provisions in force from 1 October 2009). 36 JL Himes and M Falk, ‘Just What the Doctor Ordered, A Second Opinion for Vertical Price Fixing’ (2010) 2 CPI Antitrust Journal 5. 37 No 07-CV-187, 2008 WL 3914461 (E D Tenn August 20, 2008), appeal dismissed, File No 08-6165 (6th Cir Nov 20, 2008). 38 O’Brien v Leegin Creative Leather Prods. Inc., No 04-CV-1668 (8th Judicial Dist, Sedgwick County Kan July 9, 2008), direct appeal to Supreme Court granted, File No 101,000 (Oct 6, 2008). 39 277 P 3d 1062 (Kan 2012).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 12 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 257

approach for RPM post-Leegin.40 The Kansas legislature responded by amending the Act in April 2013 to substitute the per se standard of review with a ‘rule of reasonableness’ standard as far as resale price agreements are concerned. The amending legislation also, with some exceptions, provided that the Act should be ‘construed in harmony’ with federal antitrust law and precedents.41 At a point in time, to further compound the legal uncertainty facing the business community in the US, there was even before Congress a Bill entitled Discount Pricing Consumer Protection Act intended to correct the Supreme Court’s mistaken interpretation of the Sherman Act in the Leegin decision and ‘to restore the rule that agreements between manufacturers and retailers, distributors, or wholesalers to set the minimum price below which the manufacturer’s product or service cannot be sold violates the Sherman Act’.42 This Act was never enacted; although approved by the Judiciary Committees of both congressional bodies it was never voted on by the full House and Senate. As many as 38 state attorneys general joined by attorneys general from three territories had signed a letter that was sent to Congress on 17 October 2009 expressing their support for this Bill, arguing that ‘[w]ith the Leegin decision now two years behind us, there remains no evidence that consumers are provided any tangible benefits, let alone benefits that outweigh the higher prices that result from minimum resale price fixing’.43 Indeed, the latter argument was also the one that formed the basis of Justice Breyer’s dissenting opinion in Leegin. Justice Breyer was not convinced that there was enough empirical evidence supporting the claim that RPM may be pro-competitive to any significant degree to justify overturning a decision that had survived for close to a century. While public enforcers in the US seem to be divided on the question of the RPM’s legality, commentators are generally more uniform in their approach, hailing the rule-of-reason analysis as the only plausible 40

ML Fessinger, ‘A Century Behind? The Kansas Supreme Court Opts Out of the Rule of Reason in O’Brien v. Leegin Creative Leather Prods., Inc. [277 P.3d 1062 (Kan. 2012)]’ (2012–2013) 52 Washburn Law Journal 323. 41 An Act concerning the Kansas Restraint of Trade Act; amending KSA. 50-101, 50-112, 50-158 and 50-161 and repealing the existing sections; also repealing KSA 50-108 and 50-115 Senate Bill [124] . 42 Discount Pricing Consumer Protection Act (2009) US s 2(b) (2). 43 Letter from the National Association of Attorneys General to the Senate (27 October 2009) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

258 Comparative competition law

approach to the RPM conundrum. Since so far the only competition harm attributed to RPM by the Supreme Court and economists is that it may be used to support a manufacturers’ or retailers’ cartel, Blair and Haynes argue that ‘RPM has no independent competitive significance since its purpose and effect (if successful) is to support an already anticompetitive horizontal agreement on price’ and so ‘the appropriate antitrust remedy is to attack the horizontal cartel rather than the vertical price agreement’.44 Consequently, since RPM may undoubtedly also be used as a promotional device as in Leegin, they argue that the best antitrust approach to RPM is the rule-of-reason approach.

4. THE EUROPEAN UNION The position under EU law has consistently been one that, theoretically at least, is comparable to the rule-of-reason approach. 4.1 Maximum and Recommended Resale Prices As far as maximum and recommended resale prices are concerned, these have always been allowed provided they do not de facto amount to a minimum or fixed price as a result of pressure or incentives.45 Thus, in JCB Service the General Court annulled that part of the Commission decision where an infringement of Article 101 of the Treaty on the Functioning of the European Union (TFEU) had been found due to the ‘fixing of discounts or resale prices’ because it held that there was no unequivocal evidence that the distributors were ‘subject to a strict body of rules on retail prices’46 or that JCB’s actions involved coercion even though the Court acknowledged that ‘the influence of JCB on retail sale prices was … significant’.47 Indeed, recommended and maximum resale 44

Blair and Haynes (n 34) 2–3. In Lubricantes the Court of Justice confirmed that contractual clauses are eligible for block exemption when they do no more than impose a recommended or maximum resale price and when, therefore, it is ‘genuinely possible for the reseller to determine that retail price’; conversely, such clauses are ineligible ‘where they lead, directly or by indirect or concealed means, to the fixing of a retail price or the imposition of a minimum sale price by the supplier’: Case C-506/07 Lubricantes y Carburantes Galaicos v GALP Energia España [2009] ECR I-00134 para 56. 46 Case T-67/01 JCB Service v Commission [2004] ECR II-49 para 128 (hereafter ‘JCB Service’). 47 Ibid paras 130 and 133. 45

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 259

price practices are block exempted under the current block exemption regulation48 where the market share of each of the parties to the agreement does not exceed 30 per cent. Moreover, where this market share threshold is exceeded and so the parties may not benefit from the block exemption, the maximum and recommended prices would only give rise to competition concerns where they work as a focal point for the resellers and might be followed by most or all of them or where the maximum or recommended prices soften competition or facilitate collusion between suppliers. An important factor for assessing such possible anti-competitive effects is the market position of the supplier. The stronger the market position of the supplier, the higher the risk that a maximum resale price or a recommended resale price leads to a more or less uniform application of that price level by the resellers, as they may use it as a focal point. They may find it difficult to deviate from what they perceive to be the preferred resale price proposed by such an important supplier on the market. Even in such circumstances, maximum resale prices may be exempted under Article 101(3) TFEU (conferring legality on restrictive practices that generate verifiable efficiency gains that result in consumer benefit when such restrictions are indispensable for the attainment of these benefits and do not eliminate competition)49 if they are useful in avoiding double marginalisation50 or may help to 48

Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1. 49 See, however, Gyselen who argues at 161–163 that the balancing exercise carried out under Art 101(3) TFEU is different from the American rule of reason. He asserts that under the former it is more taxing for the manufacturer to prove that pro-competitive effects outweigh negative effects than under the US rule-ofreason analysis because the Commission in its assessment under Art 101(3) tends to give more weight to the negative effects and the burden of proof on the parties claiming the benefit of Art 101(3) is heavier as apart from having to show that the restrictions generate efficiencies the onus is on them to also show that there are no less restrictive alternatives to achieve these benefits and that the anti-competitive effects are not substantial: L Gyselen, ‘Resale Price Maintenance: Growing Convergence between the US and the EC in Sight?’ in M Bulterman, L Hancher, A McDonnell and H Sevenster (eds), Views of European Law from the Mountain: Liber Amicorum Piet Jan Slot (Kluwer 2009). 50 Double marginalisation occurs when both the manufacturer and retailer have market power at their respective levels and they apply their own mark-up in prices, resulting in the possibility that supra-competitive profits are made twice. But this also inevitably leads to the sub-optimally low quantities of sale associated with high prices. A manufacturer will generally be keen on as much competition as possible on the downstream level as the lower the mark-up on that

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

260 Comparative competition law

ensure that the brand in question competes more forcefully with other brands, including own-label products, distributed by the same distributor.51 4.2 Fixed and Minimum Resale Prices Although fixed and minimum RPM practices have traditionally been considered to fall within the category of agreements or concerted practices whose very object is anti-competitive so that their harmful effect on competition may be presumed and need not be proved for a finding of illegality in terms of the first paragraph of Article 101 TFEU,52 level, the larger the sales and therefore also the profits of the manufacturer. So both undertakings and consumers would be better off by internalising the respective pecuniary externalities as prices would decrease and quantities, as well as profits, would increase. RPM can solve the double marginalisation problem as the manufacturer can set the final retail price under RPM that an integrated firm (automatically internalising the externalities) would set. As the double marginalisation problem is driven by the incentive to add a mark-up, this result can, however, also be achieved by a maximum price, which would allow firms to offer lower prices. 51 Vertical Restraints [2010] (n 1) paras 226–229. 52 Case 243/83 SA Binon & Cie v SA Agence et messageries de la presse [1985] ECR 2015, para 44 and Commission Guidelines on the Application of Article 81(3) of the Treaty [2004] OJ C101/97 para 23. See also R Whish, Competition Law (6th edn, Oxford University Press 2009) 119 and A Jones and B Sufrin, Text, Cases and Materials: EC Competition Law (3rd edn, Oxford University Press 2008) 226. Of course, as in the US (see text to n 26), it is permissible for a manufacturer to fix prices where the retailer is an agent (Vertical Restraints [2010] (n 1), para 18), though there would still be an infringement (but a horizontal rather than a vertical concerted practice) if producers colluded to enter into agency agreements with price-restricting clauses in lieu of wholesale agreements with the objective of raising retail prices or preventing the emergence of lower retail prices. See Commission Commitment Decision Case COMP/AT.39847-E-Books where the Commission at para 87 concluded that such a concerted practice ‘by its very nature’ has the potential to restrict competition. Moreover, comparable with the US Colgate doctrine (see text to n 25), the manufacturer may unilaterally impose prices on its retailers because for Article 101 TFEU to bite there must be an ‘agreement’. But see D Waelbroeck, ‘Vertical Agreements: 4 Years of Liberalisation by Regulation N. 2790/99 after 40 Years of Legal (Block) Regulation’ in H Ullrich (ed), The Evolution of European Competition Law: Whose Regulation, Which Competition? (Edward Elgar 2006) who, comparing Monsanto (n 19), where the US Supreme Court said that ‘a distributor is free to acquiesce in the manufacturer’s demand in order to avoid termination’ without this giving rise to illegal resale

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 261

it has always been possible for the parties to the agreement/practice to argue that such an arrangement is exempt under Article 101(3). This was established by the Court of Justice in Matra Hachette v Commission when it stated that ‘in principle, no anti-competitive practice can exist which, whatever the extent of its effects on a given market, cannot be exempted, provided that all the conditions laid down in Article 85(3) [now 101(3)] of the Treaty are satisfied’.53 However, the European Commission has, from its very first block exemption regulation concerning vertical agreements, blacklisted fixed and minimum RPM as hardcore restraints, irrespective of the market power of the parties concerned. This effectively lays down a presumption that these restrictive practices will be deemed not to generate any or sufficient efficiencies to offset their negative effects, unless on a case-bycase basis the RPM practice can be positively shown to satisfy Article 101(3).54 So, as observed by Giovannetti and Stallibrass, ‘in addition to the presumption of negative effects, there is also a presumption that the price maintenance with the ruling in Case T-208/01 Volkswagen AG v Commission [2003] ECR II-5141, where the General Court citing case law held that the existence of an agreement could be deduced from the conduct of the parties concerned and that measures would only be apparently unilateral and thereby still illegal under Article 101 TFEU when they receive the ‘tacit acquiescence’ of the dealers and that tacit acquiescence can be inferred from implementation in practice of the manufacturer’s initiative, argues that ‘it is clear that there is no equivalent to the Colgate doctrine in [EU] Competition law’. 53 Case T-17/93 Matra Hachette SA v Commission [1994] ECR 595, para 85. 54 The former Commission Regulation (EEC) 1983/83 on the application of Article 85(3) of the Treaty to categories of exclusive distribution agreements [1983] OJ L173/1, Recital 8; Commission Regulation (EEC) 1984/83 on the application of Article 85(3) of the Treaty to categories of exclusive purchasing agreements [1983] OJ L173/5, Recital 8; and Commission Regulation (EEC) 4087/88 on the application of Article 85(3) of the Treaty to categories of franchise agreements [1988] OJ L359/46, Art 5(e), which excluded all forms of RPM (other than recommended prices in the case of franchise agreements) from the respective block exemptions. All three block exemption regulations were replaced by Commission Regulation (EC) 2790/1999 on the application of Article 81(3) of the Treaty to categories of vertical agreements and concerted practices [1999] OJ L336/21 which, though exempting maximum RPM and the practice of supplier-recommended resale prices, still blacklisted fixed and minimum RPM in Art 4(a). The latter block exemption regulation was in turn replaced by Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1 but this maintained the blacklisting of fixed and minimum RPM in Art 4(a).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 17 /

Date: 8/9

JOBNAME: Duns PAGE: 18 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

262 Comparative competition law

agreement will not fulfil the conditions of Article 81(3) EC [101(3) TFEU]’.55 That said, as Peeperkorn, a DG Competition official, observes, the presumptions, though strong, are rebuttable.56 The question is, how easily rebuttable are these presumptions in the eyes of the European Commission so as to demonstrate an approach that is truly different from the per se illegality approach followed in the US prior to Leegin? The Commission’s past decisional practice shows that there has not been one instance where the European Commission found a fixed or minimum RPM to satisfy the conditions laid down in Article 101(3) TFEU; the Commission invariably concluding that the RPM arrangement concerned failed to generate the kind or degree of efficiencies necessary to outweigh the perceived anti-competitive effects.57 Indeed, Peeperkorn notes that ‘it seems that in particular the efficiency arguments mentioned in support of RPM are not very strong and that RPM is not an efficient instrument for bringing about these efficiencies’.58 It was pointed out above that, in the publishing sector, it is argued that RPM agreements are indispensable in order to secure the production and distribution of books, in particular non-commercial ones (an issue discussed earlier). In VBBB/VBVB59 the publishing associations concerned challenged the Commission’s refusal to grant an exemption to their RPM agreements precisely on the ground that it had failed to take account of the ‘special market structure’ of this sector and the ‘special nature of books’. The Commission had held that forms of non-price competition such as stocking, specialisation, services offered and ordering facilities were secondary to price competition – such non-price competition mattered only in the case of highly specialised books in 55

E Giovannetti and D Stallibrass, ‘Three Cases in Search of a Theory: Resale Price Maintenance in the UK’ (2009) 5 European Competition Journal 641, 643. See also L Peeperkorn, ‘Resale Price Maintenance and Its Alleged Efficiencies’ (2008) 4 European Competition Journal 201, 203. 56 Peeperkorn (n 55) 212. 57 See also Commission Guidelines on the Application of Article 81(3) of the Treaty [2004] OJ C101/97, paras 46, 79. 58 Peeperkorn (n 55) 212. Jones likewise notes that ‘the reduction of intra-brand competition and the increased transparency on prices that results from [fixed and minimum RPM] reduces downward pressure on the price of particular products and has the indirect effect of reducing inter-brand competition. This has led the Commission to find that RPM provisions generally violate Article 81(1) and do not satisfy the Article 81(3) conditions’: A Jones, ‘Resale Price Maintenance: A Debate About Competition Policy in Europe?’ (2009) 5 European Competition Journal 479, 500 (emphasis added). 59 Cases 43 and 63/82 VBVB and VBBB v Commission [1984] ECR 19.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 18 /

Date: 8/9

JOBNAME: Duns PAGE: 19 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 263

which it assumed that the large majority of consumers were not interested. It considered that ‘the price of the product is, for many consumers an important element in their decision to purchase – and this applies to books as much as to other products’. It also held that agreements on collective RPM denied consumers freedom of choice.60 To put it simply, the Court found against the associations, ruling that the special features of the market could not justify a restrictive system whose effect was to deprive distributors of all freedom of action as regards the fixing of the selling price. Moreover, the Advocate General specifically addressed the question of whether cultural considerations too may justify exemption under Article 101(3) in the following clear terms: ‘there can … never be any question of an absolute priority of cultural over economic considerations’.61 More recently, in the Sammelrevers case, the Commission cleared the German book price-fixing system on the grounds that in terms of the undertaking submitted by the parties to the agreement, there was no breach of Article 101. However, this was because the arrangements did not appreciably affect trade between the Member States. In commenting about this case, former Competition Commissioner Monti had stated that provided that a national book price-fixing system did not affect inter-state trade appreciably (in casu by not imposing fixed book prices on direct cross-border sales), the Commission would be willing to clear it, taking into account also the national interest in maintaining these systems, which are aimed at preserving cultural and linguistic diversity in Europe.62 However, Emmerich, while discussing the German national book price-fixing arrangement, questioned whether the requirements of Article 101(3) were met in such circumstances. This is because it is difficult to see how consumers could adequately participate in the advantages generated by the conduct given that book price maintenance

60

VBBB/VBVB [1982] OJ L54/36. VBVB and VBBB v Commission (n 59) 89. 62 Press Release IP/02/461 of 22/3/2002 (an unofficial English translation of the parties’ Undertaking can be viewed in (2002) 2 Competition Policy Newsletter 37). For criticism of this decision see G Monti, ‘Article 81 EC and Public Policy’ (2002) 39 Competition Market Law Review 1057, 1085 and G Monti, EC Competition Law (Cambridge University Press 2007) 104–105 who, after denouncing the Commission’s approach and the Commissioner’s statement in this case as problematic, remarks that ‘it seems that the Community’s policy is an unhelpful attempt to carve out a compromise without confronting the difficult question about how to balance competition and culture in the books sector’. 61

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

264 Comparative competition law

is aimed at granting a sufficient margin to the book traders and thereby transferring wealth from the consumer by fixing exorbitant prices.63 Moreover, as recently confirmed by the current Competition Commissioner Almunia, one should not read too much into this decision, as the Commission did not take a definitive view in this case. When asked by a Belgian MEP (Member of the European Parliament) whether the Commission would approve a Polish publishers’ plan to fix minimum prices for books in line with the approval granted to German publishers in Sammelrevers, Almunia replied as follows: Industry-wide agreements or legislation may attempt to set the retail price for books. That said, whether a law or an agreement is contrary to EU competition rules depends on a range of factual, legal and economic elements. In the 2002 German Sammelrevers case, the Commission examined but did not take a definitive view regarding the agreements between publishers and booksellers. It concluded on the facts of the case that, in light of the undertaking given by the parties involved that guaranteed the freedom of direct cross-border selling of German books to final consumers in Germany, the relevant agreements did not have an appreciable effect on trade between Member States, and therefore fell outside EU competition rules.64

4.3 Post-Leegin The Commission, perhaps influenced by the Leegin judgment, has indicated a willingness to adopt a more lenient approach to fixed and minimum RPM. Although it has retained their exclusion from the current block exemption on vertical restraints,65 in the revised accompanying guidelines the Commission highlights the potential efficiencies that such arrangements may generate and professes its willingness to apply a more rigorous assessment of the alleged negative effects when an efficiencies defence is raised.66 The former guidelines had merely reiterated that such arrangements were hardcore restrictions without acknowledging that 63

V Emmerich, ‘The Law on the National Book Price Maintenance’ [2001] European Business Organization Law Review 553. 64 Answer by Mr Almunia on behalf of the Commission to a Parliamentary question by MEP Marc Tarabella (24 October 2013). 65 Commission Regulation (EU) 330/2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices [2010] OJ L102/1, art 4 (a). 66 Vertical Restraints (n 1) paras 223 and 225. Of course, the guidelines still, and indeed in a more extensive manner than the previous guidelines, highlight the specific competition concerns attached to fixed and minimum RPM: Vertical Restraints (n 1) para 224. See also text to n 5.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 20 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 265

there may be justifications or offsetting efficiencies that are specific to such arrangements.67 The revised guidelines now make it clear that, while a fixed or minimum RPM creates the presumption that the agreement restricts competition in violation of Article 101(1) and does not satisfy the conditions for exemption under Article 101(3), in those instances where the parties substantiate their claim that efficiencies will likely result from the inclusion of RPM in their agreement and demonstrate that all the conditions of Article 101(3) are fulfilled, the Commission is required to effectively assess – and not just presume – the likely negative effects on competition and consumers before deciding whether the conditions of Article 101(3) are fulfilled.68 The revised guidelines recognise, in particular, three types of efficiencies specific to RPM especially when the RPM is supplier driven. Firstly, in the case of new product launches, RPM might help during the introductory period of expanding demand to induce and enable distributors to increase promotional efforts and expand overall demand for the product so as to make the launch a success. This will be beneficial to the manufacturer and consumers and to the competitive process more generally. Secondly, fixed resale prices may be necessary for the running of a coordinated short-term low-price campaign (two to six weeks) in franchise or other similar distribution systems, again to the benefit of consumers. Finally, the extra margin provided by RPM may also be useful to allow retailers to provide (additional) pre-sales services beneficial to consumers, in particular in the case of experience or complex products, while concurrently providing the means and the incentive to prevent freeriding among retailers on these services as this would dampen these promotional efforts.69 67

Commission Guidelines on Vertical Restraints [2000] OJ C291/1, para 47. Commission Guidelines on Vertical Restraints [2010] OJ C130/1, (n 1) para 223. 69 Ibid para 225. In practice, however, proving any of these efficiencies may be quite difficult because as Geradin, Layne-Farrar and Petit point out: ‘… in practice, the flexibility introduced in the Guidelines could prove ineffectual. The majority of the efficiency benefits arising from RPM are “qualitative” in nature. The various positive effects on distributors’ incentives are thus not easily amenable to economic quantification as generally required under Article 101(3) TFEU. In turn, it will often prove complex to balance the harmful quantitative effects of RPM against their qualitative benefits under Article 101(3). As a result, the balancing exercise will inevitably hinge on a value judgment – which by its nature is variable, imprecise, and subjective’: D Geradin, A Layne-Farrar and N Petit, EU Competition Law and Economics (Oxford University Press 2012) 479. 68

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

266 Comparative competition law

However, this more positive and effects-based approach to RPM does not seem to be shared by the national competition authorities and national courts of all the EU Member States when applying their domestic and EU antitrust provisions. This has led to legal uncertainty in Europe as in the US. For instance, three fairly recent decisions of the German Bundeskartellamt in Germany condemning minimum RPM show that German authorities continue to take a hostile and rigid approach to RPM.70 Caspary notes that in one of these cases, CIBA Vision Vertriebs GmbH, Großostheim, the German Federal Cartel Office, though citing the current European Commission guidelines, then in draft form, failed to consider the draft guidelines where they acknowledge the procompetitive effects that RPM arrangements might generate. This effectively maintains a per se illegality approach to RPM practices.71

5. CONCLUSION Whether fixed and minimum RPM may truly provide significant efficiencies so as to offset its negative effects remains controversial. While most economists theorise about the benefits of RPM,72 some commentators 70 T Caspary, ‘Swimming against the Zeitgeist? German Federal Cartel Office Issues Third Fine for Resale Price Maintenance (CIBA)’ [2010] European Competition Law Review 125; ‘Press Release on its Phonak GmbH decision’ (Bundeskartellamt, 15 October 2009) ; ‘Press Release on its CIBA Vision Vertriebs GmbH, Großostheim decision’ (Bundeskartellamt, 25 September 2009) ; ‘Press Release on its Microsoft Deutschland GmbH decision’ (Bundeskartellamt, 8 April 2009) . See also W Moeschel, ‘Vertical Price Fixing: Myths and Loose Thinking’ [2013] European Competition Law Review 233, 233. 71 Caspary (n 70) 127. See also L Atlee and Y Botteman, ‘Resale Price Maintenance and Most-Favoured Nation Clauses: The Future Does Not Look Bright’ (2013) November (1) CPI Antitrust Chronicle; F Amato, ‘RPM in the European Union: Any Developments Since Leegin?’ (2013) November (1) CPI Antitrust Chronicle. 72 See, for example, B Klein, ‘Competitive Resale Price Maintenance in the Absence of Free Riding’ (2009) 76 Antitrust Law Journal 431. Klein holds that RPM induces and guarantees increased promotional efforts even in the absence of freerider risk (the fear that discount retailers will freeride on their services) because the manufacturer also uses RPM to compensate the retailer for the promotion; the retailer will consider that his revenue will increase not only

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 22 /

Date: 13/10

JOBNAME: Duns PAGE: 23 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Resale price maintenance 267

such as Peeperkorn (as quoted above) seriously question the robustness of the arguments supporting the view that RPM may generate various efficiencies,73 as did the minority in Leegin who questioned the frequency and extent to which such efficiencies may materialise in reality. At the same time, though under EU law RPM has always been exemptible, the dearth of Commission decisions exempting RPM indicates that efficiency claims have not been sufficiently convincing to warrant the exemption of such restrictive practices. Moreover, empirical studies have not shown that allowing fixed or minimum RPM will necessarily lead to the claimed efficiencies. In the US in the period when the Miller-Tydings Fair Trade Act74 and the McGuire Act75 permitted individual states to enact fair trade laws that removed the per se illegality of minimum RPM, no particular positive effect was observed in the states that took a more permissive view towards RPM.76 Likewise, Giovannetti and Stallibrass because of increased sales due to the promotion, but also because the RPM guarantees a higher retailer margin. RPM may be viewed as a necessary tool to incentivise and compensate retailers for more intensive product promotion (greater point-of-sale promotional services through more specialised dedicated staff that can help consumers appreciate the brand more and thereby be enticed to buy it). Klein argues that for this reason RPM should be equated with exclusive territory as both are ways of compensating retailers for increased promotional efforts and so logically a competition authority should take the same approach to both practices. Moreover, Klein asserts that if one acknowledges that RPM is used not only as protection against freeriding (as is traditionally held), but also as a compensation mechanism, then the argument used by sceptics that RPM’s pro-competitive effects are rare because the freerider problem is rarely so serious as to deter dealer investment would be defeated; for this other procompetitive effect is more common and essential. Indeed, Klein holds that, perceived this way, RPM is ‘just an element of the normal competitive process’. Klein claims that, conversely, the anti-competitive harms usually associated with RPM occur in reality much more rarely than generally perceived, especially when RPM is being carried out by small manufacturers without their being coerced by dominant retailers. However, compare Ewald (n 3) 303–305 who cautions that one should not overestimate the efficiency-enhancing potential of RPM; inter alia by ignoring the effect on the infra-marginal customers (i.e. customers who would also buy the respective product at the lower level of pre-sale service) and the extent of this group in comparison to marginal customers, one would incorrectly assume that RPM is having a positive impact on overall efficiency and consumer welfare. 73 Peeperkorn (n 55) 208–212. 74 50 Stat 693. 75 66 Stat 631. 76 Indeed, Justice Breyer in Leegin Creative Leather Products, Inc. v PSKS, Inc cites the DOJ’s statement made during the 1975 Hearings on H. R. 2384

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

268 Comparative competition law

show that the removal of RPM on books in the UK in 1997, contrary to expectations, actually led to an increase in the total sales volumes for books and in retail diversity, leading the authors to remark that this ‘suggests that the efficiency benefits of RPM can (at the least) be overstated’.77 Giovannetti and Stallibrass also refer to two other UK cases involving RPM, one on children’s toys and the other on replica football kits, that demonstrate the indisputable anti-competitive effects of RPM.78 In the light of this, there is much to be said for maintaining a presumption of illegality in respect of fixed and minimum RPM; but one that allows for its rebuttal by the undertakings concerned where they can provide solid evidence of verifiable efficiency claims that may be passed on to consumers, as is the case under EU law.79

before the Subcommittee on Monopolies and Commercial Law of the House Committee on the Judiciary, 94th Cong., 1st Sess., 122 (1975) that a comparison of the prices in the fair trade law states with prices in the states that still outlawed minimum RPM showed that minimum RPM had actually raised prices by 19 per cent to 27 per cent: Leegin v PSKS (n 9). Justice Breyer also cites a study by the FTC staff in 1983 who, after studying numerous price surveys, concluded that collectively the surveys indicated that RPM in most cases leads to an increase in the prices of products sold with RPM: ‘Bureau of Economics Staff Report to the FTC’; T Overstreet, Resale Price Maintenance: Economic Theories and Empirical Evidence (Bureau of Economics 1983) 160. 77 Giovannetti and Stallibrass (n 55) 649. 78 Ibid 649–652. 79 Leegin is also criticised by some commentators for the uncertainty that they claim it has instilled – in Leegin, while placing RPM under the rule of reason, the Court did not lay down clear, coherent rules to guide the courts and authorities in the assessment that they are now obliged to carry out – leading to the risk that an excessively tolerant policy could develop, with the balance tilting towards the validity of these restrictions so that the consumer could end up paying higher prices and efficiency could be seriously undermined: see S Marco Colino, Vertical Agreements and Competition Law: A Comparative Study of the EU and US Regimes (Hart Publishing 2010) 84–87, 146. See the discussion on what should be the right rule-of-reason standard to follow and the related burden-shifting implications in a resale price maintenance analysis under US law in A Kuenzler, ‘Presumptions as Appropriate Means to Regulate Resale Price Maintenance: In Defence of Structuring The Rule of Reason’ (2012) 8 European Competition Journal 497.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 13_Chapter10

/Pg. Position: 24 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

PART III ENFORCEMENT AND SANCTIONS

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 1 /

Date: 21/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

11. Public enforcement Arlen Duke

1. INTRODUCTION Competition law can be enforced in two ways: by public enforcement agencies or by private action brought by a victim of anti-competitive conduct (considered in Chapter 14). This chapter focusses on public enforcement. The first section introduces the reader to the various public enforcement models. Drawing on jurisdictions discussed elsewhere in this volume to provide examples, these models, and the potential strengths and weaknesses of particular agency models, are examined. The second section of the chapter examines the methods employed by enforcement agencies to detect potential breaches. It also outlines the investigatory powers typically conferred upon enforcement agencies and methods used. The final section examines a range of enforcement strategies undertaken to secure compliance with competition laws.

2. ENFORCEMENT AGENCY MODELS Kovacic and Eversley are correct when they note that: The design of a jurisdiction’s administrative infrastructure can have a decisive influence on the type and quality of policy outcomes that a competition system achieves. Both older and newer competition systems have come to realize that a body of competition laws is only as good as the institutions entrusted with their implementation.1

As are Trebilcock and Iacobucci in their observation that when it comes to competition law institutions around the world, there is a ‘striking diversity of institutional designs’.2 Kovacic and Hyman also correctly 1 William E Kovacic and DeCourcey Eversley, ‘An Assessment of Institutional Machinery: Methods Used in Competition Agencies and What Worked for Them’ (ICN, Competition Policy Implementation Working Group May 2007). 2 Michael J Trebilcock and Edward M Iacobucci, ‘Designing Competition Law Institutions’ (2002) 25 World Competition 361, 361.

271

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

272 Comparative competition law

note: ‘The enforcement of competition law entails several discrete tasks: the investigation of possible wrongdoing, the decision to prosecute, the determination of culpability and the imposition of sanctions. In the design of a competition system, a jurisdiction can unbundle these functions, or combine them within a single entity.’3 Broadly speaking there are two enforcement agency models, the administrative model (under which the agency undertakes investigative, enforcement and administrative adjudicative functions) and the prosecutorial model (under which the agency undertakes investigative and enforcement functions and must prove its case by bringing a complaint before a generalist court or specialist adjudicative body). Within these two models there remain considerable differences in terms of the way in which agencies are structured. Whichever model is adopted, investigation and enforcement powers may be concentrated in a single agency or dispersed across multiple agencies. Further, the agency that oversees competition may also be responsible for overseeing and enforcing other, connected laws. The discussion below explains these models in more detail and considers their relative advantages and disadvantages. 2.1 The Administrative Model Under the administrative model, investigative and adjudicatory powers are vested in a single agency (or group of agencies), although a party found by the enforcement agency to have breached the law will almost always have a right to challenge the agency’s decision in the courts.4 Countries to adopt this model include Brazil, Japan and Mexico, the US (Federal Trade Commission (USFTC)) and the European Union as well as many European countries.5 In some countries, for example Australia and South Africa, constitutional limitations preclude the adoption of such a model.

3 William E Kovacic and David A Hyman, ‘Competition Agency Design: What’s on the Menu?’ (2012) 8 European Competition Journal 527, 535. 4 Donald I Baker, ‘Private and Public Enforcement: Complements, Substitutes and Conflicts – A Global Perspective’ in Ariel Ezrachi, Research Handbook on International Competition Law (Edward Elgar 2012) 255. 5 Including Bulgaria, Czech Republic, Denmark, Germany, Estonia, Greece, Italy, Cyprus, Latvia, Lithuania, Hungary, Netherlands, Poland, Portugal, Romania, Slovenia, Slovakia, Finland, Sweden and the United Kingdom: Commission Staff Working Paper, ‘Report on the Functioning of Regulation 1/2003’ (COM(2009)206, European Commission 2009).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 2 /

Date: 13/10

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 273

The enforcement model adopted in the European Union6 at the European level7 provides a good example of administrative arrangements.8 The Director-General for Competition (DG Competition) within the European Commission is responsible for investigating and forming a preliminary opinion on potential breaches of EU competition law. Investigation of potential competition law breaches is allocated to a particular Directorate and is managed by a case team. This team may be assisted by the Chief Competition Economist who provides economic guidance throughout the investigation process. The team’s findings may also be scrutinised by a peer review panel. If investigations uncover a potential breach, the Commission has the power to file a Statement of Objections (an administrative complaint). This statement describes the conduct and outlines the Commission’s preliminary assessment of that conduct. If the preliminary assessment suggests that a breach has occurred, the party being investigated can settle proceedings at this stage by giving an ‘undertaking’. If no undertaking is given, the case team conducts further investigation and a hearing is then conducted by a hearing officer (a Commission official outside DG Competition but who also reports to the Competition Commissioner) during which parties are entitled to make written and oral submissions. The hearing is attended by the Directorate dealing with the case. After the hearing, DG Competition drafts a preliminary decision that is vetted by Commission lawyers and an advisory committee. This preliminary decision is then submitted to the

6 Regulation 1/2003 enables national competition enforcement agencies and courts within the European Union to apply EU competition rules (Arts 5 and 6). In fact national competition enforcement agencies are obliged to apply EU law in all cases where it is applicable (Art 3): Council Regulation (EC) No 1/2003 of 16 December 2002 on the implementation of the rules on competition laid down in Articles 81 and 82 of the Treaty [2003] OJ L 1/1. 7 Competition law cases can also be initiated in national courts by both national competition authorities and by private litigants. National courts also have the power to award private damages. If the matter raises EU competition law issues, and the EU law is unclear, the matter may be referred to the European Court of Justice. It should be noted that ‘there is a great deal of variety in the Member State’s institutional arrangements’: Daniel A Crane, The Institutional Structure of Antitrust Enforcement (OUP 2011) 191. 8 Wils provides a useful overview of the EU enforcement arrangements: Wouter PJ Wils, ‘The Combination of the Investigative and Prosecutorial Function and the Adjudicative Function in EC Antitrust Enforcement: A Legal and Economic Framework’ (2004) 27 World Competition 201, 203–4.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

274 Comparative competition law

College of Commissioners who decides whether to adopt DG Competition’s recommendation by a majority vote.9 The Commission’s decision is binding upon the companies or individuals to whom it is addressed. A firm found to have breached the laws, or other persons with a special interest in the matter, can appeal to the General Court (which was, until recently, known as the Court of First Instance) on both legal and factual grounds. Those dissatisfied with the court’s determination can appeal, on legal grounds only, to the Court of Justice (formerly European Court of Justice). The USFTC provides another example of an administrative model. Its Bureau of Competition has the power to investigate unfair methods of competition (which includes conduct caught by sections 1 and 2 of the Sherman Act) and, where appropriate, initiate actions (although only with the approval of the Commissioners). Upon completing its investigations the Commissioners must decide whether to (a) close the investigation and take no action, (b) accept a settlement of the matter or (c) proceed to litigation. If a decision is made to proceed to litigation, the matter may (at first instance) be heard by an Administrative Law Judge, who is also an FTC employee10 although the FTC can also bring matters before the district courts. Decisions of the Administrative Law Judge can be appealed, on both questions of fact and law, to the USFTC itself in which case the five Commissioners sit ‘as a kind of appeals court reviewing the initial decision’11 after hearing arguments from both sides. The Commissioners reach a decision by majority vote. If the Commissioners reach a decision adverse to the respondent she may appeal the Commission’s decision to the Federal Court of Appeals on legal grounds only. 2.2 The Prosecutorial Model Under the prosecutorial model, an enforcement agency is primarily an executive body that investigates and prosecutes potential breaches of the law before an independent adjudicative body; either a generalist court or a specialist Tribunal.12 Many regimes employ specialist competition law courts including Canada, Chile, China, Denmark, Israel, South Africa and 9 The College of Commissioners ordinarily adopt the draft determination: Crane, Institutional Structure (n 7) 193. 10 At this hearing, the USFTC’s counsel and the respondent’s counsel present their arguments and Commissioners are ‘walled off’ from the matter: Wils (n 8) 207. 11 Ibid. 12 Crane, Institutional Structure (n 7) 18.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 4 /

Date: 13/10

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 275

the United Kingdom.13 Other regimes such as those in place in Australia and Ireland, as well as cases brought by the Department of Justice in the United States, require competition cases to be proven in the general court system. Often, criminal and civil prohibitions will be enforced by different bodies, requiring inter-agency cooperation. For example, in Canada, the Competition Bureau (CCB) is responsible for investigating and enforcing civil breaches and the Director of Public Prosecutions enforces the criminal prohibitions with investigative assistance from the CCB.14 Similar arrangements apply, for example, in Australia and Japan.15 In other jurisdictions, different divisions within the same enforcement agency enforce both civil and criminal prohibitions. For example, in the United States, criminal enforcement is undertaken by the Antitrust Division of the Department of Justice. The enforcement system adopted by the Antitrust Division of the Department of Justice (DoJ) provides a good example of an agency that operates under the prosecutorial model. The DoJ is able to open investigations and has extensive investigative powers. However, it has no power to find to impose sanctions. Rather, if the DoJ forms the view that conduct breaches US antitrust laws they must bring a civil or criminal suit in a generalist Federal District Court. Any decision reached by the Federal District Court is appealable, on legal grounds only, to the district Court of Appeal. A further appeal to the Supreme Court is also available, although such appeals are rare.16 As Crane has noted very few if any agencies, even those primarily executive enforcement agencies structured to preserve the separation of powers,17 are purely executory in nature. For example, almost all agencies have powers that enable them to compel testimony and the production of documents. Even the DoJ, which is in theory a purely executive body, has the power to issue investigative demands without applying to a court.18 Many executive agencies also have limited adjudicatory powers or the power to exercise quasi-judicial functions such as granting exemptions to conduct that would otherwise breach the law. For example, the Australian Competition and Consumer Commission has the 13

Ibid 224. Competition Act 1986, ss 7–8, s 23. 15 Antimonopoly Act art 96(1); Competition and Consumer Act 2010 (Cth). 16 Wils (n 8) 207. 17 Crane cites the Australian Competition and Consumer Commission and the South African Commerce Commission as examples: Crane, Institutional Structure (n 7). 18 Ibid 218–19. 14

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

276 Comparative competition law

power to ‘authorise’19 most forms of prohibited anti-competitive conduct where such conduct generates public benefits that outweigh any detriments (usually anti-competitive in nature) it causes.20 Thus while the Australian agency must prove competition law breaches in court, it takes on an adjudicatory function when dealing with requests for exemption. 2.3 Assessing the Merits of the Competing Models When it comes to the assessment of agency models there are two key factors to consider: accuracy and efficiency. It should be noted from the outset that many factors impact both efficiency and the capacity of an enforcement system to generate correct decisions. Most of these factors, such as the level of expertise of agency staff, are unlikely to favour one model over the other.21 2.3.1 Administrative model The administrative model is often said to perform well on the efficiency metric. As agencies are not required to commence legal proceedings in court, administrative costs are typically lower under such models although this will ultimately depend on the level of safeguards built into the particular administrative model.22 The highly specialised and technical nature of competition law adjudication may justify the primary adjudicatory function being performed by those who have a competition law and economics specialisation,23 although such specialisation can also occur under the prosecutorial model where claims are heard by a specialist Tribunal. Levels of administrative efficiency will not be determined solely by whether an agency follows the administrative or prosecutorial model. The processes adopted within the agency will also be influential. For example, Trebilcock and Iacobucci attribute the DG Competition’s realisation of substantial efficiencies (in the context of merger review) to the fact that even formal proceedings before the DG Competition are less judicialised and elaborate than procedures adopted by prosecutorial agencies.24 Trebilcock and Iacobucci also note that the experience in the

19

Other administrative processes (such as notification and clearance) can also be used by those seeking immunity. 20 When assessing detriments and benefits, a qualified total welfare standard is employed: Re Qantas Airways Ltd [2004] ACompT 9. 21 Wils (n 8) 214. 22 Ibid 223. 23 Crane, Institutional Structure (n 7) 39. 24 Trebilcock and Iacobucci (n 2) 381.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 277

USFTC (which also has the power to resolve matters administratively) is less encouraging as protracted delays are common because of the FTC’s decision to adopt hearings that often resemble full-scale court hearings.25 Given this, it may be that enforcement agencies should be thought of as falling on a spectrum, rather than into two or three discrete and neatly defined categories as suggested earlier in the chapter. Whilst the administrative model may generate efficiencies, ‘[a] process that relies on a single agency to initiate an investigation, issue an administrative order, and review that order raises concern about the independence … of the adjudication process’.26 It also raises concerns about prosecutorial bias. It has been said, for example, that the FTC structure violates due process.27 Allegations of prosecutorial bias have also been made against DG Competition28 although attempts to challenge the European system on the basis that it contravenes the basic human right to a fair hearing by an independent and impartial Tribunal (enshrined in Article 6 of the European Convention on Human Rights)29 have been unsuccessful, largely because affected parties have a right to challenge, in court, any decision made by DG Competition.30 Claims of prosecutorial bias are difficult to assess, especially as errors will occur under both agency models. Statistical studies are plagued with interpretation difficulties.31 However behavioural law and economics, drawing on insights from psychology, provides two theoretical explanations for such a bias. First, the behavioural economics movement has taught us that human decision making is subject to confirmation bias: we tend to search for information that confirms rather than challenges our beliefs. It is important to remember that investigations are commenced because of an initial belief or suspicion that there has been a contravention of competition law. Subsequent investigations may aim to gather evidence of an assumed breach rather than dispassionately assessing whether such a breach has occurred. Second, hindsight bias, the tendency to falsely believe that the resulting outcome is the only one that could have occurred,

25 26 27 28 29 30

Ibid 382. Crane, Institutional Structure (n 7) 221–2. Wils (n 8) 210. Crane, Institutional Structure (n 7) 200. Wils (n 8) 208. Bendenoun v France (24 Feb 1994) [46]; Jansevic v Sweden (23 Jul 2002)

[81]. 31

For an example of an empirical study, see Frank Montag, ‘The Case for a Radical Reform of the Infringement Procedure under Regulation 17’ (1996) 8 European Competition Law Review 428 discussed in Wils (n 8) 216.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 7 /

Date: 13/10

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

278 Comparative competition law

may lead to investigations being pursued for fear that unless some incriminating evidence is found, there will be a perception that the investigations should never have been made in the first place. Further, it has been said that agency staff have an incentive to find guilt because it generates impressive statistics whether or not the decisions are correct. The risk of prosecutorial bias can be reduced by providing a right to judicial review (a feature of most administrative systems) or by appropriate internal checks and balances. However, this is only true if the courts are truly independent and not overly deferential to the agency’s findings. In such circumstances, and where the hearing is de novo, prosecutorial bias will be reduced. However, in some jurisdictions such as Japan, ‘it is not certain that courts will make completely impartial and objective decisions’.32 Under Japanese administrative laws, there is a presumption that decisions of enforcement agencies are lawful. Further, Japanese ‘courts generally lack expertise in competition law and tend to adhere to JFTC decisions’33 as evidenced by the high rate at which courts uphold the initial decision of the Japanese agency.34 There is also evidence of deference in the European Union system. The EU courts initially ‘limited themselves to verifying whether the Commission has acted lawfully and … granted the Commission a substantial margin of discretion in the economic analysis of facts’.35 More recently, the courts are demonstrating a greater willingness to review complex decisions of the Commission36 although the review does not approximate a hearing de novo and it is fair to say that the intensity of review varies. This may well explain ‘the trend that the Commission wins the majority of cases that are appealed’.37 If judicial review operates in a biased fashion, it may introduce additional costs that threaten the efficiency of the system while doing little to improve accuracy. In fact, it is possible that any cost advantage of the administrative model ‘might even be inverted, as a result of internal checks and balances and frequent judicial review’.38

32 Mitsuo Matsushita, ‘Reforming the Enforcement of the Japanese Antimonopoly Law’ (2010) 41 Loyola University Chicago Law Journal 521, 528. 33 Ibid 529. 34 Ibid. 35 ‘European Union’ in American Bar Association, Competition Laws Outside the United States: Volume 1 (2nd edn, American Bar Association 2011) 10. 36 See, for example, Case T-342/99 Airtours plc v Commission [2002] ECR II-2585; Case T-80/02 Tetra Laval BV v Commission [2002] ECR II-4519. 37 ‘European Union’ (n 35) 11. 38 Wils (n 8) 226.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 279

2.3.2 Prosecutorial model Just as judicial review can reduce the risk of prosecutorial bias, the prosecutorial model, by separating investigative and adjudicative functions, is said to afford due process and in most instances it does.39 Whilst the involvement of an independent adjudicator helps overcome prosecutorial bias it can also cause an enforcement model to perform badly in terms of timeliness and process costs.40 Further, where cases are brought in a generalist court, matters may be decided by a body with an insufficient level of expertise, threatening the accuracy of outcomes. A better balance between administrative efficiency and due process is likely to be achieved where the decision is made by a specialist body, especially where the procedures of the specialist body allow for matters to be dealt with less formally than would be the case if the matter was resolved in the courts.41 In fact, Crane predicts that as ‘the complexity of antitrust increases, perhaps we should expect that more jurisdictions will migrate towards specialized antitrust courts’,42 although he also acknowledges that constitutional limitations may make this difficult if not impossible in some jurisdictions. The need to prove allegations before a specialist Tribunal means that the decision maker is independent but will also be familiar with the relevant competition laws and the economic principles that underpin those laws. Further, as Tribunals are not judicial bodies, streamlined process and procedures can be adopted to improve the speed with which cases are heard. Whether such benefits are achieved will always be determined by institutional design. Trebilcock and Iacobucci note in the period from 1986 to 2001 the Canadian Competition Bureau, a prosecutorial agency, brought less than one contested case per year before the Competition Tribunal (a specialist Tribunal).43 This led Trebilcock and Iacobucci to observe that: The resulting costs, delays and uncertainty involved in Tribunal proceedings have induced firms and the Commissioner to substitute the locus of decisionmaking … away from the Tribunal towards the Bureau where process values such as transparency, accountability and reasoned public decision-making are much diminished … The delay and cost of Tribunal Proceedings has led to de facto integration of investigative, enforcement, and adjudicative functions within the Competition Bureau.44 39 40 41 42 43 44

Columns Design XML Ltd

Trebilcock and Iacobucci (n 2) 370. Ibid. Ibid 373. Crane, Institutional Structure (n 7) 225. Trebilcock and Iacobucci (n 2) 373. Ibid 374, 376.

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

280 Comparative competition law

2.4 Multi-Agency Models Whether the administrative or prosecutorial model is adopted, competition law enforcement may be concentrated in a single agency or dispersed across multiple agencies. As Crane notes, ‘it is not unusual for different law enforcement agencies to have overlapping jurisdiction’.45 Further, the agency that oversees competition may also be responsible for overseeing and enforcing other, connected laws. In some multi-agency jurisdictions, enforcement authority is clearly separated. This separation can be achieved in a variety of ways. Agencies can be allocated responsibility to enforce particular provisions,46 particular types of offences (criminal or civil) or to enforce laws in particular industries. In other jurisdictions, such as the United States, the division is not so clear. In the United States two federal government agencies, with overlapping jurisdiction, enforce US competition laws: the USFTC (which is an independent regulatory agency) and the Antitrust Division of the Department of Justice (an executive branch).47 These bodies share authority to enforce the Clayton Act (which prohibits, among other things, anti-competitive exclusive dealing (tying), anti-competitive mergers and acquisitions and anti-competitive price discrimination). Technically, the DoJ has exclusive jurisdiction to enforce the Sherman Act (which prohibits anti-competitive arrangements and monopolisation). However, the USFTC ‘can do so indirectly under section 5’s broad mandate’.48 However, the DoJ has exclusive jurisdiction for criminal enforcement.49

45

Crane, Institutional Structure (n 7) 27. See, for example, arrangements in China: text accompanying n 69 below. 47 The State Attorneys General and private litigants also play a role in the enforcement of US competition laws. 48 Section 5(2) of the Federal Trade Commission Act (codified at 15 USC §45(a)) gives the USFTC the power to prevent the use of unfair methods of competition, which includes conduct that would fall within ss 1 and 2 of the Sherman Act. In Fashion Originators’ Guild of America v FTC 312 US 457 (1941) the US Supreme Court held that ‘[i]f the purpose and practice of the combination of garments manufactured and their affiliates runs counter to the public policy declared in the Sherman and Clayton Acts, the Federal Trade Commission has the power to suppress it as an unfair method of competition’. 49 This is because only the Department of Justice can directly enforce the Sherman Act, the only statute that imposes criminal liability for anti-competitive conduct. 46

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 10 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 281

As most of the USFTC and DoJ’s civil jurisdiction is duplicative, ‘both agencies could theoretically investigate and sue over the same activity’.50 The agencies consult before opening any investigation but duplication can lead to confusion about jurisdiction. It can also cause in-fighting. This could explain the fact that the decided trend in recent times has been to convert multiple agencies to single agencies as witnessed in France, Estonia, Portugal, Spain and Brazil.51 More recently, the United Kingdom has combined the competition functions of the Office of Fair Trading (those functions being enforcing cartel and first-phase review body for mergers and market studies) and the Competition Commission (whose role it was to conduct investigations referred to it by the OFT)52 to create a single competition regulator, the Competition and Markets Authority.53 In the European Union, member states also have their own competition laws, often leading to parallel application of national and EU laws to the same anti-competitive conduct. As of 2004, the local enforcement agencies must apply EU competition law to cases where the conduct affects trade between two member states.54 Local laws cannot be applied to such cases if the conduct does not breach EU law. One reason to have agencies with overlapping functions is to ‘foster interagency competition that increases the output and improves the quality of the product that the agencies are intended to supply’.55 Multi-agency models provide ‘insurance against the possibility that any single agent will fail to execute its responsibilities for reasons of laziness, resource constraints or political influence’. It has also been argued that ‘[m]ultiplicity … creates the basis for rivalry, which can lead to improvements in system wide performance’.56 However this will not always be the case. In the United States, somewhat ironically, the agencies ‘essentially have a market-division approach to jurisdiction, with each agency laying claim to the industries with which they have greater experience’.57 50

Crane, Institutional Structure (n 7) 42. Kovacic and Hyman (n 3) 528. 52 Both of which closed on 1 April 2014. 53 See Stephen Wilks, ‘Institutional Reform and the Enforcement of Competition Policy in the UK’ 7 European Competition Journal 1. 54 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 L 1/1. 55 William E Kovacic, ‘Downsizing Antitrust: Is it Time to End Dual Federal Enforcement’ (1996) 41 Antitrust Bulletin 505, 510. 56 Kovacic and Hyman (n 3) 532. 57 Crane, Institutional Structure (n 7) 43. See also Daniel A Crane, ‘Technocracy and Antitrust’ (2008) 86 Texas Law Review 1159. 51

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

282 Comparative competition law

Further multiplicity can also detract from the coherence of national competition policy if the agencies’ enforcement priorities do not align. A single agency may also be vested with functions beyond competition policy. For example, the Australian Competition and Consumer Commission, the Canadian Competition Bureau and the USFTC also enforce consumer protection laws. Whilst single-purpose agencies may be better placed to establish coherent policies that focus solely on competition issues, multi-purpose agencies can ‘realise synergies and lower the costs associated with coordinating policy between … separate institutions with related functions’58 provided the combined functions are true policy complements. 2.5 Agencies in Emerging Jurisdictions Many jurisdictions are currently in the process of implementing entirely new or significantly revised competition law enforcement regimes. The US and EU experiences with the prosecutorial and administrative regimes respectively appear to have influenced the design of these newer regimes. It is worth noting that often in emerging jurisdictions, the agency model employed is not overly determinative of efficiency or accuracy. Many agencies are operating in economies that lack a competition culture and are only just starting to embrace market principles.59 Competition laws are often implemented as conditions of trade arrangements rather than because the jurisdiction is genuinely committed to free market principles. 2.5.1 Japan The Japan Fair Trade Commission (JFTC) is in the process of moving from an administrative enforcement model to a more prosecutorial style model. The JFTC is an ‘independent administrative commission’,60 modelled on the USFTC.61 It conducts administrative investigations and where it believes that the law has been contravened it has the power to ‘restore competitive order immediately by giving cease-and-desist orders

58

Kovacic and Hyman (n 3) 533. Michael Trebilcock, ‘Designing Competition Institutions: Values, Structures and Mandates’ (2010) 41 Loyola University Chicago Law Journal 455, 467. 60 Japan Fair Trade Commission, ‘For Fair and Free Market Competition’ (Japan Fair Trade Commission 2010) 3 . 61 Matsushita (n 32) 524. 59

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 12 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 283

when illegal acts have been detected’.62 The JFTC also has the power to impose surcharge payment orders (administrative fines).63 Parties issued with such orders have two options. They can accept the order, in which case a ‘recommendation decision’ will be issued, which essentially makes the cease-and-desist order legally binding. Alternately, they can reject it. Currently, if the recipient rejects the order, the JFTC will conduct a hearing during which the recipient presents arguments as to why the cease-and-desist order was incorrectly issued.64 The JFTC (through its five Commissioners and Chair) issues a decision. Those dissatisfied with the JFTC’s decision are able to appeal to the Tokyo High Court, which has the power to revoke the JFTC’s decision and any penalties imposed.65 The JFTC is said to be moving to a hybrid system because in December 2013 Japanese competition laws were amended. By 2015 the JFTC hearing procedure will be abolished. In its place, the Tokyo District Court will be given exclusive jurisdiction to hear appeals from JFTC cease-anddesist orders.66 As Crane has observed, the ‘impetus for [the] reform echoes some of the ongoing themes in the European discussion about separation of power, transparency and due process’.67 The lack of separation between investigation and adjudication created concerns about the legitimacy of the agency’s decision making. Echoing the discussion of the accuracy/ efficiency trade-off discussed above, some argued for the preservation of the original arrangements on the basis that the JFTC has developed considerable expertise and, because of that expertise, is best placed to adjudicate such matters.68 2.5.2 China China’s enforcement system is quite unique. Three separate agencies enforce China’s competition laws but jurisdiction is divided on the basis

62

Japan Fair Trade Commission (n 60). Ibid. 64 Further see Crane, Institutional Structure (n 7) 221. 65 Japan Fair Trade Commission (n 60). 66 Japan Fair Trade Commission, ‘New Year Message from Chairman Sugimoto’ (Japan Fair Trade Commission Website January 2014) . 67 Crane, Institutional Structure (n 7) 221. 68 Matsushita (n 32). 63

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

284 Comparative competition law

of the form of anti-competitive conduct.69 The Ministry of Commerce oversees merger regulation. The State Administration for Industry and Commerce covers abuse of dominance and cartel behaviour except for price-related conduct. The National Development and Reform Commission regulates price-setting conduct as abuses of dominance involving such conduct. As Crane has noted, ‘China is perhaps the only country to separate enforcement duties based on the nature of the offence’.70 Parties dissatisfied with the decision of an enforcement agency have a right to challenge the decision in the courts but they must first apply for administrative reconsideration.71 2.5.3 India The Competition Commission of India adopts an administrative model. It has the power to issue cease-and-desist orders and impose monetary penalties. It may issue orders or directions as it sees fit.72 However, such orders can be appealed to the Competition Appellate Tribunal, a specialist competition court. The Supreme Court of India hears appeals from the Tribunal. Crane suggests that ‘[b]y creating an independent court to hear Commission appeals, India’s structure may address, in part, the transparency and separation of powers concerns raised by the unified administrative model while still capturing the efficiencies created by a unified enforcement agency’.73

3. DETECTION AND INVESTIGATION 3.1 Detection Potential breaches of competition laws come to the attention of enforcement agencies through a variety of avenues, including complaints from those affected by the conduct in question (such as customers, competitors

69 See generally Xiaoye Wang, ‘Highlights of China’s New Anti-Monopoly Law’ (2008) 75 Antitrust Law Journal 133 cited in Crane, Institutional Structure (n 7) 215. 70 Crane, Institutional Structure (n 7) 215. 71 Anti-Monopoly Law (China) s 53. 72 Competition Act (India) s 27. 73 Crane, Institutional Structure (n 7) 224.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 285

or suppliers),74 information provided by other government agencies,75 information provided by competition law enforcement agencies in other countries as well as self-reporting (often prompted by leniency policies, an issue explored in greater depth below). Further, agencies may also adopt a proactive approach and monitor markets of concern. 3.1.1 Leniency policies Increasingly, enforcement agencies employ immunity or cooperation policies to detect potential breaches. Immunity policies offer total immunity from prosecution, usually to the first party to report. Such policies are designed to uncover information that may not otherwise be available to the enforcement agency. Cooperation policies rarely offer complete immunity, but they provide incentives to cooperate with enforcement agencies when potential breaches are being investigated. Thus, even if a party is not the first to report the conduct, they may receive a discount on penalties imposed where they cooperate with the enforcement agency throughout its investigations. Due to the inherent secrecy of cartel conduct, and the growing sophistication of cartel members, immunity policies are an important tool for uncovering such conduct. In fact, many immunity policies are often limited to cartel conduct, although that is not the case in jurisdictions such as the European Union76 and China where immunity from prosecution from all offences is offered.77 In particular in regard to cartels, ‘[a]lmost all government enforcement agencies now rely very heavily on some participant in a cartel breaking ranks and coming forward to blow the whistle on its conspiring competitors, in return for a real anticipated reward’.78 For example, the EU estimates that one-third of its decisions were triggered by leniency applications79 and the Japanese enforcement 74

The EU estimates that one in 10 of its decisions are based on complaints received: Commission Staff Working Paper, ‘Report on the Functioning of Regulation 1/2003’ (COM(2009)206, European Commission 2009) [21]. 75 For example, an agency that regulates foreign investment may be aware of potential merger activity. 76 On 30 June 2008 the European Commission introduced a new settlement procedure applying solely to cartels: Commission Regulation (EC) 622/2008 of 30 June 2008 amending Regulation (EC) 773/2004 [2008] OJ L 171. 77 Anti-Monopoly Law (China) s 46. The State Administration for Industry and Commerce has issued rules that explore the conditions on which immunity will be granted. 78 Baker (n 4) 25. 79 Commission Staff Working Paper, ‘Report on the Functioning of Regulation 1/2003’ (COM(2009)206, European Commission 2009) [21].

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 18 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

286 Comparative competition law

agency reported that it received 26 applications for immunity in the three months after the introduction of its immunity policy in 2006.80 Put simply, such policies offer rewards (such as immunity from prosecution or the imposition of more lenient penalties) in exchange for access to information that may never otherwise be revealed to or uncovered by enforcement agencies. They change ‘the cost/benefit calculation of cartelization’81 and other forms of anti-competitive conduct. Barnett, a former Assistant General of the Antitrust Division of the United States, describes it well when he notes that immunity policies ‘convert cartel members into cooperative witnesses, so that prosecutors can gain access to background information, testimony, and the documents that otherwise might be destroyed’.82 Barnett83 notes that immunity policies require three elements to work properly: (1)

It must be possible to impose severe sanctions on firms that do not obtain immunity. An immunity program must provide significant benefits as compared with the alternate strategy of staying in the cartel and risking prosecution. There must be a genuine fear that the matter would be detected. Whilst leniency and immunity policies can add to the possibility of detection, they are unlikely to be effective unless cartelists believe there is a risk of detection if they do not self-report. The policies must be predictable and transparent. Published policies that make firm commitments about when immunity will be conferred are likely to be more effective.

(2)

(3)

These criteria are well accepted and form the basis of US immunity policies, which in turn form the basis of such policies in many other jurisdictions. However, some are now beginning to question the criteria. For example, Beaton-Wells has recently noted that reviewing leniency policies against such criteria fails to directly examine the effectiveness of policies against their ‘objectives of facilitating detection, prosecution and

80 Thomas O Barnett, Criminal Enforcement of Antitrust Laws: The US Model, presented at Fordham Competition Law Institute’s Annual Conference on International Antitrust Law and Policy (September 2006) 9. 81 Ibid. 82 Ibid. 83 See also Scott D Hammond, Remarks at ICN Workshop on Leniency Programs: Cornerstones of an Effective Leniency Program (2004) 4–5.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 16 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 287

deterrence of cartel conduct’.84 Rather, it is generally assumed that such policies meet these objectives if the previously listed criteria are met. Based on an empirical study of Australian cartel regulation, Beaton-Wells argues that whilst conventional thinking about the design and administration of immunity policies may make sense in theory, it may not necessarily be realisable or even desirable in practice. In particular, Beaton-Wells questions the assumed link between the severity of sanctions and the likelihood of self-reporting, noting that in the Australian context self-reporting actually decreased after the introduction of criminal sanctions for hard-core cartel conduct.85 Beaton-Wells posits that factors such as corporate culture of compliance, rather than a fear of detection, may in fact be a stronger driver of immunity applications. Beaton-Wells also questions the conventional wisdom that discretion or uncertainty is necessary to the effective functioning of such policies, noting that practitioners often ‘value the opportunity to negotiate an outcome with the [enforcement agency]’.86 Immunity policies around the world are relatively consistent because, as noted above, most have been based on the United States DoJ policies and experiences. Corporate leniency policies typically required, at minimum, that: (1) (2) (3)

(4)

the applicant was the first to provide evidence to the enforcement agency;87 the applicant took, upon discovery of the cartel, prompt and effective action to bring the conduct to an end; the applicant reports the wrongdoing with candour and completeness and cooperates with the enforcement agency throughout the investigation; the applicant did not coerce another party to participate in the activity.

84 Caron Beaton-Wells, ‘Leniency Policies: Effectiveness-Testing’ in N Charbit and E Ramundo (eds), William Kovacic, An Antitrust Tribute Liber Amoricum – Vol II (Institute of Competition Law 2014) 303. 85 In the four-year period prior to the introduction of criminal sanctions (2005–2009) 47 markers were received. In the four years since the introduction of criminal sanctions, only 36 markers were received. 86 Beaton-Wells (n 84). 87 Many leniency policies will include a ‘marker system’. The marker system allows a party to report cartel conduct even if they may not have collated the necessary information and evidence to meet the threshold for immunity. Upon reporting, the applicant places a marker, which protects its place in the queue.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 17 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

288 Comparative competition law

Many immunity policies also include a requirement that the informant makes restitution, although often this requirement is not strictly enforced as agencies focus on uncovering and deterring anti-competitive conduct, leaving those harmed to seek compensation by way of private action. Most policies provide immunity from prosecution by the enforcement agency. However, immunity rarely extends to private civil suits in which compensation is sought for harm caused by the illegal conduct.88 Concerns about potential civil suits that may flow once the conduct is revealed have the potential to undermine the effectiveness of immunity policies. By bringing the contravention to the attention of enforcement agencies, the applicant must trade off protection from civil/criminal prosecution (which may not occur if the cartel is not detected) against the possibility that it will face private claims when others involved in the conduct are prosecuted and the anti-competitive conduct is exposed. Some jurisdictions prioritise the effective operation of the immunity policy over assisting victims to recover compensation by way of private action. For example, the Australian Competition and Consumer Commission has adopted a policy of not disclosing information gathered by way of immunity application to private litigants. The Commission adopts a paperless process so as not to generate ‘any new documents which may involve admissions or assist others’.89 Under most corporate immunity policies, when immunity is granted to a company, its employees, directors and officers receive definitive immunity provided they are prepared to admit their involvement in the illegal conduct and cooperate with the enforcement agency.90 Nevertheless individual immunity policies still have a role to play. An individual may need to invoke an individual leniency policy when the company she works for does not wish to apply for immunity or does not qualify for corporate immunity.91 88 Although note that in the United States an immunity applicant may have any civil damages awarded against them de-trebled provided the requirements of the Antitrust Criminal Penalty Enhancement and Reform Act of 2004. One such requirement is that the immunity applicant exhibited ‘satisfactory cooperation’ to plaintiffs in the civil action. See also Baker (n 4) 259. 89 Australian Competition and Consumer Commission, ‘ACCC Immunity Policy Interpretation Guidelines’ (ACCC 2009) [57]. See also Caron BeatonWells and Brent Fisse, Australian Cartel Regulation: Law, Policy and Practice in an International Context (Cambridge University Press 2011) 407. 90 See for example US Corporate Leniency Policy, C. 91 In addition to the requirements discussed in the text, many corporate policies require that the admission of liability is a truly corporate act as opposed to confessions by particular employees. Where a company is not willing to make

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 18 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 289

3.2 Investigatory Powers Broadly speaking, competition agencies have similar investigatory powers available to them although the scope of those powers and the circumstances in which they may be exercised varies from jurisdiction to jurisdiction. 3.2.1 Compulsory acquisition of information and documents The primary investigatory tool relied upon by enforcement agencies is the power to compulsorily acquire information and documents,92 although ordinarily enforcement agencies will first make voluntary requests before invoking such powers. Generally speaking, when requiring such disclosure the enforcement agency is obliged to state the legal basis for the complaint, describe the information requested and warn about the consequences of non-compliance or dishonest response. In some jurisdictions, such as Canada, the enforcement agency must apply to the court for a warrant.93 Other agencies are able to exercise such powers of their own volition. For example, in the United States the Attorney-General or the Assistant Attorney-General in charge of the Antitrust Division of the Department of Justice has the power to issue such a demand if he or she has ‘reason to believe that [the person to be issued with the request] may be in possession … of documentary material, or may have information, relevant to a civil antitrust investigation’.94 The Australian Competition and Consumer Commission has the same powers if the Commission is satisfied of the factors listed in the previous sentence.95 In other jurisdictions the enforcement agency is given even wider discretion. In the EU, the European Commission can, by decision, require entities to disclose such information.96 The exercise of such powers is not qualified by any express conditions;97 however the

such an admission, individuals are still able to seek immunity under individual leniency policies. 92 United States: 15 USC 34, s 1312; 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 L 1/1 art 18. 93 Competition Act (Canada) s 11. 94 15 USC 34, s 1312(a). 95 Competition and Consumer Act 2010 (Cth) s 155. 96 ‘Brazil’ in American Bar Association (n 35) 10. 97 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 L 1/1 art 18(3).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

290 Comparative competition law

Commission is required to state the legal basis of its request, providing an implicit threshold requirement. 3.2.2 Search and seizure powers Most enforcement agencies also have search and seizure powers, often referred to as ‘dawn raid’ powers. Agencies are typically allowed to enter premises, examine books and other business records, take or obtain copies of such books or records, seal or detain evidence and ask for explanations on facts or documents relevant to the purpose of the inspection. Again some agencies are granted such power to use at their discretion. Others must have the use of such powers approved by a court. The European Commission has such powers, which are given to it so that it can ‘carry out the duties assigned to’ it.98 The Chinese authorities are also granted such powers once authorised by the head of the enforcement agency.99 In Japan, investigatory attorneys are given power to compulsorily seize any documents and other information necessary for investigation and can order private parties to produce them.100 The Australian Competition and Consumer Commission, the US Department of Justice Antitrust Division, the Canadian Competition Bureau, the Japanese Fair Trade Commission and the Brazilian agencies must apply ex parte for a warrant.101 In jurisdictions that impose criminal sanctions for competition law breaches, investigators are given further powers. In many jurisdictions, including Australia,102 the United Kingdom, and the United States,103 investigators can apply to a court of competent jurisdiction for an order authorising the interception of wire or oral communications. They can also undertake surveillance of business premises or individuals. In the United States, the DoJ proceeds with criminal investigations through the grand jury process. The task of the grand jury is to decide if someone should be charged for a crime. To fulfil the task the grand jury is given broad investigatory powers, including the power to subpoena documents and witnesses. Thus grand juries serve both an investigatory and an accusatory function.

98

Ibid art 20(1). Anti-Monopoly Law (China) art 39. 100 Matsushita (n 32) 531. 101 Australia: Competition and Consumer Act 2010 (Cth) s 154; Canada: Competition Act s 15(1)(a)–(b); Japan: arts 102–104. 102 Telecommunications (Interception and Access) Act 1979 (Cth). 103 18 USC 2516(1)(r). 99

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 20 /

Date: 13/10

JOBNAME: Duns PAGE: 23 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 291

4. COMPLIANCE AND ENFORCEMENT 4.1 Promoting Compliance The primary goal of enforcement agencies, whatever form they take, is to promote compliance with competition laws. The primary aims of an enforcement agency include stopping unlawful conduct and deterring future offending conduct. Very often these aims can be achieved without the need to incur the expense associated with a contested judicial proceeding. As the now disbanded UK Office of Fair Trading notes: There are many instruments that a competition authority can use short of reaching a formal infringement decision that may influence companies’ behaviour. These include opening administrative proceedings, accepting commitments, informing companies that a complaint has been made, issuing warnings, conducting market studies, publishing codes of conduct or guidelines and advocacy.104

‘A competition policy program need not invariably place the enforcement of antitrust commands atop its agenda.’105 In fact, most agencies do not. This is best illustrated by the enforcement pyramid approach. The pyramid diagrammatically illustrates the range of compliance options available to the enforcement agency and illustrates ‘the flexibility of [the enforcer’s] approach in using education, guidance and advice to secure compliance’.106 The enforcement pyramid comes from responsive regulation theory,107 which: presumes that cooperative compliance, will work most of the time with most firms [and that] deterrence will be the strategy that is most likely to work when cooperative compliance fails … The basic idea is one of organising compliance strategies in a regulatory pyramid where more cooperative strategies are deployed at the base of the pyramid and progressively more punitive approaches are used as the regulator moves up the hierarchy of strategies. 104 Office of Fair Trading, ‘The Deterrent Effect of Competition Enforcement by the OFT: A Report prepared by Deloitte’ (OFT962, Office of Fair Trading 2007) [2.16]. 105 Kovacic and Eversley (n 1) 5. 106 Office of Fair Trading, ‘Statement of Consumer Protection Enforcement Principles’ (OFT1221rev, Office of Fair Trading 2012) 13. 107 John Braithwaite, To Punish or Persuade (SUNY Press 1985). See also Ian Ayres and John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (OUP 1992).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

292 Comparative competition law There is a presumption in favour of starting at the base of the pyramid and only escalating up the pyramid if the regulated firm refuses to respond in a spirit of cooperative compliance. The objective is that firms and individuals will comply even without enforcement action … through internalisation and institutionalisation of compliance norms.108

Moreover, if agencies are patient and fair in escalating matters up the enforcement pyramid, and provide plenty of warning of the inevitability of escalation and sufficient time to comply before moving up a rung in the enforcement pyramid, the agencies’ reputation for justice as well as their ‘reputation for ruthless retaliation against recalcitrance’109 will be promoted. The enforcement pyramid approach is employed in many jurisdictions and even those that do not explicitly adopt it exercise the range of compliance strategies commonly subsumed in an enforcement pyramid. The enforcement pyramid adopted by the Australian Competition and Consumer Commission has Education, Advice and Persuasion at its base. Voluntary self-regulation is next followed by Settlement. Court cases are at the top.110 A similar prioritisation of approaches is seen in the more detailed enforcement pyramid applied in the consumer protection context in the United Kingdom.111 The proportion of space at each layer represents the proportion of enforcement activity at that level. Education, which sits at the base of the pyramid, is the most commonly employed strategy. Education campaigns

108

Christine Parker, John Braithwaite and Natalie Stepanenko, ‘ACCC Enforcement and Compliance Project: Working Paper on ACCC Compliance Education and Liaison Strategies’ (Centre for Competition and Consumer Policy, Australian National University 2004) 3. 109 Ayres and Braithwaite (n 107) 43. 110 In its latest guidelines the Commission does not present this information by way of a pyramid as it has previously done: Australian Competition and Consumer Commission, Compliance and Enforcement Policy (21 February 2014) cf Australian Competition and Consumer Commission, Compliance and Enforcement Policy (7 April 2009). 111 The Office of Fair Trading employed an enforcement pyramid that includes from base to top: education, guidance and advice, self-regulation, partnering to support compliance efforts (including the implementation of compliance programmes), dialogue warnings, undertakings, penalties and civil and criminal penalties: The Office of Fair Trading, Statement of Consumer Protection Enforcement Principles (OFT1221rev, Office of Fair Trading 2012).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 293

attempt to ‘coax compliance by persuasion’112 and aims to raise awareness of the law. This is achieved through methods such as guidelines explaining the law, training programmes, public awareness campaigns and public speeches. Often agencies work with particular businesses or industry associations to develop codes of conduct, an example of voluntary self-regulation which is ‘aimed at nipping non-compliance in the bud’.113 Such an approach engages businesses in the process and engenders commitment towards the goals sought to be achieved by competition laws. Businesses have the incentive to participate in such endeavours as this may ward off more stringent government regulation. Enforcement agencies also promote self-regulation by encouraging businesses to implement compliance programmes. Even when they believe an offence has been committed, enforcement agencies often settle matters. Enforcement action is expensive and time consuming and where the conduct is not overly serious, and there is little chance that the business will re-offend, there is little to be gained from devoting resources to prosecution. Administrative settlements allow an agency to divert its limited funds to other, more effective compliance strategies further down the pyramid. Where the conduct is more serious, the matter may be resolved by a consent judgment. In fact, the US Department of Justice Antitrust Division concludes most matters by consent judgments.114 Consent judgments often need to be approved by the courts. Whilst courts do not simply ‘rubber stamp’ consent agreements, they are typically very deferential when granting approval. When deciding between commencing court action and resolving the matter by way of settlement, agencies consider a range of factors including the nature of the alleged contravention, the impact of the conduct on third parties, recidivism and the apparent good faith of the party involved. Many agencies also have the power to accept courtenforceable undertakings from parties to bring anti-competitive conduct to an end and correct any harm it may have caused to competition.115 Even though they sit at the top of the pyramid, and therefore are the least used of all the compliance strategies, court actions are incredibly 112

John Braithwaite, ‘Convergence in Models of Regulatory Strategy’ (Paper delivered at a Public Seminar entitled ‘Occupational and Safety and Environmental Protection’ convened by the Institute of Criminology, 25 October 1989). 113 Parker, Braithwaite and Stepanenko (n 108) 23. 114 Spencer W Waller, ‘Prosecution by Regulation: The Changing Nature of Anti-trust Enforcement’ (1998) 77 Oregon Law Review 1383, 1408. 115 See, for example, Australia – s 87B Competition and Consumer Act 2010 (Cth).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

294 Comparative competition law

important. First, in some instances other compliance strategies are likely to be ineffective. Examples of such conduct include serious, international cartels and recidivism. As Braithwaite notes: More costly punitive attempts at control are thus held in reserve for the minority of cases where persuasion fails. When it does fail, the most common reason is that an actor is being a rational calculator about the likely costs of law enforcement compared with the gains from breaking the law. Escalation through progressively more deterrent penalties will often take the rational calculator up to the point where it becomes rational to comply.116

Second, enforcement actions also have a role to play in supporting the effectiveness of other compliance strategies. Ayres and Braithwaite describe well the theory that underpins the compliance pyramid, which has been seen to play out in practice, they note: Escalation up [the] pyramid gives the state greater capacity to enforce compliance but at the cost of increasingly flexible and adversarial regulation. Clear communication in advance of willingness by the state to escalate up the pyramid gives incentives to both industry and regulatory agents to make regulation work at lower levels of interventionism … The irony proposed [is] that the existence and signalling of capacity to get as tough as needed can usher in a regulatory climate that is more voluntaristic and nonlitigious than is possible when the state rules out adversariness and punitiveness as an option. Lop the tops [in the competition context, enforcement actions are at the top of the pyramid] and there is less prospect of self-regulation, less prospect of persuasion as an alternative to punishment.117

Braithwaite describes this as a ‘paradox’ in that ‘to the extent that [agencies] can absolutely guarantee a commitment to escalate if steps are not taken to prevent the recurrence of lawbreaking, then escalation beyond the lower level of the pyramid will rarely occur’.118 As Baker notes, ‘it is important that a public agency take a sufficient number of enforcement actions to remind businesses and their advisors that the agency enforcement should be regarded as a relevant risk’.119

116 John Braithwaite, ‘Fasken Lecture: The Essence of Responsive Regulation’ (2011) 44 University of British Columbia Law Review 475, 484. 117 Ayres and Braithwaite (n 107) 39. 118 Braithwaite, ‘Fasken Lecture’ (n 116) 489. 119 Baker (n 4) 247.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 24 /

Date: 13/10

JOBNAME: Duns PAGE: 27 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 295

The aforementioned paradox also explains the significant penalties that can be imposed on those who breach competition laws120 as well as the recent trend towards the criminalisation of cartel conduct.121 4.2 Political Influences Enforcement priorities may also be set by reference to political ideology or goals other than the promotion of economic efficiency, especially in emerging jurisdictions. When assessing enforcement priorities or strategies, it is important to keep in mind the different cultural and political forces that were the impetus for introducing competition laws. Whilst many countries introduce competition laws because of a genuine commitment to free market ideology, some jurisdictions (again, often emerging economies) introduce such laws because they are required to by international bodies such as the World Trade Organization or the World Bank. In jurisdictions that have been coerced into introducing competition laws, political forces are ‘sometimes as concerned with limiting the spread of rampant Western-style market competition as they are with stimulating it’.122 For example, the Japanese Fair Trade Commission struggled to enforce competition laws initially due to resistance from ministries and business, which is not surprising given that the laws were effectively forced upon the country by the United States after the Second World War.123 Further, in many parts of the world there remain ‘profound doubts as to the wisdom of market competition in the first place’.124 For example, China’s recently introduced competition laws are underpinned by objectives broader than the pursuit of economic efficiencies. The law is to assist China to establish and effect the ‘state practice socialist market economy’,125 regulate excessive government power, protect the 120

For example, in the United States fines of up to $100 million can be imposed, as can jail sentences of up to 10 years. In the EU, fines of up to 10 per cent of the infringer’s total turnover in the preceding year can be imposed. In Australia, infringers can be fined the greater of (a) $10 million; (b) three times the benefit derived or (c) where it is not possible to calculate the benefit derived, 10 per cent of annual turnover (or, where the infringer is part of a corporate group, 10 per cent of the group’s annual turnover). 121 See further Chapter 12. 122 Crane, Institutional Structure (n 7) 211–12. 123 Hiroshi Iyori and Akinori Uesugi, The Anti Monopoly Laws and Policies of Japan (Federal Legal Publications 1994) 213–19. 124 Crane, Institutional Structure (n 7) 212. 125 The Constitution of the People’s Republic of China has, since 1993, required that the state practise a socialist market economy.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

296 Comparative competition law

country’s security and to reduce the gap between social incomes.126 Often, non-efficiency criteria will trump efficiency concerns, whose benefits are less immediate. For example, it has been suggested that China’s competition laws may aim to stymie foreign investment rather than prohibit only anti-competitive mergers.127 Similarly, it has also been suggested that the Russian government uses competition policy to advance nationalistic objectives, not to promote economic efficiency.128 Even where economic development underpins competition laws, certain forms of anti-competitive conduct may escape punishment. For example, Crane has suggested that India’s concern to promote ‘long-term infrastructure rather than short-term consumer welfare … may result in the allowance of anti-competitive [behaviour] that ostensibly contribute[s] to “development” goals’.129 Political pressure or intervention is not limited to developing economies. Even in well-established jurisdictions it is not uncommon for short-term, highly public political issues to influence the development and enforcement of competition law. In 2011 the Australian federal government introduced two prohibitions against ‘price signalling’ (which the most recent review into Australia’s competition laws has sensibly recommended be repealed even though the provisions are yet to be applied). Responding to public concern over the collective market power of the ‘Big 4’ banks, both the Labour government and the Liberal coalition rushed to provide the solution. The government Bill became law and, demonstrating the political motivations underpinning the introduction of these prohibitions, the scope of the prohibitions was limited to the banking industry. During times of crisis it is not uncommon for competition law enforcement (particularly in the merger context) to be relaxed. For example, in the recent Global Financial Crisis, there were calls for ‘emergency’ mergers in the banking and automotive sectors to be allowed. Bush notes that political influence can, for example, ‘pressure

126 Yong Huang, ‘Coordination of International Competition Policies: An Anatomy Based on Chinese Reality’ in Andrew T Guzman (ed), Cooperation, Comity and Competition Policy (OUP 2011) 229, 233. 127 Mark Williams, ‘Foreign Investment in China: Will the Anti-Monopoly Law be a Barrier or Facilitator’ (2009) 45 Texas International Law Journal 127, cited in Crane, Institutional Structure (n 7) 213. 128 See OECD Economic Survey of Russia 2009: Improving Regulation in Russia’s Goods and Services Markets, cited in Crane, Institutional Structure (n 7) 213. 129 Crane, Institutional Structure (n 7) 213–14.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 297

antitrust agencies to approve mergers designed to “rescue” distressed industries, particularly if the firms in question are deemed “too big to fail”’.130 In fact, government intervention can extend beyond pressure with some governments having a right to clear mergers on public interest grounds.131 The treatment of the merger between Lloyds and HBOS132 in the United Kingdom provides an example of stability concerns trumping competition concerns during a period of crisis. The merger saw one of the main ‘challenger’133 banks acquired by one of its larger, established competitors. Under the recently replaced regime that applied in the United Kingdom at the time, mergers that raised significant competition concerns were investigated, at first instance, by the Office of Fair Trading. If the Office of Fair Trading found that the merger raised competition concerns, the merger would ordinarily be referred to the Competition Commission for formal investigation. Even though the Office of Fair Trading found that the Lloyds–HBOS merger raised competition concerns134 the merger was not referred to the Competition Commission because the Secretary of State for Business and Enterprise intervened and cleared the merger on public interest grounds, namely that the merger would promote stability in banking markets.135 Such an 130 Darren Bush, ‘Too Big To Bail: The Role of Antitrust in Distressed Industries’ (2010) 77 Antitrust Law Journal 277, 280. 131 The Australian Competition Tribunal also has the power to authorise mergers on public interest grounds. However, there is an explicit requirement that public benefits (which may be economic, such as economies of scale, or non-economic, such as environmental) outweigh any anti-competitive detriments the merger is likely to generate. That said, financial stability may count as a public benefit. 132 HBOS failed because the majority of its funding was wholesale (more than half of which was short term) and its equity capital was low. Increasingly unable to replace maturing wholesale funding, it was acquired by Lloyds in early 2009. 133 A challenger bank is a bank seeking to expand its market share, which suggests that its acquisition would impact on the competitive dynamics in the markets affected. 134 Office of Fair Trading, ‘Anticipated Acquisition by Lloyds TSB plc of HBOS plc – Report to the Secretary of State for Business Enterprise and Regulatory Reform’ (Office of Fair Trading 2008). See further Arlen Duke and Leela Cejnar, ‘Competition and the Banking Sector: Friend or Foe?’ 7 Law and Financial Markets Review 152. 135 Section 42 of the Enterprise Act 2002 gives the Secretary of State the power to give the Office of Fair Trading an intervention notice if he or she believes that the case raises public interest considerations. Relevant public

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 27 /

Date: 13/10

JOBNAME: Duns PAGE: 30 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

298 Comparative competition law

approach is not supported by most competition law scholars and enforcement agencies who generally believe that the competition laws (which aim to bring about efficiency gains in the long term) should not be subordinated to promote short-term stability in financial markets.136 This is because there is a ‘tenuous connection between economic emergencies and antitrust … there is no evidence that antitrust law has ever so affected investment incentives that it caused or contributed to an economic crisis’.137 Nevertheless: The pressures of an economic crisis can make short-term stability look particularly attractive. Faced with depositors who might lose their savings or workers who might lose their jobs, the policymakers’ objective might understandably be to prevent the imminent harm. Where such stability can be quickly achieved through industry consolidation, there is a temptation to discount future costs of crisis-averting mergers.138

4.3 International Cooperation Finally, it is necessary to consider the enforcement implications of globalisation. As Fox has noted: ‘Antitrust law has moved from a national enterprise to an international enterprise. Markets transcend national boundaries, and many problems appear to require supranational or cooperative solutions.’139 With this in mind, international enforcement efforts are becoming better coordinated. There is increasing and better cooperation between enforcement agencies through fora like the OECD (Organisation for Economic Co-operation and Development), the ICN (International Competition Network) and regional competition networks

interest considerations are set out in s 58. Although the stability of financial markets was not listed as a relevant public interest consideration at the time of the Secretary of State’s intervention, this was not problematic. This is because s 42(3) expressly provides that the public interest categories do not need to be specifically listed at the time of intervention. This allows for the passage of legislation that retroactively adds public interest categories to s 58. 136 See further Daniel A Crane, ‘Did We Avoid Historical Failures of Antitrust Enforcement During the 2008–2009 Financial Crisis?’ (2010) 77 Antitrust Law Journal 219. 137 Howard A Shelanski, ‘Enforcing Competition During an Economic Crisis’ (2010) 77 Antitrust Law Journal 229, 230. 138 Ibid. 139 Eleanor M Fox, ‘Antitrust Without Borders: From Roots to Codes to Networks’ in Guzman (ed) (n 126).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 28 /

Date: 13/10

JOBNAME: Duns PAGE: 31 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Public enforcement 299

as well as through multi- or bilateral cooperation agreements140 and provisions in free trade agreements. Ling identifies the three basic forms of cooperation.141 First, there is procedural cooperation over investigation, obtaining and sharing evidence and regular consultations. Second, there is positive comity, which allows one country to request the other country to remedy anti-competitive conduct originating in the other country but which affects the first/requesting country. Third, there is convergence, which would involve the unification of competition law rules at an international level. Ling then notes ‘[a]ll three of these cooperative methods have now developed to a certain degree in international antimonopoly cooperation, but to varying extents’.142 Cooperation agreements promote the first and second forms of cooperation. The best-known examples are the 1991 and 1998 agreements between the EU and the US, which have been very effective. Cooperation between agencies is also facilitated by instruments like bior multilateral assistance agreements, extradition treaties and free trade agreements (which often include basic provisions regarding cooperation in competition law matters). Such agreements typically provide for various forms of procedural cooperation and can also impose positive comity obligations. Enforcement agencies may also coordinate searches, enforcement actions and exchange information. The United States agencies, for example, have entered into multilateral cooperation agreements with approximately 80 jurisdictions.143 Such agreements allow for an increased level of consultation, cooperation and coordination among agencies. The third form of cooperation identified by Ling, convergence, is promoted through less formal channels. The ICN, devised by the United States and described by Swaine as a form of ‘“soft” multilateral cooperation’,144 provides an international forum focussing on competition law enforcement and international cooperation among enforcement agencies. It aims to spread best practice and encourages both procedural and

140 Many cooperation agreements of OECD member countries are based on the 1995 OECD recommendation. 141 Dong Ling, ‘Cooperation, Comity and Competition in China’ in Guzman (ed) (n 126) 121, 137. 142 Ibid. 143 Antitrust Division Manual, Fourth edition, VII-32 (4th edn, US Department of Justice 2012). 144 Edward T Swaine, ‘Cooperation, Comity and Competition Policy’ in Guzman (ed) (n 126) 1.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 29 /

Date: 8/9

JOBNAME: Duns PAGE: 32 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

300 Comparative competition law

substantive convergence.145 The ICN involves those outside government circles in its deliberations including academics, private sector lawyers and economists.146 The OECD Competition Committee frequently holds roundtables on competition law issues at which senior agency representatives present their views on these issues. If consensus is reached, the OECD prepares recommended practices147 as part of a broader agenda to foster international cooperation and convergence.148 It also provides peer reviews of member countries’ competition law systems. Although they have no law-making power, ‘less formal institutions are becoming generators of soft law that may harden into world law’.149 As Geradin, Reysen and Henry note: [T]he work carried out by the above-mentioned soft law antitrust institutions has proved fruitful. The high-level discussions on contemporary and pressing competition law issues have led to the adoption of numerous (nonbinding) recommendations and best practices providing a useful point of reference for both practical enforcement competition laws, and cooperation between a large number of competition authorities and their respective interlocutors.150

145

Antitrust Division Manual, Fourth Edition, VII-33. See also Damien Geradin, Marc Reysen and David Henry, ‘Extraterritoriality, Comity and Cooperation in EU Competition Law’ in Guzman (ed) (n 126) 21. 146 Daniel Sokol, ‘International Antitrust Institutions’ in Guzman (ed) (n 126) 201. 147 Ibid 187, 198–9. 148 Geradin, Reysen and Henry (n 145) 21, 42. 149 Eleanor M Fox and Michael J Trebilcock, ‘Introduction’ in Eleanor M Fox and Michael J Trebilcock (eds), The Design of Competition Law Institutions (OUP 2013) 12. 150 Geradin, Reysen and Henry (n 145) 21, 42.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 14_Chapter11

/Pg. Position: 30 /

Date: 13/10

JOBNAME: Duns PAGE: 1 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

12. Criminalizing cartels: A global trend? Gregory C. Shaffer, Nathaniel H. Nesbitt and Spencer Weber Waller *

1. INTRODUCTION For most countries, competition law is a relatively recent phenomenon. In 1950, only ten jurisdictions had competition laws. Today more than 120 jurisdictions do and more than eighty of these systems have commenced since 1980.1 Officials from these jurisdictions now interact regularly through networked organizations, such as the International Competition Network (ICN) and the Organisation for Economic Co-operation and Development (OECD), as well as the United Nations Conference on Trade and Development (UNCTAD). There is no binding multilateral treaty mandating that countries have a competition law system that prescribes certain behavior, or create specialized institutions to enforce these prescriptions, although the North American Free Trade Agreement and a number of other free trade agreements signed by the United States (US) have short provisions that require countries to adopt and enforce laws against anti-competitive business conduct. Yet such prescriptions and institutions now exist in all major economies, creating what can be viewed as a transnational legal order, a collection of formalized legal

* The views expressed here are ours and do not reflect the views of Hogan Lovells. We thank Caron Beaton-Wells, John Connor, Dan Gifford, Christopher Harding, Hugh Hollman, and Robert Kudrle for their comments, and Jennifer Fair, Isaac Swaiman, and Mary Rumsey for research assistance. 1 Rachel Brandenburger, ‘The Many Facets of International Cooperation at the Antitrust Division’ (International Bar Association Midyear Conference, Madrid, Spain, June 15, 2012) ; John Fingleton, ‘The International Competition Network: Planning for the Second Decade’ (9th Annual ICN Conference, Turkey, April 2010) .

301

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

302 Comparative competition law

norms and associated organizations and actors that authoritatively order the understanding and practice of law across national jurisdictions.2 Combating international cartels is one of the central goals of this transnational legal order, and there appears to be greater normative convergence on this antitrust issue than any other among governments. Countries around the world have increased sanctions against cartels, including in many cases adopting criminal sanctions for the first time, with the term cartels now commonly harnessed to the unsavory epithet ‘hard core’ to signify cartels engaged in price fixing, output limitations, market divisions and bid rigging.3 More than 30 countries have criminalized cartel conduct in some form. All but five have done so since 1995 and over 20 since 2000, and the list is growing.4 Many states have initiated prosecutions, several have secured convictions, and a few have imposed jail time for these offenses. Others are significantly increasing the amount of fines for cartel behavior, such that they can be viewed in punitive terms, whether they are formally of a criminal or administrative law nature. Around 60 countries now combine enhanced sanctions with a leniency program pursuant to which the first to confess is immunized from public criminal or civil prosecution, adopting a carrot and stick approach to destabilizing and deterring cartels. There is, in short, a global trend toward enhanced sanctions combined with common enforcement techniques. Former US Deputy Assistant

2

Gregory Shaffer, ‘Transnational Legal Process and State Change’ (2012) 37 Law and Social Inquiry 229; Terence Halliday and Gregory Shaffer, ‘Transnational Legal Orders’ in Terence Halliday and Gregory Shaffer (eds), Transnational Legal Orders (CUP 2015). 3 See, for example, Christopher Harding, ‘Business Collusion as a Criminological Phenomenon: Exploring the Global Criminalisation of Business Cartels’ (2006) 14 Critical Criminology 181, 182, 188: ‘The closing years of the century then witnessed an apparent sea change … Europe, and also jurisdictions elsewhere throughout the world, seemed to be engulfed by a project of criminalization.’ For an excellent book on the phenomenon, see Caron BeatonWells and Ariel Ezrachi (eds), Criminalising Cartels: Critical Studies Of An International Regulatory Movement (Hart Publishing 2011). 4 See also Janet L McDavid and Megan Dixon, ‘Antitrust Update: Criminal Antitrust Enforcement in a Down Economy’ (2011) Antitrust update, 1011, 1015: ‘There are now over 100 countries with antitrust regimes, and over two dozen that have criminalized cartel activity.’ Reporting over 20: Caron Beaton-Wells, ‘Cartel Criminalisation as Cultural Change: A Report from Findings of a Survey of the Australian Public’ (Japanese Fair Trade Commission October 5, 2010) 6 .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 303

Attorney General for Criminal Enforcement, Scott Hammond thus proclaimed, ‘[i]n the last two decades, the world has seen the proliferation of effective leniency programs, ever-increasing sanctions for cartel offenses, a growing global movement to hold individuals criminally accountable, and increased international cooperation among enforcers in cartel investigations.’5 Although the US remains the primary user of criminal law as an enforcement tool, and the enforcement record outside the US is relatively slim, the debate over criminalization is active.6 This debate reflects a general shift in government attitudes toward cartels, which earlier were viewed in more benign (or even positive developmental) terms, since cartels could contribute to price stability and labor peace on account of the long-term employment that stable firms with ensured profit margins can offer.7 This state of affairs raises a series of questions. What spurred the trend toward increased sanctioning and criminalization of cartel activity? Has the move been almost entirely driven by criminalization evangelists such as the US Department of Justice (DOJ), working unilaterally, bilaterally, and through transnational networks? Is the primary explanation cognitive in terms of shared norms of being a ‘modern’ regulatory capitalist state, as stressed by global polity theory?8 Or does the shift simply reflect a rational response to the rise of international cartels operating in more than one jurisdiction as part of economic globalization? In other words, is the trend toward criminalization a functional response to regulatory difficulties that transcend national borders in an economically interdependent world, as rational institutionalist theory stresses? Or are organic, bottom-up, national processes also at work? Are we witnessing a change in ideological and moral sensibilities around the globe regarding cartel conduct? What is the impact of formal legal change on actual practices within countries, both in terms of government investigations 5 Scott D Hammond, ‘Criminal Policy Speeches: The Evolution of Criminal Antitrust Enforcement Over the Last Two Decades’ (Department of Justice – Antitrust Division February 25, 2010) 1 . 6 Beaton-Wells (n 4) 6. 7 ‘[E]conomic theorists and political leaders of the interwar period had often praised the potential value of cartels’: David Gerber, Global Competition Law: Law, Markets and Globalization (Oxford University Press 2010) 178. See also Christopher Harding and Julia Joshua, Regulating Cartels in Europe (Oxford University Press 2010). 8 For an excellent assessment of different explanations of policy diffusion, see Beth Simmons, Frank Dobbin and Geoffrey Garrett (eds), The Global Diffusion of Markets and Democracy (Cambridge University Press 2008).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

304 Comparative competition law

and private behavior? What are the mechanisms driving the criminalization of cartel activity and the enforcement of anti-cartel policies, and what are the primary obstacles to change? What, in sum, does the criminalization trend tell us about transnational legal processes and their limits?9 This chapter reviews and assesses trends in the criminalization and enforcement of cartel offenses around the world, highlighting their notable features. It notes the change in focus in the US toward addressing international cartels in the 1990s, followed by a shift in focus toward criminal penalties, including jail time for foreign individuals. The US has advocated these changes in international fora, most notably the OECD and the ICN. It has been relatively successful in its efforts to convince countries to enhance sanctions against cartel activity, and to use new enforcement tools such as leniency programs. We thus note the role of US advocacy in these developments. Yet we do not attempt to answer whether the change is rational interest-based or cognitive, since we find that there is support for both explanations.10 Rather, we focus on developments in formal law compared to actual enforcement, the mechanism of diffusion through networks, and the challenges posed for formal criminalization policies to be applied. Criminalizing competition offenses and actually enforcing those prohibitions are two different propositions. Section 2 addresses changes in the law – the ‘law on the books’ – in a number of jurisdictions, starting with the US and the European Union (EU). It highlights the enactment around the globe of increasingly severe penalties against cartel activity, and the increased adoption of criminal sanctions, which follows the US lead. Section 3 examines enforcement trends in the US, the EU, and other countries, starting with the increased focus of US enforcement policy on international cartels and on the application of criminal sanctions against individuals. Section 4 addresses transnational cooperation efforts in combating cartels, focusing particularly on the role of the US-initiated ICN, as well as the ongoing substantive and peer-review work of the OECD. Section 5 examines two new developments regarding enforcement techniques, the enhanced threat of extradition for cartel activity on account of mutual criminalization, and the proliferation of leniency programs to destabilize cartels, which are 9

For an analytic framework for assessing transnational legal processes and their impact, see Shaffer (n 2); Halliday and Shaffer (n 2). See also Harold Hongju Koh, ‘Bringing International Law Home’ (1998) 35 Houston Law Review 623. 10 To start, rational interest-based responses operate within particular cognitive frames, which affect the policy goals pursued.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 305

used as a carrot to complement the stick of more severe sanctions. Section 6 addresses the central impediments to these trends, and in particular the challenge of obtaining sufficient social and political support for criminalizing cartel activity and the impact of different institutional legacies, especially where criminal and competition law responsibilities are divided among different authorities.

2. LEGAL CHANGE ACROSS COUNTRIES: THE LAW ON THE BOOKS 2.1 Overview of Policy Developments While 20 years ago cartel prosecutions were largely the province of the US DOJ, a broad range of countries have now adopted enhanced sanctions against cartel offenses, including criminal sanctions now being available in over 30 countries. Around 35 countries have criminalized cartel conduct in some form; most of them since the mid-1990s, and the list is growing. We compile these countries as of the end of 2012 in Table 12.1 at the end of this chapter, noting the date of criminalization, the conduct covered, and the sanctions provided. We also highlight those countries that appear to be more serious about criminal prosecution in practice, although in many cases we do not yet know how sanctions will be applied because of the novelty of the legal trend. The leader in the move toward enhancing sanctions against cartels has been the US, now joined by the EU. Although the EU itself applies only administrative fines to undertakings and not individuals, these fines are punitive enough that EU Advocates-General view them as ‘quasicriminal,’ and over half of the EU member states have criminalized at least some forms of cartel activity. More recently, other OECD countries have followed suit, as have the BRICS – Brazil, Russia, India, China and South Africa. As John Connor writes, ‘with the adoption of an antitrust law in China in 2007, virtually all the world’s leading economies have made cartels illegal.’11 Whereas cartel activity earlier was not viewed as problematic, and seen as possibly even conducive, to national economic development, most leading economies have now significantly stiffened sanctions against cartels, reflecting a sense of social opprobrium, at least 11

John M Connor, ‘Latin America and the Control of International Cartels’ in Eleanor M Fox and D Daniel Sokol (eds), Competition Law and Policy in Latin America (Hart Publishing 2009) 291, 309.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

306 Comparative competition law

in the books. The trend toward punitive administrative sanctions has facilitated the move toward criminal sanctions as an enhanced deterrent. Yet scholarly attempts to assess global patterns of criminalization are limited.12 Christopher Harding rightly notes that any such attempt will necessarily be plagued by problems of definition and legislative complexity.13 Some systems, for example, prosecute cartel offenses as both administrative and criminal offenses. It thus may be difficult as a practical matter to disentangle administrative fines and surcharges from ‘criminal fines.’ The growing trend toward ‘criminalization’ (broadly construed) also masks crucial differences between jurisdictions. Harding concludes that empirical study reveals ‘very much a patchwork of criminalization’ involving local legal traits.14 Apart from Canada and the US, little of other countries’ ‘criminal legislation pre-dates the 1980s, and much of it has a more partial or tentative character.’15 For example, criminal sanctions can be imposed for bid rigging in Austria, Germany, Hungary, Italy, and Poland, but such conduct can be viewed as more akin to fraud and thus distinct (for such jurisdictions) from that of cartel offenses. In some jurisdictions, only criminal fines may be imposed, and in many others, only fines are used in practice, so that they lack the social sanction and potential deterrence value of jail time. Canada, for example, prosecutes individuals, but even where successful, the individuals typically only receive a conditional sentence pursuant to which they provide community service.16 While European, Latin American, and Asian jurisdictions have all criminalized competition offenses, ‘there is no systematic pattern.’17 Criminalization of cartel conduct is of a recent vintage, and more limited in scope, outside the US.18 Harding identified only nine countries in 2006 that are ‘to some extent self-consciously targeting cartel activity by means of criminal law within the context of the recent international campaign against cartels,’ listing Brazil, Canada, France, Germany, 12 Outside of North America, this feature fits with the general paucity of academic attention to business crime. Harding’s 2006 study is directly on point: Harding (n 3) 198. 13 Ibid 189. 14 Ibid. 15 Ibid 190. 16 Telephone Interview with Department of Justice (US) official (June 10, 2011). 17 Harding (n 3) 190. 18 Ibid 190–91.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 307

Ireland, Japan, Norway, the United Kingdom (UK), and the US.19 This figure, however, has increased since then. An October 2010 Cartels Workshop organized by the ICN placed the figure at more than 20 countries;20 we show that the figure was more than 30 in 2012, and the list is growing. 2.2 US Developments The US has long been the global leader in aggressively pursuing competition policy.21 The US first extended the jurisdictional reach of its law to combat foreign-based cartels. As this move was insufficient, the US has since attempted to export anti-cartel policies and enforcement techniques abroad, in particular through the OECD and the ICN discussed in Section 4. The US has criminalized agreements and conspiracies to restrict competition since 1890 with the passage of the original Sherman Antitrust Act, although it would be over 70 years before jail sentences for antitrust violations became common.22 Originally only a misdemeanor, violation of the Act became a felony in 1974. The US subjects both corporations and individuals to punishment, casting ‘[t]he net … widely in terms of both conduct and persons.’23 Cartel participants face large fines, possible imprisonment, and treble damages in civil actions. The Antitrust Criminal Penalty Enhancement and Reform Act of 2004 significantly increased the maximum penalties for antitrust offenses.24 The Act increased the maximum corporate fine from $10 million to $100 million; increased the maximum individual fine from $350,000 to $1 million; and raised the maximum jail sentence from three to ten years. The longest 19

Ibid 191. Beaton-Wells (n 4) 6. 21 Harding (n 3) 190. 22 Sherman Antitrust Act, ch. 647, §§ 1–8, 26 Stat. 209 (1890) (codified as amended at 15 USC. §§ 1–7 (1982)). The Act became law on July 2, 1890. See Donald Baker, ‘Why Is the United States So Different from the Rest of the World in Imposing Serious Criminal Sanctions on Individual Cartel Participants?’ (2011) 12 Sedona Conference Journal 301, 302–04. See generally John Herling, The Great Price Conspiracy: The Story of the Antitrust Violations in the Electric Industry (Greenwood Press 1962): detailing revival of criminal antitrust enforcement against nationwide electrical equipment cartel in the early 1960s. 23 Harding (n 3) 190. 24 Pub. L. No. 108-237, 118 Stat. 661 (2004) (codified as amended in scattered sections of 15 USC). 20

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 7 /

Date: 13/10

JOBNAME: Duns PAGE: 8 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

308 Comparative competition law

prison sentence for an individual to date has been 60 months.25 If the top end of the range applicable under the US Sentencing Guidelines exceeds the statutory maximum, then an alternative sentencing provision available for federal felonies may be employed which can increase fines to twice the amount of harm caused or the gains derived.26 The DOJ routinely invokes this provision to justify fines in excess of the statutory maximum. The largest criminal fine under this provision has been $500 million.27 The Antitrust Division of the DOJ relies heavily on its amnesty and leniency program in its criminal enforcement efforts.28 The amnesty program grants automatic immunity from criminal prosecution if an entity or individual reports illegal conduct which the Antitrust Division was not previously aware of and truthfully cooperates with the Division in its investigation and any resulting prosecution. The cooperating entity must further not have been the ringleader of the cartel nor have coerced others to participate. Discretionary immunity or reduced sentences and fines are available to entities and individuals who cannot meet the full criteria set forth above. To further induce cartel offenders to come forward, recent legislation also amended the leniency program to limit the civil damages exposure of an amnesty applicant from treble damages to the actual damages caused, in addition to providing immunity from criminal prosecution.

25 US Dep’t of Justice, Press Release, Former Sea Star Line President Sentenced to Serve Five Years in Prison For Role in Price-Fixing Conspiracy Involving Coastal Freight Services Between the Continental United States and Puerto Rico, (Dec. 6, 2013), available at . Recently, the Department of Justice secured twoyear prison sentences for two Japanese executives – the longest sentence of a foreign national who has submitted to the United States’ jurisdiction: Department of Justice, ‘Antitrust Division Update’ (Department of Justice Website 2013) . 26 Criminal Fine Improvements Act, Pub. L. No. 100-185, 101 Stat. 1279 (1987) (codified as amended at 18 USC 3571(d) (2006)). We thank Professor Caron Beaton-Wells for this point. 27 ‘DOJ AU Optronics Corporation Executive Convicted for Role in LCD Price-Fixing Conspiracy’ (Department of Justice Website September 18, 2012) . AU Optronics was fined $500 million, matching the largest fine imposed against a company violating US antitrust laws. 28 Department of Justice (US) ‘Corporate Leniency Policy’ (Department of Justice 1993); Department of Justice ‘Leniency Policy for Individuals’ (Department of Justice 1994).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 309

2.3 EU and EU Member State Developments Historically, Europe has approached antitrust regulation very differently than the US. While the US approach has been characterized by criminalization and treble damage civil claims, Europe’s approach had long been administrative, involving a more consensual process which was often based on voluntary adoption of practices, and which avoided penal sanctions and critical language suggesting moral approbation.29 Europe’s approach began to change in the early 1980s, and has since become increasingly ‘American’ in its tough-minded approach toward cartels, focusing on their adverse economic impact on consumers.30 Although EU authorities are still unable to impose prison terms or fine individuals for antitrust violations, and there are significant constitutional and institutional limitations on their ability to do so, they have become much more aggressive in anti-cartel enforcement and there is now a debate as to the advisability of adding a criminal law component to EU competition policy.31 The EU has significantly increased its penalties for cartel offenses in recent years to the extent that EU Advocates-General have called them ‘quasi-criminal.’32 EU authorities now can assess fines up to 30 percent of a company’s annual sales in connection with the prohibited activities 29

Gerber (n 7) 159–203; see generally David Gerber, Law and Competition in 20th Century Europe (Oxford University Press 1998). 30 Harding (n 3) 186–87. Gerber noting ‘growing confidence in the intellectual foundations of US-style substantive law analysis within the [EU] competition directorate’: Gerber (n 7) 201. 31 cf Ingeborg Simonsson, ‘Criminalising Cartels in the EU: Is There a Case for Harmonisation?’ in Beaton-Wells and Ezrachi (eds) (n 3); Peter Whelan, ‘A Principled Argument for Personal Criminal Sanctions as Punishments under EC Cartel Law’ (2007) 4 Competition Law Review 7; Katalin J Cseres and others (eds), Criminalization of Competition Law Enforcement: Economic and Legal Implications for EU Member States (Edward Elgar 2006). 32 Opinion of Advocate General Kokott, Case C-97/08, Akzo Nobel NV and Others v. Commission of the European Communities [2009] ECR I-8237, 39: ‘The consequence of the sanctionative nature of measures imposed by competition authorities for punishing cartel offences – in particular fines – is that the area is at least akin to criminal law’; Joined Opinions of Mr. Advocate General Vesterdorf, Case T-1/89, Rhone-Poulenc SA v. Commission [1991] ECR II-867, 3: referring to the ‘substance of the [competition] cases, which all broadly exhibit the characteristics of a criminal law case.’ See also, noting the punitive and quasi-criminal nature of the penalties involved, Philip Marsden, ‘Checks and Balances: EU Competition Law and the Rule of Law’ (2009) 22 Loyola Consumer Law Review 51, 55.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 9 /

Date: 13/10

JOBNAME: Duns PAGE: 10 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

310 Comparative competition law

multiplied by the number of years in which the offenses occurred. Fines also now include an ‘entry fee’ – or automatic fixed penalty – of between 15 and 25 percent of the participant’s annual sales in the affected sector.33 While the fines do not change the maximum penalty – 10 percent of a company’s total turnover in the preceding business year (whether or not in connection with the prohibited activity) – the revisions make it more likely that the fines will approach that limit.34 Among the more significant developments in the EU are the refinement of its leniency program in 2006 and the establishment of its cartel settlement procedure in June 2008, in both cases following the US lead.35 The leniency program, unlike the US model, is not backed by the threat of criminal sanction. Yet the program has proven crucial to uncovering cartels in Europe, with the majority of cases in recent years stemming from evidence obtained from a leniency applicant.36 The settlement procedure is designed to streamline the EU’s handling of cartel cases and thus free resources for new investigations. The alleged cartelist is able to review the EU’s evidence against it and determine whether to acknowledge its involvement and accept liability for violations of EU competition law in EU territory. Cooperating parties receive an automatic 10 percent reduction in penalties pursuant to the settlement program, but can obtain full immunity if they are the first to confess pursuant to the leniency program. The European Commission, through its Directorate-General for Competition (DG Comp), and EU national competition authorities established the European Competition Network (ECN) in 2004 to increase cooperation between EU and national authorities, and more effectively 33

European Commission (EC) Guidelines On The Method Of Setting Fines Imposed Pursuant To Article 23(2)(A) Of Regulation No 1/2003 [2006] OJ C210 2. 34 Gary R Spratling and D Jarrett Arp, ‘International Cartel Investigations: Evaluating Options and Managing Risk in Multi-Jurisdictional Criminal Antitrust Investigations’ [2010] Antitrust Counseling & Compliance 229, 241–43, 338, 1788 PLI/Corp 229, 242–43. 35 Harding and Joshua (n 7) 243; Commission Regulation (EC) 622/2008 amending Regulation (EC) No. 773/2004, as regards the conduct of Settlement Procedures in cartel cases [2008] OJ L171/3. 36 See Edmond noting that all cases adopted in 2010 were via immunity applications: Charlotte Edmond, ‘For Whom the Whistle Blows’ (Legal Week September 22, 2011). See also Alan Riley, ‘The Modernisation of EU Anti-Cartel Enforcement: Will the Commission Grasp the Opportunity?’ (Special Report, Center for European Policy Studies January 2010) 5 .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 10 /

Date: 10/9

JOBNAME: Duns PAGE: 11 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 311

share enforcement tasks.37 The underlying regulation also strengthened authorities’ investigatory powers by, for example, establishing their right to seal business premises and books or records, and to interview persons who may have useful information.38 The ECN has helped competition policy gain stature throughout the EU, and enhanced the authority of the DG Comp and the National Competition Authorities.39 The major focus of sanctions for EU member state authorities likewise is the imposition of administrative fines against corporate enterprises, embedded in a juridical framework.40 Starting in the late 1990s, and increasing more recently, however, some EU member states have begun to embark on a criminalization project. Over half of the EU member states now criminalize certain cartel offenses, including Austria, Belgium, Denmark, Estonia, France, Germany, Greece, Hungary, Ireland, Italy, Poland, Portugal, Romania, Slovakia, Slovenia, Spain, and the UK, and the list appears to be growing, although Austria, the Netherlands, and Luxembourg have decriminalized competition law (except for bid rigging in Austria).41 As discussed in Section 4, however, how such criminalization initiatives will operate in practice remains in question as the policies remain circumscribed compared to the US model.

37

Stephen Wilks, ‘The European Competition Network: What Has Changed?’ (Paper for EUSA Conference, Montreal, May 17–19, 2007) . 38 See Chavez (n 16) 964–65. See also Council Regulation (EC) 1/2003 On the Implementation of the Rules on Competition Laid Down in Articles 81 and 82 of the Treaty, arts. 19, 20 [2003] OJ L 1, 14–15. 39 Gerber (n 7) 190, 200. 40 Ibid 182. In addition the European Union has issued a directive requiring member states to enact effective private damage remedies for competition violations into national law by the end of 2016. Directive 2014/104/EU of the European Parliament and of the Council of 26 November 2014 on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union, available at http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/?uri=CELEX:20 40104&from=EN. 41 Stating that more than half of all EU member states have some form of criminal sanctions for some forms of cartel conduct: Philipp Girardet, ‘“What if Uncle Sam wants you?”: Principles and Recent Practice Concerning US Extradition Requests in Cartel Cases’ (2010) 1 Journal of European Competition Law and Practice 286, 287; Wouter Wils, ‘Is Criminalization of EU Competition Law the Answer’ in Cseres and others (eds) (n 31) 60, 74.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 11 /

Date: 13/10

JOBNAME: Duns PAGE: 12 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

312 Comparative competition law

2.4 Developments in other Countries In recent years, a wide range of other jurisdictions, at least formally, provide jail time for cartel offenses. Individuals now face potential imprisonment for cartel activity in Australia, Brazil, Canada, Iceland, Indonesia, Israel, Japan, Korea, Norway, Russia, Thailand, and Zambia, in addition to in the US and a majority of EU member states. Moreover, cartel participants are subject to complementary private civil damage actions in a growing number of countries, including Australia, Brazil, Canada, Japan, and New Zealand, as well as in Belgium, Germany, Ireland, and the UK within the EU. The global trend is toward providing for enhanced sanctions against cartels. Many OECD members (in addition to the US and EU member states) have stiffened penalties for engaging in cartel activities, which vary in their civil/administrative or criminal law nature. For example, Canada’s Budget Implementation Act of 2009 substantially changed the criminal enforcement regime for cartel offenses by raising the maximum penalty from 5 to 14 years in prison and the maximum fine from $10 to $25 million. Price fixing, market allocation, and output restrictions are now per se offenses under the law.42 And as of early 2012, an individual convicted and sentenced to prison under certain provisions of the Act no longer enjoys the prospect of serving his or her sentence in the community.43 Australia recently increased maximum fines to the greater of $10 million or three times the value of the benefit derived from the cartel. Where value cannot be determined, the law provides for a fine of 10 percent of annual turnover.44 In 2009, the Australian Parliament criminalized various cartel offenses, providing for up to 10 years in prison and a fine of $220,000 (Aus), and New Zealand is in the process of following Australia in this respect.45 In the summer of 2009, Japan

42 Spratling and Arp (n 34) 250; Bill C-10, An Act to implement certain provisions of the budget tabled in Parliament on 27 January, 2009 and related to fiscal measures, 2nd Sess., 40th Parl., 2009 [hereinafter Bill C-10]. Bill C-10 became S.C. 2009, c. 2 upon passage through Parliament. 43 Elisa Kearney and others, ‘Canadian Government Restricts Availability of Conditional Sentences (“House Arrest”)’ (Mondaq March 20, 2012) . 44 Spratling and Arp (n 34) 243. 45 A Bill imposing criminal sanctions for cartel behavior was introduced into Parliament in October 2011. Commerce (Cartels and Other Matters) Amendment Bill, 2011 No. 341-1, Digest No. 1942 (Oct. 13, 2011)

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 12 /

Date: 10/9

JOBNAME: Duns PAGE: 13 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 313

increased criminal sanctions for cartel offenses, changing its maximum prison sentence for cartel conduct or bid rigging from three to five years. Japan also raised the statute of limitations from three to five years, and restructured its new leniency program.46 In 2005 Korea likewise revised its competition laws to increase fines against cartel participants from a maximum of 5 percent to a maximum of 10 percent of sales in related goods or services, and to facilitate use of a leniency program.47 More recently, in 2011, Mexico’s Congress approved a new law introducing criminal sanctions of up to ten years in prison for collusion, as well as the ability to engage in surprise inspections, known as ‘dawn raids.’48 Even Switzerland, ‘where cartels were “endemic” to the economy,’ has recently passed a law providing for administrative fines of up to 10 percent of a firm’s total combined revenue for the preceding three years.49 The BRIC countries have also stiffened penalties for engaging in cartel activities. Brazil has emerged as the new leader in Latin America in combating cartels. Since 2003, Brazil’s competition system has eliminated overlapping functions, streamlined cartel investigations, and enhanced authorities’ enforcement tools through granting them the power to conduct ‘dawn raids’ and to use the leverage provided by new leniency and settlement programs.50 Federal and state prosecutors conduct cartel prosecutions, in cooperation with the agencies forming part of the Brazilian Competition Policy System (BCPS).51 The BCPS’s anti-cartel

: criminalizing price fixing, output restrictions, market allocation and bid rigging, and providing criminal penalties of, inter alia, up to seven years’ imprisonment for an individual. See also Chavez (n 16) 965–66; Ministry of Economic Development, ‘New Zealand, Cartel Criminalisation Discussion Document’ (Ministry of Economic Development (NZ) 2010) . 46 Spratling and Arp (n 34) 252, 309–10. 47 Ibid 243, 311. 48 ‘Monopolies in Mexico: Compete – Or Else’ (Economist May 7, 2011) 41. See also Omar Guerrero Rodríguez and Alan C Ramírez, ‘Mexican Competition Law Aligned Incentives for Effective Cartel Enforcement’ (Competition Policy International, October 2012) . 49 Chavez (n 16) 943. 50 OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2010) 7. 51 Ibid 7, 37.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 13 /

Date: 10/9

JOBNAME: Duns PAGE: 14 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

314 Comparative competition law

program has grown steadily, especially since 2006.52 Brazil’s program for criminally prosecuting cartels is recently among ‘the most active of all countries,’ and includes fines and prison sentences ranging from two to five years.53 Brazil’s leniency program, created in 2000, offers full immunity to the first cartelist to confess, or partial immunity where enforcers were already aware of the cartel. Brazil also introduced a Cartel Settlement Program under which competition authorities enter into settlements with companies that lost the race to apply for leniency. The settlement program has led to a number of settlements since 2007.54 Brazil has emerged as a regional expert in anti-cartel enforcement – having recently shared its growing expertise with Argentina, Chile, Paraguay, and El Salvador.55 Brazil’s competition authorities regularly engage in bilateral training and consultations with other national authorities. Among the other BRICs, China’s new anti-monopoly law went into effect in 2008, and in the summer of 2009, China’s industry and commerce agency announced new procedures governing antitrust investigations and enforcement.56 China’s Anti-Monopoly Law does not expressly provide for criminal penalties, but other criminal law provisions can and have been used against cartel participants (such as for ‘obstruction’ of justice).57 Similarly, under a law effective in late 2009, Russia now applies criminal sanctions for antitrust violations, and certain offenses are punishable by up to six years’ imprisonment.58 In India, ‘after a long and troubled gestation,’59 a new Competition Act began to take effect in 2009, which stiffens sanctions, including fines of up to 10 percent of an enterprise’s turnover,60 and prison terms for obstruction of 52

Ibid 14–15. Ibid 18. See also Spratling and Arp (n 34) 253. 54 OECD (n 50) 17. 55 Ibid 50–51. 56 Brendan Pierson, ‘China Adopts New Rules to Enforce Monopoly Law’ (Law 360 June 9, 2009) . 57 Jonathan Gowdy, ‘China Issues New Anti-Monopoly Rules and Procedures on Pricing Conduct’ (Mondaq January 20, 2011) ; Martyn Huckerby and others, ‘China’s Antitrust Tigers Grow Teeth’ (Mondaq September 1, 2010) . 58 Spratling and Arp (n 34) 254. 59 Aditya Bhattacharjea, ‘India’s New Competition Law: A Comparative Assessment’ (2008) 4 Journal of Competition Law and Economics 609, 609. 60 The Competition Act, 2002, No. 12 of 2003, art. 27. 53

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 14 /

Date: 13/10

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 315

justice.61 Sometimes associated as one of the expanded ‘BRICS,’ South Africa also in 2009 enacted criminal liability for directors and managers for certain competition law offenses,62 and has now created a cartel division within its enforcement agency.63 The impact of these legal changes, however, remains an open question in many countries for the reasons we discuss in Section 4.

3. ENFORCEMENT TRENDS ACROSS COUNTRIES 3.1 Overview of Policy Developments The prevalence of cartels is difficult to measure. Conspiracies are meant to go undetected, and many, of course, do. Yet because of more robust law enforcement tools and shifting attitudes about cartels, authorities are detecting more cartels than in the past. While many countries have law on the books, to what extent do different jurisdictions effectively enforce those laws? Answering this question is difficult. As Harding observed, it is still too early to assess the impact of ‘the enforcement of criminal law relating to cartels, when so much of this is of recent origin,’ but ‘some indications at a more general and comparative level [suggest] that implementation and enforcement [do] not match the rhetoric of law enactment.’64 Scholars and policymakers have attempted to get at the enforcement question in different ways. Economist John Connor has collected empirical data on enforcement in a large multi-jurisdictional sample.65 The ICN and OECD offer survey evidence regarding the perspectives of cartel

61

Ibid art 42. See also Bhattacharjea (n 59) 626. Competition Amendment Act of 2009 § 16 (S. Afr.). However, the Amendment Act provides that the President can determine when it enters into force; as of the beginning of 2015, it had not yet come into force. Global Legal Group, ‘The International Comparative Legal Guide to Cartels & Leniency’ (International Comparative Legal Guides 2011) 218 . 63 Noting appointment of official to the ‘newly established Cartels Division’: ‘Commission Appoints Head of Cartel Unit’ (AllAfrica.com May 6, 2011). 64 Harding (n 3) 190, 192. 65 See, for example, John M Connor, ‘Effectiveness of Antitrust Sanctions on Modern International Cartels’ (2006) 6 Journal of Industrial Competition and Trade 195. 62

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

316 Comparative competition law

authorities.66 Critically, however, empirical work on cartel enforcement can measure only detected cartels. One must therefore bear in mind that any conclusions drawing from this empirical work inevitably suffer from selection bias. Methodological difficulties notwithstanding, the overall global trend is toward enhanced enforcement, including higher fines and an increased emphasis on individual accountability with a view toward deterrence, although the level of enforcement varies significantly by jurisdiction as does its deterrent value. Connor undertook, to our knowledge, the first attempt to assess quantitatively, using a large multi-jurisdictional sample, the magnitude and pattern of global antitrust sanctions imposed against ‘international cartels,’ by which he refers to cartels involving members from more than one country.67 Connor presents data on such cartels discovered from 1990 to 2005 in a 2006 study, and maintains an annually updated database of all publicly reported sanctions. He selects 1990 as a date that roughly captures the beginning of the current level of sanctions and the harmonization of antitrust laws among the US, EU, and Canada. From 1990 to 2005 authorities took a total of 387 legal actions against 260 international cartels. Legal action includes ‘the launching of an official investigation, the filing of a private antitrust damages suit, or the imposition of one or more legal sanctions.’68 The US, Europe, and Canada commenced the vast majority of these legal actions – 128, 101 (including action by both EU member states and the EU) and 56, respectively. In contrast, only 17 actions were initiated in Asia, and only two in Latin America, against international cartels during this period. Authorities secured a total of 285 ‘convictions’ in these cases, which Connor defines to include consent decrees, settlement agreements, and warnings. He concludes that the data suggests that ‘antitrust authorities are by and large cautious about 66

ICN, ‘Trends and Developments in Cartel Enforcement’ (9th Annual ICN Conference, Istanbul, Turkey, 29 April 2010) (hereafter ‘ICN Trends’). The survey highlights six recent changes impacting cartel enforcement across the globe: (1) increased penalties, (2) changes in investigative powers, (3) new violation provisions or new definitions of what constitutes cartel offenses, other significant changes or development, (4) changes in leniency programs, (5) shifting perceptions of the importance of cartel enforcement, and (6) the use of the ICN’s Cartel Enforcement Manual to advance cartel enforcement. 43 of 46 respondents to the ICN survey, for example, noted that increased penalties have impacted cartel enforcement in that jurisdiction in the last ten years – the highest of any reported factor. 35 respondents pointed to enhanced investigative powers and leniency provisions as a major development. 67 Connor (n 65) 197. 68 Ibid 199.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 16 /

Date: 13/10

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 317

opening formal investigations in the sense that 90–95% of the cases investigated conclude with sanctions of some sort.’69 More recently, Connor updates and summarizes the data through 2008 with a view to uncovering trends across countries over time.70 Between January 1990 and November 2008 he finds that there were 516 ‘formal official investigations’ of suspected international cartels around the world, meaning that there were 129 new formal investigations (516–387) between 2005 and 2008 alone.71 This figure regarding ‘investigations’ includes cartels that were subject to raids, grand jury hearings, class actions, and indictments. At least 6,000 companies have been alleged or proven to be members of international cartels, including about 2,900 ultimate parent companies with known names and locations. At least 1,620 corporations have been fined.72 These cartels alone affected a total of around $16 trillion in sales, with the largest number trading in industrial goods, followed by business and consumer services.73 Connor finds that the trend across countries toward increased discovery of international cartels is pronounced, with rates of discovery being 15 times higher in 2005–2008 than before 1994, and having steadily increased between those dates.74 Enforcement patterns vary widely among states and regions. The number of indictments in the US and Canada peaked in the late 1990s, although this shift could suggest that past enforcement efforts and the use of criminal sanctions have been successful deterrents. At the same time, the severity of applied US sanctions has increased.75 In parallel, the EU has commenced an increasingly large number of cases, resulting in increasingly high sanctions. National competition authorities within the EU have been the biggest prosecutors of international cartels since 2000, and collectively they surpass all other agencies in terms of the number of 69

Connor derived the 90–95% figure in view of the fact that of 102 legal actions without convictions, 81 were still underway at the time of publication and only 21 were dismissed: ibid 200. 70 John M Connor, ‘Cartels & Antitrust Portrayed: Private International Cartels from 1990 to 2008’ (American Antitrust Institution, Working Paper No. 09-06, 2009) . 71 For about 20% of the cartels, no adverse government decision or private settlement had been reached at the time of his review. Investigations in 32 of the 516 were closed without sanctions because of insufficient evidence. Email from John Connor (June 6, 2011). 72 Connor (n 70) 5, 25. 73 Ibid 7, 12–14. 74 Ibid 18. 75 Ibid 17.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 17 /

Date: 8/9

JOBNAME: Duns PAGE: 18 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

318 Comparative competition law

international cartels uncovered.76 Other regions have also engaged in increased cartel enforcement, although cartel detections in Africa, Asia and Latin America are comparatively modest.77 Total global penalties assessed against international cartels from 1990 to 2008 are approximately $63.3 billion.78 Government fines account for more than half of this total ($35 billion), though private settlements stemming from civil suits (particularly from the US) are also significant, totaling approximately $29 billion.79 The European Commission has imposed the vast bulk of global fines, followed by national competition authorities within the EU, the US DOJ, and US state AttorneysGeneral.80 Far behind the US and EU, Korea leads the rest of the world in terms of fines imposed (with approximately $750 million), followed by Africa (less than $500 million), Latin America (approximately $300 million), other Asian countries, and Oceania, each with less than half of the Latin American total.81 Individuals increasingly risk being held accountable, especially in the US.82 At least 435 individual executives have been penalized, and 989 charged worldwide as of December 2008.83 Americans account for nearly one-third of all penalized executives.84 Although the number of individuals charged in the US fluctuated between 1990 and 2008, the likelihood of sentences has become greater and the amount of fines and length of prison sentences have become increasingly severe since 1990.85 The US, however, remains ‘almost unique’ when it comes to prison sentences: only Israel is another significant jurisdiction in that respect, followed by Japan.86 3.2 US Developments The US has historically taken the global lead in cartel enforcement. In the last two decades, it has done so regarding international cartels and has increasingly focused on criminal sanctions since the late 1990s, 76 77 78 79 80 81 82 83 84 85 86

Columns Design XML Ltd

/

Ibid 8–9, 17, 20–21. See also ibid 17. Ibid 51. Ibid. Ibid 55. These figures are approximate. See ibid 71. Ibid 80–97. Ibid 83–84. Ibid 82. Ibid 88–96. Ibid 82.

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 18 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 319

including incarceration of foreign defendants.87 The US reports that it has imposed more than 90 percent of fines in the past few years in connection with the prosecution of international cartel activity.88 The result has been increasing US prosecution of foreign defendants. Connor’s data shows that in the US, ‘prior to 1995, less than 2% of corporations accused of criminal price fixing were foreign-based firms; after 1997, more than 50% were non-US corporations.’89 The DOJ is typically investigating about 50 international cartels at any one time.90 Total annual criminal fines increased dramatically starting in 1997, with a record of $1.1 billion in FY 1999.91 Total criminal fines again exceeded $1 billion in 2009 and 2012.92 Though the DOJ does not target particular geographic regions or industry sectors, recent years have seen particularly robust enforcement against Asian corporations. Since 2005 the DOJ has imposed cartel fines of $10 million or greater on more companies headquartered in Asia than on those headquartered in every other country combined.93 The Antitrust Division of the DOJ emphasizes ‘that the most effective way to deter and punish cartel activity is to hold culpable individuals accountable by seeking jail sentences.’94 As one US official states, ‘our defendants routinely offer to pay large fines in lieu of going to jail, a plea that we reject, but they don’t offer to go to jail in lieu of paying a large fine.’95 Connor reports that the DOJ secured prison sentences for a total 87

John M Connor, ‘Global Antitrust Prosecutions of International Cartels: Focus on Asia’ (2008) 31 World Competition 575, 582. As noted above, Congress made violation of the Sherman Act a felony, rather than a misdemeanor, in 1974. There were some minor criminal domestic cases in the 1980s, which helped lay the groundwork for the major international cases in the 1990s. We thank Spencer Waller for stressing this point. 88 Chavez (n 16) 945. 89 Connor (n 87) 588. 90 Hammond (n 5) 3. 91 Chavez (n 16) 959–60. 92 Ibid. See also Department of Justice, Antitrust Division, ‘Spring 2012 Update’ (Department of Justice (US) 2012) . 93 Ron Knox, ‘The Longer Arm of the Law’ (Global Competition Review October 2, 2012) . 94 Hammond (n 5) 11. 95 R Hewitt Pate, ‘Criminal Policy Speeches: International Anti-Cartel Enforcement’ (Department of Justice – Antitrust Division November 21, 2004) . We thank Bob Kudrle for this.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 19 /

Date: 13/10

JOBNAME: Duns PAGE: 20 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

320 Comparative competition law

of 284 individuals in cartel cases from 1990 to 2007, of which an increasing number and proportion are foreigners.96 Since May 1999 nearly 50 foreign defendants have served or are serving prison sentences in the US for international cartel offenses or obstructing a cartel investigation.97 The percentage of defendants sentenced to jail in cartel cases, and the number of individuals per corporate defendant, have increased. The 1990s saw an average of 37 percent of defendants involve a jail sentence, whereas the 2009 average was 80 percent.98 Sentences are also getting harsher.99 The DOJ achieved its highest average jail sentence for all defendants in fiscal year 2007, with an average sentence of 31 months. That same year, it imposed a record of 31,391 total jail days against individuals, mostly pursuant to guilty pleas.100 3.3 The Vitamins Cases and their Impact In the late 1990s the US took the lead in a landmark case that had a significant impact on global anti-cartel practices. The ensuing cases in multiple jurisdictions illustrated the potentially grave consequences of cartel conduct for businesses, the public, and regulators. The prosecutions received significant media attention. The vitamins case grew out of the earlier lysine cartel which was captured on live video thanks to the informant Mark Whitaker, later portrayed in the 2009 motion picture ‘The Informant!’, starring Matt Damon. As US Deputy Assistant Attorney General Scott Hammond has observed, ‘the prosecution of the 96 John Connor, ‘Anti-Cartel Enforcement by the DOJ: An Appraisal’ (2008) 5 Competition Law Review 89, 111. 97 Observing further that 102 US citizens have been sentenced to prison terms for cartel activity since 1999: Gregory J Werden and others, ‘Criminal Policy Speeches: Recidivism Eliminated: Cartel Enforcement in the United States Since 1999’ (Department of Justice (US) – Antitrust Division September 22, 2011) www.justice.gov/atr/public/speeches/speech-criminal.html 4–5: see also Hammond (n 5) 7. 98 Hammond (n 5) 8; see also Connor (n 96) 111: noting the rise in the average number of individuals receiving prison sentences between 1990 and 2007. 99 Noting an average of 8 months’ imprisonment between 1990 and 1999, a 19-month average between 2000 and 2009, and a 24-month average in 2010– 2011: Department of Justice (US) Antitrust Division, ‘Spring 2012 Update’ (Department of Justice (US) 2012) . 100 Hammond (n 5) 7–8. See also Connor (n 96) 111: noting the increasing harshness of prison sentences.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 20 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 321

vitamin cartel … helped [to] trigger a rethinking of the adequacy of competition laws around the world.’101 The vitamins cases began with a federal class action complaint in Alabama and a federal grand jury investigation in Texas. The December 1997 Alabama complaint alleged a conspiracy among the three major vitamins manufacturers to fix prices and allocate sales.102 In May 1999 the DOJ announced plea agreements involving major pharmaceutical manufacturer F Hoffmann-La Roche Ltd and the German chemical manufacturer BASF Aktiengesellschaft. Stretching more than nine years, the conspiracy affected more than $5 billion in commerce. The government’s sentencing recommendation detailed an ‘extremely well organized operation,’ involving at least quarterly meetings and once-a-year meetings among high-level corporate officials to set an annual ‘budget,’ and project global sales volumes and prices. The conspirators’ efforts to conceal the plot were extensive, including the destruction of most documents after meetings and the disguising of conspirators’ identities in those documents not destroyed.103 Prosecutions in multiple countries followed the US litigation. In the end, the corporate cartelists agreed to pay unprecedented fines, globally totaling well over $3 billion (in June 2010 USD). This total breaks down as follows: in the US, $900 million in criminal fines and over $1.1 billion in civil settlements; in the EU, €790.5 million in fines; in Canada, $94.7 million (Can.) in fines and a $132.2 million national class action settlement; in Australia, $26.5 million (Aus.) in fines and a $30.5 million (Aus.) class action settlement (the first in Australia);104 in Brazil, R$17.6 million in fines.105 The US DOJ criminally prosecuted 12 corporations and 14 individuals. Eleven executives, including six Europeans, went to prison in the US. It was ‘the first time a foreign executive agreed to serve time in US prison for his participation in an international cartel.’106 While other jurisdictions imposed then-record national fines (Canada, the EU, Australia, and Korea, for example), only the US imposed jail time. Overall, the

101

Hammond (n 5) 9–10. See Harry First, ‘The Vitamins Case: Cartel Prosecutions and the Coming of International Competition Law’ (2001) 68 Antitrust Law Journal 711, 713–14. 103 See ibid 714–15. 104 Connor (n 87) 594. 105 Chavez (n 16) 936–37. 106 Hammond (n 5) 7. 102

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 21 /

Date: 13/10

JOBNAME: Duns PAGE: 22 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

322 Comparative competition law

vitamins prosecution resulted in greater attention to anti-cartel enforcement around the globe, and spurred the DOJ to increase international anti-cartel cooperative efforts, as discussed in Section 4. 3.4 EU and EU Member State Developments The EU initially tended to follow the US lead in sanctions against international cartels. The EU’s fines in cases involving the lysine, citric acid, vitamins, sodium gluconate, and graphite electrodes cartels lagged behind prosecutions in the US by two to five years.107 For example, the US DOJ announced its first indictment in the citric acid conspiracy in 1996, whereas the EU announced its decision in that case in 2001. Over time, however, the EU has steadily become more active in investigating and sanctioning cartels. Although the EU itself still does not impose a criminal law sanction for cartel activity, it has been increasingly aggressive in seeking large administrative fines. Many EU cartel prosecutions involve activity only in EU territory, whereas most of the US cases involve international or global cartels (that is, cartels involving perpetrators in more than one country, or foreign cartels whose activities have effects in more than one continent), suggesting two things. One, that the EU still faces significantly greater cartel activity within its own territory108 and, two, that the US continues to engage in a more active extraterritorial approach to enforcement. There has been ‘an explosion of enforcement’ against cartels in Europe since the mid-1990s, linked once more to the use of leniency programs.109 Over 90 percent of EU fines against cartels have been imposed since 1995.110 The EU overtook the US in 1999 in terms of the amount of fines imposed. As of 2009 the EU had investigated more than 4,300 companies and penalized more than 1,550 of them with most penalties paid by European firms.111 And those figures continue to grow.112 Fines totaled €1.756 billion in 2001 and a record €3.334 billion in 2007. In 2012 the EU secured the highest cartel fine to date: approximately €1.471 107

Chavez (n 16) 960. For example, Connor notes that more than half of discovered cartels operated within Europe: Connor (n 65) 9. 109 Harding and Joshua (n 7) 145. 110 Connor (n 87) 589. 111 Connor (n 65) 50. 112 See generally European Commission, ‘Cartel Cases’ (European Commission Website) . 108

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 323

billion in connection with the TV and computer monitor tubes cartel.113 Indeed, the amount in fines collected by the EU in recent years is many times the amount collected by the Antitrust Division of the DOJ (although it must be recalled that the DOJ also uses criminal sanctions, and the US system also includes more extensive private domestic damage suits).114 On May 19, 2010 the EU announced its first settlement under its 2008 settlement procedure, pursuant to which ten producers of Dynamic Random Access Memory (DRAM) chips, agreed to pay a total of €331,273,800 in fines.115 Participants received a 10 percent settlement discount for admitting fault and adhering to settlement procedures. Six settlements have thus far been reached under the 2008 procedure.116 A few EU member states now impose criminal sanctions against cartels under domestic law, although a number limit criminal sanctions to the offense of bid rigging. Between October 2005 and December 2009 the Irish competition authority had secured 33 criminal convictions for cartel participants, although no one has gone to jail as the courts have uniformly suspended the prison sentences.117 This figure includes the first prison sentence in Europe for a non-bid rigging offense, a six-month suspended sentence for the central figure in a cartel involving the home heating oil industry.118 Ireland prosecuted a total of 24 individuals, and

113 European Commission, ‘Statistics’ (European Commission December 5, 2012) . 114 Compare ibid with Antitrust Division, Department of Justice, ‘Spring 2012 Update’ (Department of Justice 2012) . See also Spratling and Arp (n 34) 237. 115 James Kanter, ‘An Old Chip Cartel Case Is Brought to a Swift End’ New York Times (May 20, 2010) B13. 116 For example, European Commission, ‘Antitrust: Commission Fines Producers of Water Management Products €13 million in Sixth Cartel Settlement’ (European Commission Website June 27, 2012) (IP/2/704); European Commission, ‘Antitrust: Commission Fines Producers of Washing Powder €315.2 Million in Cartel Settlement Case’ (European Commission Website April 13, 2011) (IP/11/473); European Commission, ‘Antitrust: Commission Fines DRAM Producers €331 Million for Price Cartel, Reaches First Settlement in a Cartel Case’ (European Commission Website May 19, 2010) (IP/10/586). 117 Patrick Massey and John D Cooke, ‘Competition Offences in Ireland: The Regime and its Results’, in Beaton-Wells and Ezrachi (eds) (n 3) 120. 118 Ibid 122; Spratling and Arp (n 34) 251. The conviction came in the spring of 2006.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

324 Comparative competition law

secured 18 convictions, in connection with that cartel.119 Although Germany has only criminalized the offense of bid rigging, it indicted over 260 persons for this offense during the first 11 years of the statute (between 1998 and 2008), and more than 180 of these individuals were convicted, many of whom are serving jail time.120 Even in tiny Estonia, authorities initiated eight criminal cases in 2009, three of which were referred to the public prosecutor for court proceedings. Estonia continues to prosecute several cartel cases each year and has been aided by its adoption of a leniency program in 2010.121 Finally, the UK has begun to prosecute cartelists under its 2002 Enterprise Act. The 2007 marine hose prosecution was the first pricefixing case brought under the 2002 law, and resulted in the first criminal sanctions ever imposed for competition law violations in the UK. The UK sentenced three executives involved in the marine hose conspiracy to prison terms of 30, 24, and 20 months, following a US plea bargain (which made the prosecution much easier and the case thus unique).122 Soon after, in August 2008, the government charged four current and former British Airways executives with fixing prices of fuel surcharges for passenger flights.123 That trial, however, collapsed in 2010 after a series of procedural and evidentiary failings on the part of the prosecution.124 Many commentators view this episode as a severe setback for the 119

‘Manager of Galway heating oil company sentenced for price-fixing’, Irish Times (May 4, 2012) . 120 Florian Wagner-von Papp, ‘What If All Bid-Riggers Went To Prison and Nobody Noticed? Criminal Antitrust Law Enforcement in Germany’, in BeatonWells and Ezrachi (eds) (n 3) 158. 121 Estonian Competition Authority, ‘Annual Report 2010’ (Estonian Competition Authority 2010) ; Estonian Competition Authority, ‘Annual Report 2009’ (Estonian Competition Authority 2009) . 122 Donald Baker, ‘An Enduring Antitrust Divide Across the Atlantic over Whether to Incarcerate Conspirators and When to Restrain Abusive Monopolists’ (2009) 5 European Competition Journal 145, 165. See also Michael O’Kane, ‘Does Prison Work for Cartelists? The View from Behind Bars’ (2011) 56 Antitrust Bulletin 483: interview of defendant by his counsel in the marine hose litigation. 123 Spratling and Arp (n 34) 253. 124 See Julian Joshua, ‘Shooting the Messenger: Does the UK Criminal Cartel Offense Have a Future?’ (Antitrust Source August 2010) 1–2 : describing factors leading to the collapse. Office of Fair

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 24 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 325

criminalization agenda in the UK, calling into question its viability.125 Wide-ranging reforms to the UK antitrust regime came into effect in 2014, including the creation of a new competition body and refinements to the definition of cartel activity, which may facilitate future prosecutions.126 3.5 Developments in Other Countries Other countries around the world have adopted new competition laws to combat cartels, reorganized their national competition agencies and devoted more resources to anti-cartel enforcement. These changes are certainly encouraged, if not catalyzed, by increasing transgovernmental exchange among national antitrust authorities within a particular normative framework which portrays global cartel conspirators as evil, contemptuous of the law, and exploitative of customers.127 Countries around the world have been attracted by US and EU enforcement models and successful prosecutions. Canada has become more active in prosecuting cartel participants, although the defendants have generally received a suspended sentence so the law has not had nearly the same bite as that of its North American neighbor. Nonetheless, from 1998 to 2008, Canada convicted 11 individuals of cartel offenses, nine were required to pay fines (between CDN$10,000 and CDN$250,000), and the other two received suspended prison sentences. The 2009 amendment to the Canadian Act creates a per Trading, ‘Project Condor Board Review’ (Office of Fair Trading 2010) 9–16 : setting forth the Office of Fair Trading’s explanation for the case’s collapse following an internal investigation. 125 See Joshua stating ‘The collapse on 10 May 2010 of the first contested jury trial under the UK’s lackluster criminal cartel regime was not only a humiliating public failure for the OFT, the investigating and prosecuting agency; it could also mark the beginning of the end of the whole criminalisation project.’: Julian Joshua, ‘DOA: Can the UK Cartel Offence Be Resuscitated?’, in BeatonWells and Ezrachi (eds) (n 3) 129. 126 Stephen Pollard and others, ‘UK Government Proposes Changes to Competition Laws, Including Criminal Cartel Offence’ (Wilmerhale, April 5, 2012) . 127 See Connor (n 11) 310. For examples of such portrayal by US antitrust authorities, see Thomas O Barnett, ‘Criminal Policy Speeches: Perspectives on Cartel Enforcement in the United States and Brazil’ (Department of Justice (US) April 28, 2008) ; Hammond (n 5).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

326 Comparative competition law

se criminal prohibition against price fixing with a maximum sentence of 14 years, which could facilitate further criminal prosecutions.128 Indeed, the last few years have witnessed a steep increase in conduct challenged by the country’s antitrust enforcer.129 After the amendments came into effect in March 2010, for example, companies implicated in a polyurethane foam cartel agreed to a fine of CDN$2.5 million, for only five months of illegal conduct, under the law’s new conspiracy provision.130 In Australia, survey evidence suggests a majority of that country’s public now views antitrust offenses in moral terms, following an extensive public-relations campaign by the Australian competition authority.131 The survey indicates that 42 percent of the public believe that cartel conduct should be a crime, and as of 2009 it now is.132 In addition to the fines imposed in connection with the vitamins cartel, Australian authorities fined six participants in the air transportation cartel in 2008 and initiated proceedings against participants in the marine hose cartel in June 2009.133 Australia has since aggressively pursued civil penalties – securing fines in the tens of millions of dollars, for example, in connection 128 Elisa Kearney and Mark Katz, ‘Anti-Cartel Enforcement In Canada – Still More Bark Than Bite’ (Davis 2009) 4–6 . 129 Ron Knox, ‘Canada’s Antitrust Bar, Global Competition Review’ (Global Competition Review November 6, 2012) . 130 Competition Bureau, ‘Backgrounder – Polyurethane Foam’ (Competition Bureau Website January 6, 2012) . Additional charges filed under the old price-fixing conspiracy provision resulted in a fine of CDN$10 million for 15 months of illegal conduct: ibid. See also ‘Canada Chocolate price-fixing: AP’ Washington Post (Washington June 6, 2013) . 131 See Beaton-Wells, noting that a greater percentage of respondents approved of punishment for moral reasons than for reasons of economic harm: Beaton-Wells (n 4) 17; See generally Melbourne Law School, ‘Cartel Project’ (Melbourne Law School Website 2014) ; Compare Andreas Stephen, ‘How Dishonesty Killed the Cartel Offence’ (2011) 6 Criminal Law Review 446; Andreas Stephen, ‘Survey of Public Attitudes to Price-Fixing and Cartel Enforcement in Britain’ (2008) 5 Competition Law Review 123: analyzing effects of British public’s failure to perceive cartels as morally blameworthy. 132 Beaton-Wells (n 4) 25. 133 Chavez (n 16) 942.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 26 /

Date: 13/10

JOBNAME: Duns PAGE: 27 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 327

with a cardboard box and air freight cartel. As of 2015, however, the government has not yet initiated a criminal prosecution for cartel conduct under the new law.134 Overall, Asian cartel enforcement has sharply increased, with approximately triple the number of investigations initiated in the 2005–2008 period compared to 1995–2004.135 In 2002 South Korea fined participants in the graphite electrode cartel – the Korea Fair Trade Commission’s first assessment of fines against an international cartel since its creation in 1981.136 Korea also imposed small fines on three of the 16 vitamins cartels in 2003.137 Korea’s leniency program is increasingly important; in 2010, the program played a role in 17 of 25 investigations leading to fines.138 In 2007 Japan imposed prison sentences (ranging between 18 months and three years) on five executives for bid rigging. The prosecution followed and was the first based upon a leniency application as part of Japan’s new leniency program.139 Despite this enforcement activity, Asia is still a weak link in international enforcement given the size of Asian economies and the profit potential for cartels.140 Since 2003 Latin American competition authorities have been increasingly active in anti-cartel activities, and ‘[b]y nearly all measures, Brazil has the largest and most effective anti-cartel authority in Latin America.’141 Brazil adopted a ‘National Anti-cartel Strategy’ (ENACC) in October 2009, as part of its second national Anti-Cartel Enforcement Day, in an effort to shape social perceptions of cartel activity and public

134 Sharon Henrick and Trish Henry, ‘Australia: Cartels’ [2012] The AsiaPacific Antitrust Review: noting fines of A$36 million and more than A$52 million in connection with cardboard box and air freight industry cartel, respectively. 135 Connor (n 70) 23. To be more precise, the term ‘investigations’ here encompasses all government and private legal actions, including formal investigations, fines, damage suits, and consent decrees: ibid 16. 136 Hwan Jeong, ‘Expansion in the Extraterritoriality of Korean Law’ (Asia Law September 2009) ; Chavez (n 16) 943. 137 Connor (n 65) 201; Jeong (n 136). 138 Kim Da-ye, ‘Two Faces of Secret-for-Pardon Policy’, Korea Times (Seoul July 29, 2012), : discussing differences between US and Korean leniency programs. 139 Spratling and Arp (n 34) 242, 252. 140 Connor (n 87) 593. 141 Connor (n 11) 315.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 27 /

Date: 8/9

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

328 Comparative competition law

awareness of government anti-cartel policies.142 The ENACC has since ‘evolved into a network of government and criminal enforcers, headed by a council created to coordinate administrative investigations and criminal prosecution.’143 Brazil’s criminal enforcement efforts have been robust. Brazil fines more hardcore cartels annually and imposes higher average corporate cartel fines than any other country in the region; it is also alone in Latin America in regularly fining cartel managers.144 Brazil fined participants of the vitamins cartel, and its Secretariat of Economic Law within the Justice Ministry has prosecuted participants in the air cargo and marine hose cartels.145 Moreover, Brazilian antitrust authorities have succeeded in reaching a number of settlement agreements – in connection with, for example, the marine hose, driving schools, and cement and beef cartels – pursuant to Brazil’s settlement program adopted in 2007.146 As of mid-2012 more than 250 executives faced criminal penalties for cartel behavior, and the government had sentenced 34 to jail time ranging from two to five years.147 Elsewhere in Latin America, Mexico – which has one of the largest and best-funded antitrust authorities in Latin America148 – fined participants in the citric acid cartel. Although Mexico’s competition authority has been preoccupied with mergers and monopoly cases, and has done little to attack cartels,149 this situation could change following passage of a new antitrust law in 2011, which includes criminal sanctions and more substantial investigatory tools. Focusing mostly on merger control and abuse of dominance, Argentina’s and Chile’s antitrust authorities are relatively underdeveloped. Argentina generally prosecutes only one or two hardcore cartels per year and imposes negligible fines.150 In the Middle East, Israel has been active in prosecuting cartel offenses, possibly reflecting closer normative ties to US antitrust policy. Israel secured jail sentences of between three and nine months for four 142 OECD (n 50) 19. The Enforcement Day program included senior enforcement officials from the US and EU: ibid. 143 Bruno Peixoto, ‘Brazil: Cartels and Leniency – The Antitrust Review of the Americas’ (Global Competition Review 2013) . 144 Connor (n 11) 316. 145 Chavez (n 16) 942–43. 146 OECD (n 50) 11, 17. 147 Peixoto (n 143). See also OECD (n 50) 18. 148 Connor (n 11) 317. 149 Ibid 318. 150 Ibid 316–17.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 28 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 329

executives and one economic advisor in its prosecution of a price-fixing and market division cartel involving the floor tile industry.151 Israel’s prosecution of cartel offenses in the liquefied petroleum gas market led to a plea agreement involving jail sentences and fines for three defendants, including a prison term for the CEO of one of Israel’s largest gas distribution companies.152 In more recent cases, Israeli authorities have even arrested executives implicated in cartel activity at the start of the investigation to prevent the executives from interfering.153 Egypt’s first criminal cartel prosecution since the creation of its Department to Protect Competition and Prohibit Monopoly came in early 2008 against 20 cement company executives. By summer’s end, an Egyptian court convicted the executives, fining them the equivalent of $1.9 million dollars each for price fixing and agreeing to divide the market.154

4. INTERNATIONAL COOPERATION EFFORTS Transnational initiatives have spurred the diffusion of legal norms regarding cartels and enforcement techniques to uncover and deter them. Throughout the post-World War II era, the US has played a central role in global business regulation.155 And so it has since the early 1990s in the global trend toward strengthening enforcement tools against cartel offenses, including the addition of criminal sanctions. The global trend coincided with the US DOJ’s aggressive stance toward international cartels dating from the early 1990s.156 The US efforts were supported by a general ideological shift in government attitudes toward market competition following the fall of the Berlin Wall in 1989. The US has worked through a number of fora to foster the diffusion of anti-cartel norms, national institution building to address cartels, and transgovernmental cooperation efforts. It has worked particularly through the OECD, the 151

Spratling and Arp (n 34) 252. Ibid. 153 Katy Oglethorpe, ‘Freshfields Panel Examines Global Cartel Enforcement Issues’ (Global Competition Review October 4, 2012) . 154 Spratling and Arp (n 34) 252–53. 155 See John Braithwaite and Peter Drahos, Global Business Regulation (Cambridge University Press 2000); Daniel W Drezner, All Politics is Global (Princeton University Press 2007) 5–6; Shaffer (n 2). 156 Harding (n 3) 194. 152

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 29 /

Date: 8/9

JOBNAME: Duns PAGE: 30 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

330 Comparative competition law

ICN, bilateral treaties, memoranda of understanding, and informal relations.157 The EU and EU member states have increasingly also played important roles. 4.1 OECD The OECD has addressed competition issues going back to its Competition Law and Policy Committee in 1961.158 Much of its efforts related to urging its members to cooperate in competition matters.159 Its efforts significantly intensified against cartels in 1998 when the OECD issued its ‘Council Recommendation Concerning Effective Action against Hard Core Cartels’, constituting the first multilateral statement defining and condemning hardcore cartels as pernicious. The OECD condemned ‘hard core cartels’ ‘as the most egregious violations of competition law.’160 The second report on the Recommendation’s implementation concluded in similar terms that ‘cartels are unambiguously bad,’ and the third report cited with approval a 2004 US Supreme Court opinion that cartels are the ‘supreme evil of antitrust.’161 Under its anti-cartel program, the OECD sponsors meetings of national authorities, publishes policy briefs and booklets designed to encourage particular enforcement tools such as the use of leniency programs (discussed in Section 5.2), collects information on sanctions, and compiles lists of best practices.162 In parallel, the OECD (complemented by a parallel program in UNCTAD) issues peer review reports on individual countries’ efforts to detect, investigate, and prosecute domestic and

157

Spencer Weber Waller, ‘The Internationalization of Antitrust Enforcement’ (1997) 77 Boston University Law Review 343. 158 Hugh Hollman and William E Kovacic, ‘The International Competition Network: Its Past, Current and Future Potential’ (2011) 20 Minnesota Journal of International Law 274. 159 See, for example, ‘Recommendation of the Council concerning Co-operation between Member Countries on Anticompetitive Practices affecting International Trade (9C(95)130/FINAL)’ (OECD Website July 27, 1995) . 160 OECD, ‘Hard Core Cartels: Third report on the implementation of the 1998 Council Recommendation 7’ (OECD 2005). 161 Ibid 7, 8. 162 See Chavez (n 16) 963–64.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 30 /

Date: 8/9

JOBNAME: Duns PAGE: 31 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 331

international cartels.163 It does so with the assistance of the countries’ national authorities who become key intermediaries for the conveyance of global anti-cartel norms and practices. The peer review process presses them to cast a critical eye on their own policies, and can provide leverage for them in domestic contests over policy reform. To expand its reach, the OECD established a broader Global Forum on Competition in 2001, and established regional competition centers in Eastern Europe and Asia while sponsoring a Latin American Competition Forum.164 The OECD’s 2005 report on the implementation of its 1998 Recommendation noted ‘aggressive enforcement efforts at very high levels, finding [that] competition authorities in more countries than ever bring important cases that resulted in significant sanctions.’165 The report concluded that ‘more countries are catching up and improving their enforcement regimes in line with developments in the most advanced jurisdictions,’ and that ‘[c]ooperation among authorities in investigations of cartels has reached unprecedented levels.’166 The report detailed illustrative examples of cartel investigations, discussed efforts to raise public awareness of cartel-related harm, reviewed international efforts to cooperate in investigations and enforcement actions, and detailed best practices for formal information sharing.167 It stressed, for example, that ‘[m]aking the public aware of the harm caused by cartels is an important part of a country’s overall effort to combat cartels,’ including through a ‘strong media relations programme,’ and it pointed to ‘[t]he programmes developed in Canada and the US’ as ‘good examples of what competition authorities can do to educate the public.’168 The report noted the particular importance of ‘sanctions against natural persons, placing them at risk individually for their conduct,’ and included a subsection on ‘A Trend Towards Criminalisation.’169

163 Hollman and Kovacic (n 158). UNCTAD also engages in peer review exercises, and complements some of the OECD’s work within its own organizational focus dedicated to serving developing countries. 164 Ibid. 165 OECD, ‘Hard Core Cartels’ (n 160) 8. 166 Ibid 8, 30. 167 Ibid 12–35. 168 Ibid 8, 16, 18–19. 169 Ibid, 28.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 31 /

Date: 8/9

JOBNAME: Duns PAGE: 32 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

332 Comparative competition law

4.2 ICN The ICN was created in 2001 under US instigation following an announcement of 14 competition agencies.170 It is a network of competition law officials and non-governmental advisors who have so far come predominantly from the private sector.171 As of May 2015, the ICN is comprised of 132 national, regional, and other territorial antitrust agencies operating in 119 jurisdictions, and the list continues to grow.172 It is now the central node for the diffusion and building of consensus around competition law norms and practices, facilitating the coordination of transnational regulatory efforts.173 In 2004 the ICN created a Cartel Working Group. The Working Group’s mandate ‘is to address the challenges of anti-cartel enforcement across the entire range of ICN members and amongst agencies with differing levels of experience. At the heart of antitrust enforcement is the battle against hardcore cartels directed at price fixing, bid rigging, market sharing and market allocations.’174 The Working Group contains sub-groups on legal frameworks and enforcement techniques and has produced (among other projects) an anti-cartel enforcement manual for its members’ use. Though the OECD did much of the early work on ‘hard core cartels,’ and continues to be important for policy analysis, the ICN has served to diffuse ideas and build relationships among OECD and non-OECD regulators and practitioners. The ICN’s aim is to foster ‘procedural and substantive convergence’ of competition law policy through sustained interaction, capacity building,

170

Eleanor Fox, ‘Linked-In: Antitrust and the Virtues of a Virtual Network’ (2009) 143 International Law 151, 160. For a recent retrospective of the network, see Paul Lugard (ed), The International Competition Network at Ten: Origins, Accomplishments and Aspirations (Intersentia 2011). 171 Hollman and Kovacic (n 158). 172 See ICN, The International Competitions Network to Focus on its Membership Needs, May 1, 2015, available at http://www.international competitionnetwork.org/uploads/library/doc1027.pdf (last checked, June 5, 2015). For an earlier figure from 2010, see John Fingleton, ‘The International Competition Network: Planning for the Second Decade’ (ICN 9th Annual Conference, Istanbul, Turkey, April 27, 2010) 2–3: giving a figure of 112 agencies from 99 jurisdictions in 2010. 173 Hollman and Kovacic (n 158). 174 ICN ‘Working Groups’ (International Competition Network Website 2014) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 32 /

Date: 13/10

JOBNAME: Duns PAGE: 33 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 333

and the sharing of practices.175 It facilitates deliberation among national competition authorities regarding preferred approaches to use against cartels, as well as to sort out differences.176 The ICN produces practical guidance, such as the Manual on Anti-Cartel Enforcement Techniques, and organizes workshops and teleseminars, including an annual ICN Cartel Workshop. The Manual on Anti-Cartel Enforcement Techniques, for example, contains chapters on searches, raids and inspections; drafting and implementation of effective leniency programs; digital evidence gathering; cartel case initiation; investigative strategy; and interviewing techniques. In 2005 the Cartel Working Group published the paper ‘Defining Hard Core Cartel Conduct, Effective Institutions, Effective Penalties’. The paper stresses ‘the significance of the personal liability of the decision-makers,’ and notes that ‘[s]ome agencies also emphasize the effectiveness of criminal sanctions as a deterrent.’177 The ICN arguably has been successful in contributing to change in legislation and administrative practice across jurisdictions regarding cartels. ICN member agencies cite the ICN’s Cartel Enforcement Manual as a crucial tool in developing national cartel enforcement strategies.178 US Deputy Assistant Attorney General Scott Hammond points to the Cartel Working Group’s annual workshops as particularly useful, providing a venue for anti-cartel enforcers to meet, learn from one another, and develop working relationships that form the basis for future cooperation.179 These informal connections, he writes, have led to ‘pick up the phone’ cooperation between competition regulators in different jurisdictions over time.180 The European Commission estimates that the ICN conducts 90 percent of its work by teleconference or email.181 Antitrust officials from national agencies around the world have increasingly shared resources and coordinated investigations, creating a sense of a shared 175

‘Memorandum on the Establishment and Operation of the International Competition Network’ (International Competition Network Website 2014) . 176 Fingleton (n 172) 5–6. 177 ICN, ‘Network Cartels Working Group, Defining Hard Core Cartel Conduct, Effective Institutions, Effective Penalties’ (International Competition Network 2005) 3–4 . 178 ICN (n 66). 179 Hammond (n 5) 14–15. 180 Ibid 15: discussing collaboration between US and UK regulators. 181 Chad Damro, ‘The New Trade Politics and EU Competition Policy: Shopping for Convergence and Co-operation’ (2006) 13 Journal of European Public Policy 867, 879.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 33 /

Date: 8/9

JOBNAME: Duns PAGE: 34 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

334 Comparative competition law

professional enterprise. Coordinated activity has expanded greatly, including over the timing of raids and the execution of warrants.182 The ICN complements bilateral cooperation among competition agencies and governments through antitrust cooperation agreements and Mutual Legal Assistance Treaties (MLATs), which can be used against cartels. The US currently has either a cooperation agreement or MLAT with Australia, Brazil, Canada, Chile, China, Colombia, the EU, Germany, India, Italy, Israel, Japan, Mexico, Russia, and the UK.183 National agencies, for example, cooperated successfully to secure plea agreements in connection with the airlines cartel, and coordinated investigative efforts in connection with the marine hose cartel.184 These agreements reflect the growing role of international coordination in cartel investigations. Transnational legal norms are more likely to have impacts where they are clear, coherent, and deemed legitimate.185 The ICN and OECD provide fora in which networks of national competition law officials participate. These international initiatives have expanded their access for officials from jurisdictions around the world. Although the OECD and ICN can be viewed as rivals, the norms that they have conveyed are coherent, and the documents that codify them are precise and elaborate. The officials who participate in the ICN and OECD initiatives act as intermediaries between the national context and the global one.186 Where they develop professional identities as competition law enforcers, and where they meaningfully participate in these transnational processes, they more likely view the documents that emerge from them to represent a legitimate expression of a transnational policy consensus. In such situations, they are more likely to actively press for concomitant domestic legal change and engage in enforcement efforts. It is through this web of global, regional and bilateral networks that anti-cartel enforcement ideas diffuse. The OECD and ICN are just the most important among transgovernmental and transnational networks of legal norm diffusion. The broader network of international competition law and international competition 182

Hammond (n 5) 15. See also Spratling and Arp (n 34) 255–61. Department of Justice (US) ‘Antitrust Cooperation Agreements’ (Department of Justice (US) Website 2014) www.justice.gov/atr/public/international/intarrangements.html. 184 Hammond (n 5) 15–16. 185 Shaffer (n 2). 186 On intermediaries, see Bruce Carruthers and Terence C Halliday, ‘Negotiating Globalization: Global Scripts and Intermediation in the Construction of Asian Insolvency Regimes’ (2006) 31 Law & Social Inquiry 521, 537; Shaffer (n 2). 183

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 34 /

Date: 8/9

JOBNAME: Duns PAGE: 35 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 335

law enforcers includes the UNCTAD, which plays an important role for developing nations, as well as a wide variety of non-governmental and academic institutions, programs, and events. Non-governmental advisors representing business and consumer groups, academia, and the legal and economic professions are significantly involved in these processes, so that the processes are not simply ‘transgovernmental,’ but more broadly transnational. Diffused common norms through both public and private actors facilitate transgovernmental coordination, and the practical experience of successful coordination in a common enterprise, in turn, spurs further normative convergence across and within nation states.

5. TRENDS IN ENFORCEMENT TECHNIQUES Multilateral and bilateral cooperative efforts, such as through the ICN, OECD, and bilateral agreements, have circulated new enforcement tools to combat cartels. We address two, the facilitation of extradition through the criminalization of cartel activities; and the use of leniency programs to induce the breakup of cartels. 5.1 Extradition and the Case of Ian Norris Extradition requires dual criminality.187 This means that the offense at issue must be a crime in both jurisdictions. One consequence of the criminalization of cartel offenses is that more countries may extradite individuals for prosecution in other jurisdictions. The jurisdiction most likely seeking such extradition and most likely to imprison individuals is the US. If extradition to the US becomes easier, decisions for participants in cartels about international business travel become more complex. In March 2010 Ian Norris became the first foreign national extradited to the US to face charges stemming from an antitrust investigation.188 Norris obstructed the DOJ’s investigation into a carbon components 187 Dual criminality, a requirement of most US extradition treaties, requires that the offense at issue must be a crime in both jurisdictions. See generally Girardet (n 41). 188 A good overview from which this summary is built can be found at Akin Gump Update, ‘The Antitrust Division Succeeds in Its First-Ever Extradition of a Foreign National’ (Akin Gump March 26, 2010) ; see also, for example, Julian M Joshua, Peter D Camesasca and Youngjin Jung, ‘Extradition and Mutual Legal Assistance Treaties: Cartel Enforcement’s Global Reach’ (2008) 75 Antitrust Law Journal 353, 354–60.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 35 /

Date: 8/9

JOBNAME: Duns PAGE: 36 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

336 Comparative competition law

manufacturing cartel in which he participated. In November 2002 UK industrial manufacturer Morgan Crucible and its US subsidiary Morganite admitted to wrongdoing in connection with a price-fixing conspiracy. Norris headed Morgan Crucible during a part of the conspiracy and throughout the cover-up. The US indicted him on both counts, and alleged that his offenses were flagrant. He was found to have assembled a document-destruction task force whose sole purpose was to destroy evidence of the price-fixing conspiracy. That task force created a script, designed to mislead investigators, for use by Morgan-entity executives questioned about the conspiracy. After the DOJ investigation became public, Norris instructed his co-conspirators to use the script if questioned about the illicit meetings. The DOJ’s attempt to extradite Norris to the US initially failed. The House of Lords held in 2008 that price fixing was not a crime at the time in question (as it is now in the UK), and thus the dual criminality requirement of the US-UK extradition treaty was not satisfied. The House of Lords nonetheless remanded the case on the question of whether extradition for obstruction of justice was proper. The lower courts rejected Norris’s argument that extradition would violate the European Convention on Human Rights in view of his (and his wife’s) age and poor health – a result unanimously affirmed on appeal by the UK Supreme Court (which in 2009 assumed the role of court of final resort from the House of Lords). The European Court of Human Rights rejected his appeal on human rights grounds in February of 2010. On March 23, 2010, the UK extradited Norris to the US to stand trial in the Eastern District of Pennsylvania.189 A federal jury convicted him of conspiring to obstruct justice on July 27, 2010,190 and in December 2010 a judge sentenced him to 18 months in a US prison.191 The Norris extradition is unlikely to be a one-off event. In 2014, a former marine hose executive was extradited from Germany to face charges of participating in a worldwide bid-rigging conspiracy making this the first extradition for a direct antitrust crime, rather than the obstruction justice charges in the Norris case. Extraditions will likely 189

Ibid. ‘DOJ News Release on Ian Norris Conviction’ (Main justice July 27, 2010) . 191 ‘Former Morgan Crucible Chief Ian Norris Sentenced to 18 Months in US Jail’, The Telegraph (London, December 10, 2010 . 190

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 36 /

Date: 8/9

JOBNAME: Duns PAGE: 37 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 337

become more commonplace as a growing number of countries criminalize cartel activity. As a result, avoiding travel to the US may no longer be sufficient to insulate cartel members from the prospect of spending time in a US prison. Even short of extradition, growing criminalization means substantial travel restrictions and the use of border controls and Interpol to monitor the location of defendants and fugitives.192 5.2 Leniency Programs Antitrust officials point to the proliferation, convergence and coordination of leniency provisions to destabilize cartels as another major development in anti-cartel enforcement around the world. Leniency programs generally provide immunity to the first cartelist to admit liability and cooperate with authorities, with the aim of creating a ‘race to confess.’193 They are used as a carrot to complement the stick of enhanced sanctions, such as criminalization or substantial fines. In some ways, they can be viewed as ‘a motor of penal expansion,’ as leniency becomes more attractive when the alternative of sanctions becomes more threatening.194 The program was first developed in the US and has been zealously used there. Gary Spratling, former US Deputy Assistant Attorney General for cartel prosecution, reports that the use of leniency is linked to 90 percent of US cartel fines imposed since 1997.195 The US has aggressively sought to promote leniency programs, and the ICN has publicized procedures to aid national competition agencies in implementing them. In 1990 only the US had a leniency program on the books. Today around sixty jurisdictions do,196 including Australia, Brazil, Canada, Czech Republic, the EU, France, Germany, Ireland, Japan, South Korea, New Zealand, Sweden, and the UK.197 The DOJ’s Hammond thus describes this phenomenon as ‘the single most significant development in cartel enforcement.’198 The diffusion of leniency programs as an anti-cartel

192

Hammond (n 5) 14. ICN Cartel Working Group, ‘Anti-Cartel Enforcement Manual’ (ICN 2009) 2. 194 Harding and Joshua (n 7) 297. 195 Spratling and Arp (n 34) 287. 196 Telephone Interview with US DOJ official (10 June 2011); see also Hammond (n 5) 2–3. 197 Hammond (n 5) 2–3; see also Chavez (n 16) 968–78: listing jurisdictions and discussing the programs extensively. 198 Hammond (n 5) 2–3. 193

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 37 /

Date: 8/9

JOBNAME: Duns PAGE: 38 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

338 Comparative competition law

enforcement tool again attests to the normative power of the US model diffused through the ICN. Nonetheless, some might question whether the parallel trend toward global criminalization should enhance or hamper the effectiveness of leniency programs.199 While Hammond rightly stresses that the threat of harsh sanctions in a jurisdiction should induce cartel members to seek immunity, the cartel participant must also be wary of potential sanctions in other jurisdictions where the cartel’s activities have transnational effects. Where the potential sanction in another jurisdiction is penal, the participant’s decision may become more complicated, especially in light of institutional divergences for enforcing criminal sanctions and administering amnesty and leniency programs in different jurisdictions so that decisions may lie outside the control of a national antitrust authority.200

6. IMPEDIMENTS TO IMPLEMENTATION: SOCIAL NORMS AND INSTITUTIONAL STRUCTURES Notwithstanding the expansion of criminal provisions and the increasingly robust enforcement of cartel prohibitions around the world, actual enforcement will face severe practical limits in most jurisdictions. The implementation of transnational legal norms will ultimately depend on local factors, and in particular, domestic political and social attitudes regarding cartel behavior, and domestic institutional structures, capacities, and legacies.201 Cultural attitudes may be the single biggest hurdle to enforcement trends. Not all members of the public are convinced that cartel offenses merit the criminal penalty of jail time, which is advocated most vocally by the US. Criminalization implicates moral judgments that vary with socio-cultural context. US antitrust law has long exhibited a moral dimension, which facilitates the use of criminal sanctions against individuals in cartel cases.202 In contrast, ‘there appears to be (at least outside North America) no strong feeling on the part of the wider public about 199

See Jason D Medinger, ‘Antitrust Leniency Programs: A Call for Increased Harmonization as Proliferating Programs Undermine Deterrence’ (2003) 52 Emory Law Journal 1439. 200 Jonathan T Schmidt, ‘Keeping US Courts Open to Foreign Antitrust Plaintiffs: A Hybrid Approach to the Effective Deterrence of International Cartels’ (2006) 31 Yale Journal of International Law 211, 232–40, 245–50. 201 Shaffer (n 2). 202 Baker (n 122) 155, 158.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 38 /

Date: 8/9

JOBNAME: Duns PAGE: 39 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Criminalizing cartels 339

the inherent criminality of price fixing and like practices.’203 National competition authorities outside of North America recognize this uphill battle and thus view public education about the evils of cartel offenses as a central component of their missions.204 Presentations in October 2010 by competition authorities from Australia and Japan at the ICN’s Cartel Workshop in Yokohama, Japan reflect this concern.205 The 2010 ICN survey extensively reports ‘factors influencing perception of importance of cartel enforcement.’206 To the extent that public opinion supporting criminalization is lacking in many jurisdictions, the transnational trend can be understood as more of a top-down than a bottom-up process. As a result, the impact at the time of implementation could be limited. Another significant hurdle is institutional, involving particular institutional heritages, structures, and capacity challenges. Jurisdictions in Europe and Asia that have an institutional heritage of using administrative agencies that apply administrative fines against enterprises, as opposed to criminal sanctions against individuals, are unlikely to change significantly, at least in the short term. In addition, for most jurisdictions, criminal law enforcement involves a separate institution from cartel enforcement, creating institutional coordination challenges. A competition agency may wish to retain its monopoly on enforcement, and public prosecutors may not trust competition authorities. In contrast, the US entrusts both civil and criminal enforcement against cartels to a specialized division within the DOJ, which has created a particular institutional legacy that facilitates the use of criminal sanctions. Also, many competition systems are in their infancy and lack institutional capacity. Competition law is particularly recent in China and Egypt, for example, and it has recently undergone a major overhaul in states such as Brazil, India, and Mexico. Finally, there are jurisdiction-specific disincentives, such as constitutional and evidentiary hurdles, which complicate the pursuit of aggressive enforcement.207 In sum, both institutional legacies and political and social attitudes toward cartels will affect the 203 Harding (n 3) 197. See also Gerber noting the lack of cultural roots of competition policy in Japan: Gerber (n 7) 213, 217. 204 See Beaton-Wells noting the Australian competition authority’s extensive campaign in this respect: Beaton-Wells (n 4). 205 See ibid; Hideo Nakajima, Director General Investigation Bureau, Japan Fair Trade Commission, ‘Outreach Activities by the JFTC: Focusing on Cartel Awareness’ (5 October 2010) 9: discussing the ‘necessity of building up public support for cartel enforcement’. 206 See ICN (n 66) 53–65. 207 Harding (n 3) 193–94.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 39 /

Date: 13/10

JOBNAME: Duns PAGE: 40 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

340 Comparative competition law

application of anti-cartel law in practice, especially as regards criminalization. In many cases, we are skeptical that enforcement practice will meaningfully follow the spread of formal legal policies.

7. CONCLUSION Countries in every region of the world, including virtually all of the world’s leading economies, have significantly enhanced sanctions and, in a growing number of cases, criminalized cartel offenses, often only recently. Many states have initiated prosecutions, several have secured convictions, and a few have imposed jail time for these offenses. The US DOJ has played a central role as a unilateral enforcer against international cartels, as a collaborator with other national competition agencies in enforcement, and as an anti-cartel advocate in international fora. International venues such as the ICN now play an important role in offering guidance to national competition agencies, and in providing a forum for policy deliberation, information sharing, cooperation, and professional socialization. Harry First, former Chief of New York state’s Antitrust Bureau, goes so far as to declare that we already ‘have international “law” [against cartels] without ever having adopted one at the international level.’208 Since criminal law lies at the heart of state sovereignty, the global trend toward criminalization of cartel offenses is quite remarkable. Yet the criminalization and enforcement records outside the US are hardly uniform. Given the novelty of legal changes in so many countries, and the challenges of institutional capacity for many, we are skeptical regarding actual enforcement in many countries that have formally adopted enhanced sanctions. Even in states that have criminalized cartel offenses, lingering questions remain about the propriety of criminalization and imprisonment. Much of the criminalization trend thus appears to be a product of transnational enforcement interests more than of domestic bottom-up processes. While countries appear to be moving toward convergence on enhanced sanctions, including criminal penalties against individuals, national competition agencies outside of North America are – to borrow from Harold Koh – grappling with the task of bringing home transnational legal norms and practices for combating cartels.209

208 209

Columns Design XML Ltd

/

First (n 102) 727. Koh (n 9).

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

/Pg. Position: 40 /

Date: 8/9

Columns Design XML Ltd

/ 1890 (became a felony in 1974)

1930v

United States

Italy

Job: Duns-Comparative_competition_law

/

Norway

Austria

France

1932 v 1947 (amended in 2005 to give investigatory powers) 1953 (became primarily civil in 1986) 1959 v (originally criminalized all hardcore cartels, in 2002 decriminalized except for bid rigging) 1960

1889

Canada

Poland Japan

Datei of criminalization of cartel activities

Country

Yes (became more active in the 1980s)

Administrative fines only

No

Amount of criminal fines not specified

Administrative fines only Max. 5 million yen Max. €75,000

Corporation max. US$100 million or twice the gain from the illegal conduct or twice the loss to the victims; individual max. US$1 million Min. € 103,00 Max. € 1032,00

Max. CDN$25 million

Criminal finesiii

Yes (became more active in early 1990s) No Yes (became more active recently) No

Yes (became more active in late 1990s) Yes (became more active in 1990s)

Serious about using criminal laws?ii

Table 12.1 Criminalization of cartel conduct by country, listed chronologically

Max. 6 years

Max. 3 years

Max. 4 years

Min. 3 months Max. 5 years Max. 3 years Max. 5 years

Max. 10 years

Max. 14 years

1926 (new laws in 1993 and 2004)

1945 (new laws in 1986 and 2001) 1951 (new laws in 1984 and 2005)

1990 (new law in 2007) 1947

1990

1890

1889 (new law in 1986)

Imprisonmentiv Date of Competition Law

JOBNAME: Duns PAGE: 41 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

341

Division: 15_Chapter12

/Pg. Position: 1 /

Date: 8/9

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 15_Chapter12

1973 (revision in 1995 separating general cartels and bid rigging) 1977 (initially provided for imprisonment but later abolished and reinstated in 2009) 1980 1988 1990 1991 (amended in 1999 to only criminalize repeat offenders)

Spain

Greece

South Korea

Kenya Brazil

Taiwan

1993

1961

Israel

Iceland

Datei of criminalization of cartel activities

Country

Table 12.1 (continued)

Yes (became more active in late 2000s)

No Yes (became more active post-2003) No

Yes (became more active in 2008)

Yes

No

Yes (became more active in late 1990s)

Serious about using criminal laws?ii

Amount of criminal fines not specified

Max. KES 10 million Min. 1/30 minimum wage Max. 1,800 times minimum wage Max. TWD 100 million (but only if repeat offender)

Max. KRW 200 million

Min. fine of 1 year (percentage of income determined by judge) Max. fine of 2 years Min. €100,000 Max. €1,000,000

Corporation max. NIS 4 million plus NIS 28,000 for each day offense persists; individual max. NIS 2 million plus NIS 14,000 for each day such offense persists

Criminal finesiii

Max. 6 years

Max. 5 years Min. 2 years Max. 5 years Max. 3 years (but only if repeat offender)

Max. 3 years

Min. 2 years

Min. 1 year Max. 3 years

Max. 5 years

1993 (new law in 2005)

1988 (new law in 2010) 1962 (new laws in 1990, 1991, and 1994) 1991

1980

1977 (new law in 2011)

1963 (new laws in 1989 and 2007)

1959 (new law in 1988)

Imprisonmentiv Date of Competition Law

JOBNAME: Duns PAGE: 42 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

342

/Pg. Position: 2 /

Date: 8/9

Yes (became more active in 2010) No Yes No

1995 (major revision was made to criminal law in 2008 but no prosecution to date)

1996 1996

1996 1997vi 1997 v 1999 1999 2002

2002

Slovenia

Columns Design XML Ltd

Ireland

/

Romania

Russia

Denmark

Germany Indonesia

Thailand Barbados

Job: Duns-Comparative_competition_law

/

Estonia

Yes

No No

No

Yes

No

No

1994

Zambia

No

1993

Slovakia

Corporation: Max. €16 million or 10% of revenue Individual: Min. 30 days of salary Max. 500 days of salary

Max. RUR 1 million or 5 years of salary Amount of criminal fines not specified None Min. Rp. 25 billion Max. Rp. 100 billion Max. THB 6 million Corporation max. BBD 500,000; individual max. BBD 150,000 (only imposed if fail to end cartel after warning)

Min. 500 Max. 30,000

Max. 500,000 penalty units (1 penalty unit is currently K180) Corporation: Min. €50,000 Max. 200 times damage caused or illegal gain obtained Individual: Min. 30 days of salary Max. 500 days of salary Max. €4 million or 10% of revenue

Administrative fines only

Max. 5 years Max. 6 months Max. 3 years Max. 6 months (only if fail to end after warning) Max. 3 years

None

Min. 6 months Max. 3 years Max. 7 years

Max. 10 years

Min. 6 months Max. 5 years

Max. 5 years

Max. 6 years

1993 (new laws in 1998 and 2001)

1979 (new law in 1999) 2002

1937 (new laws in 1990 and 1997) 1923 (new law in 1958) 1999

1991 (new law in 2006)

1953 (new laws in 1991 and 2002) 1996

1993 (new law in 2008)

1991 (new laws in 1994 and 2001) 1994 (new law in 2010)

JOBNAME: Duns PAGE: 43 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

343

Division: 15_Chapter12

/Pg. Position: 3 /

Date: 8/9

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law 2011

Mexico

/

Yes

Yes Min. 1,000 days of salary Max. 3,000 days of salary

Administrative only Corporation max. AUS$ 10 million or 3 times the gain from the illegal conduct or, if that gain cannot be defined, 10% of revenue; individual max. AUS$220,000 Amount of criminal fines depends on damages Max. R 500,000

Unlimited fines

Criminal finesiii

Min. 3 years Max. 10 years

Min. 2 years Max. 8 years Max. 10 years

Max. 5 years Max. 10 years

Max. 5 years

1955 (new laws in 1979 and 1998) 1924 (new laws in 1992 and 2011)

1991 (new law in 2001)

1948 (new laws in 1998 and 2002) 1990 (new law in 1996) 1906 (new laws in 1965, 1971, 1974, and 2010)

Imprisonmentiv Date of Competition Law

ii

All years are based upon the date the law was enacted. A country is deemed to meet this standard if it has prosecuted individuals under their criminal law, or, in the event that it has only recently criminalized cartel conduct, it has stressed that it is its intention to do so. This category involves, in part, a subjective judgment. iii All criminal fines are based upon penalties in the current law. iv All terms of imprisonment are based upon penalties in the current law. v These countries have limited criminalization of cartel activity to bid rigging in public tenders. This table provides an entry point for understanding the state of criminalization in countries. While we note those countries that criminalize only bid rigging, the table does not break down countries in terms of whether the criminal offense focuses specifically on cartels or more broadly on competition law violations. The table also does not note countries that have completely decriminalized cartel violations, such as Finland. The table also does not note those jurisdictions where fines are formally administrative (and not criminal), but where their stringency can arguably be viewed as punitive and thus penal in practice, such as the EU, Finland, Germany and the Netherlands. vi The country only imposes criminal fines and do not imprison violators. vii South Africa passed legislation to criminalize cartel activity in 2009 but it has not yet gone into force because of Presidential inaction.

i

2009vii

2009

Czech Republic

South Africa

No Yes

2005 v 2009

Hungary Australia

Yes

Yes

2002

United Kingdom

Serious about using criminal laws?ii

Datei of criminalization of cartel activities

Country

Table 12.1 (continued)

JOBNAME: Duns PAGE: 44 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

344

Division: 15_Chapter12

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

13. International governance of competition and the problem of extraterritorial jurisdiction Brendan Sweeney

1. INTRODUCTION One of the challenges facing competition regulation is the ever-increasing internationalization of commerce. Radical improvements in transport and communications have ensured that commercial activities occurring in one state can have dramatic effects in other states. Unlike commerce, however, competition law remains essentially domestic. This disconnect between the realities of modern commerce and the source of its regulation has forced states to confront the boundaries of their legitimate regulatory reach and to investigate new regulatory models. In an interconnected world it is unrealistic to expect states to adopt a strict territorial approach to protecting their legitimate interests. A state has a legitimate right to intervene in activities that are directed at its economic well-being even where those activities occur outside its territorial boundaries. This is so because the state has a legitimate right to protect its citizens against economic exploitation. At the same time, the state in which the activities occur clearly has a legitimate right to regulate activities occurring within its borders. This right is rooted in the notion of state sovereignty, which traditionally accords the state authority over activities occurring in its own territory. Thus we have a situation of concurrent jurisdiction. These occasions of concurrent jurisdiction are constantly being created. This is because of regime diversity (both in competition law and, more generally, legal systems1) and because competition rulings are based primarily on domestic considerations (that is, local welfare or other considerations, not the welfare or otherwise of foreigners), these occasions are also sometimes contested. Determining how to govern these occasions is one of the compelling issues confronting states. It is 1 See for example Hannah Buxbaum, ‘Transnational Regulatory Litigation’ (2006) 46 Virginia Journal of International Law 251.

345

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

346 Comparative competition law

suggested that jurisdiction (the extraterritorial reach of domestic competition laws) is best understood within the context of an emerging scheme of international governance. At this point a word on terminology is needed. Jurisdiction is a word with many meanings. This chapter is concerned with just one of those meanings, subject-matter jurisdiction (also called prescriptive jurisdiction or legislative jurisdiction). Subject-matter jurisdiction in this chapter means the extraterritorial reach of domestic competition law.2 The chapter is not concerned with issues of personal jurisdiction, evidencegathering or remedial enforcement.3 An examination of these jurisdictional matters involves considerations of a more general systemic nature than merely competition law and is beyond the scope of this chapter. The chapter is organized as follows. First, it presents an overview of the current situation with regard to governance of transnational competition. The aim is to show jurisdiction in its broader context. Secondly, the chapter examines the extraterritorial application of US antitrust law and the reactions it has inspired. The reasons for concentrating in this way on the US are quite simple: the US was the first state to have competition laws; it was the first state to apply those laws extraterritorially; it is still the state that practises the broadest extraterritoriality; more than any other state it encourages private actions (where government restraint is more difficult to apply). Thirdly, the chapter looks at the situation in the European Union (EU). Fourthly, it considers the approach to extraterritorial reach in a selection of other states including Japan, Korea, China, Canada, Brazil and Australia.

2 The use of the word ‘jurisdiction’ is further complicated by the fact that the extraterritorial reach of domestic law is not regarded in some states as being properly characterized as a jurisdictional issue. For example, US courts have recently held that the extraterritorial reach of the Sherman Act is not a jurisdictional matter but a question going to the merits of a case: see Animal Science Products Inc. v China Minmetals Corp. 654 F.3d 462 (3rd Cir 2011), cert denied 132 S Ct 1744 (2012); Minn-Chem, Inc., v Agrium Inc. 683 F.3d 845 (7th Cir 2012); cert denied 134 S.Ct. 23 (2013). The distinction can have important practical implications but is not important for this discussion. 3 These issues are discussed in Brendan Sweeney, The Internationalisation of Competition Rules (Routledge 2010). See also Brendan Sweeney, ‘Combating Foreign Anti-Competitive Conduct: What Role for Extraterritorialism?’ (2007) 8 Melbourne Journal of International Law 35–87.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 347

2. THE ROLE OF DOMESTIC JURISDICTION IN THE INTERNATIONAL GOVERNANCE OF COMPETITION International governance possibilities represent a continuum. At one end lies a multilateral competition agreement with binding rules and some form of supranational enforcement mechanism (for example, a World Trade Organization (WTO)-type solution). This type of solution is most unlikely (at least in the foreseeable future). A much weaker version was floated at the WTO by the EU, but after several years of discussion failed to achieve any traction.4 Today few commentators favour any formal multilateral competition agreement. Those who see a role for the WTO are generally careful to set tight limits for its involvement. At the other end of the spectrum lies a world based on the unilateral application of domestic competition laws extraterritorially. When the only real competition regime was the US version, various US actors, from the courts to administrative authorities, experimented with unilateralism. However, a number of headline cases from the 1970s and 1980s demonstrated both the potential for conflict inherent in unilateralism and the weapons that other states may deploy to counter that unilateralism.5 Between the two ends of the continuum lie an infinite number of possibilities. Solutions become easier as states adopt corresponding policies. In this respect there has been an incredible growth in the number of countries adopting some form of competition regime.6 However, one should not expect that harmonization is a natural progression nor that it can prevent all conflict. States may apply similar regulations to the same conduct and come up with quite different solutions. For some states, competition law and policy is a new and often radical departure from previous practice; such states are on a steep learning curve. Many states do not have the capacity to fully enforce their competition laws, it

4 An attempt to craft a minimum plurilateral agreement at the WTO failed. In 2003 the Ministerial Conference at Cancun in Mexico removed competition policy from the Doha agenda. See ‘Doha Work Programme’ (WTO Doc WT/L/579, WTO 2004) [1(g)] (Decision Adopted by the General Council on 1 August 2004). 5 See for example the Westinghouse Uranium case discussed at text accompanying n 32. 6 In 1990 about 20 states had a competition law. Now over 120 states boast such a regime.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

348 Comparative competition law

may take years for understanding and capacity to develop.7 Nor should we expect competition policies to trend towards increasing uniformity. States have different socio-economic structures with different policy needs.8 States are at different levels of development and have different views on how to pursue economic development. The science of economics is not so settled as to permit of definitive answers. Indeed, the notion of competition itself has no universally accepted meaning. Consequently, we should not be surprised that competition policies differ. Finally, states differ in their institutional choices – the respective role of administrators, courts and private parties varies from state to state. For example, the US relies much more heavily on the private sector to enforce its competition law than do other states. The EU puts significant trust in a very powerful regional regulator. These variances are not simply regulatory choices, but often reflect deep-seated systemic differences (which in turn are often culturally driven).9 Over the past two decades a patchwork solution to the problem of international competition governance has emerged. This solution includes a mix of treaties, informal agreements (state to state and agency to agency) and networks (political, administrative, professional and academic). These networks involve a range of domestic governmental and non-governmental actors. They operate in a variety of ways through a variety of institutions such as the Organisation for Economic Co-operation and Development (OECD), United Nations Conference on Trade and Development (UNCTAD), Asia-Pacific Economic Cooperation (APEC) and the World Bank. The most important of these networks is probably the International Competition Network (ICN).10 7

The ICN, UNCTAD and the OECD all operate capacity-building programmes. 8 For example, Gal has persuasively argued that small economies will rationally adopt different competition policies to large economies. See Michal S Gal, Competition Policy for Small Market Economies (Harvard University Press 2003). 9 For a brief overview of differences see Albert A Foer and Jonathon W Cuneo, ‘Toward an Effective System of Private Enforcement’ in Albert A Foer and Jonathon W Cuneo (eds), The International Handbook of Private Enforcement of Competition Law (Edward Elgar 2010) 592–3. 10 The ICN is dedicated solely to competition policy issues. Its work is done by a series of working groups; and a conference is held annually. Input is sought not only from members, but also from other interested state and non-state actors, as well as international institutions such as the OECD and UNCTAD. Decisions, however, are made solely by members – that is, by competition authorities. The ICN has so far produced an impressive list of guidelines and best practices on

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 4 /

Date: 13/10

JOBNAME: Duns PAGE: 5 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 349

Global governance through a mix of governmental and quasigovernmental networks fits neatly into theories of global governance espoused by some liberal scholars. Anne-Marie Slaughter, for example, views network governance as the future of international governance (the ‘new sovereignty’).11 This new sovereignty, according to Buxbaum, is ‘a reconceptualization of sovereignty that de-emphasizes the notion of absolute control over particular territory, but nevertheless maintains the importance of statehood and of an international community of states.’12 Developing international rules is de-emphasized in favour of a soft approach. The emphasis is on domestic law and how domestic law is to be adapted and applied to meet the needs of international commerce. Given the central role of domestic law, the issue of the scope of domestic jurisdiction becomes important. The issue may be posed thus: what sort of approach to jurisdiction is most likely to achieve the balance between domestic needs and international harmony? As previously indicated, a strict territorial approach to jurisdiction is inappropriate. Commercial interdependence and the emergence of global markets demand a more sophisticated notion of jurisdiction than one that is characterized entirely by geographical boundaries. If a simple geographically based

issues such as pre-merger notification, merger regulation, conducting cartel investigations and assessing dominance or market power in unilateral conduct cases. These guidelines and best practices are nonbinding: member states are free to use them as they wish. Perhaps the greatest success of the ICN is more difficult to measure – the mutual understanding and trust forged between members of a network confronted regularly with similar problems. The ICN website is at www.internationalcompetitionnet work.org. See generally Oliver Budzinski, The Governance of Global Competition (Edward Elgar 2008) 142; Oliver Budzinski, ‘The International Competition Network: Prospects and Limits on the Road towards International Competition Governance’ (2004) 8 Competition and Change 223; D Daniel Sokol, ‘Monopolists without Borders: The Institutional Challenge of International Antitrust in a Global Gilded Age’ (2007) 4 Berkeley Business Law Journal 37. 11 However, not everyone agrees that the new sovereignty is either legitimate or effective: Anne-Marie Slaughter, A New World Order (Princeton University Press 2004); see, for example, Kenneth Anderson, ‘Squaring the Circle? Reconciling Sovereignty and Global Governance through Global Government Networks’ (2005) 118 Harvard Law Review 1255. See also the critique of network governance by Pierre-Hugues Verdier, ‘Transnational Regulatory Networks and Their Limits’ (2009) 34 Yale Journal of International Law 113. Realists are naturally sceptical. 12 Buxbaum (n 1) 307.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 5 /

Date: 13/10

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

350 Comparative competition law

jurisdiction is rejected, some form of extraterritorial jurisdiction is inevitable.13 Determining the limits of that jurisdiction becomes the task.14 When balancing domestic preferences against the possibilities of international conflict, it should not be assumed that conflict within the international regulatory space is inevitable. In fact, considering the scale of international commerce and the fact that over 100 states now have some form of competition rules, the occasions of ‘open’ conflict are quite few. Even though states may have different regulatory objectives, there are reasons why one state may allow another to regulate the international space which they both inhabit. First, it may be the ‘cheap’ or practical option. Domestic enforcement has costs and limitations (remedies against foreign entities are not always enforceable); although the spill-over benefits from another state’s enforcement may not produce an optimal outcome, they may be an acceptable second-best outcome. Secondly, the state may defer to another state because it recognizes the other state has a greater interest in the matter. This is not just a unilateral exercise of comity; in return for deferring on this occasion, the state may expect a reciprocal response on some other occasion when its interests are paramount. Ultimately comity relies on a commitment to reciprocity. Thirdly, deference may be the realpolitik option. Small and developing states are probably not in a position to demand jurisdictional priority vis-à-vis more powerful states. Thus the exercise of jurisdiction is a complex matter. Ultimately domestic jurisdiction should be understood as part of a broader governance approach that utilizes a variety of tools: domestic jurisdictional rules, positive as well as traditional comity,15 mutual recognition agreements, cooperation agreements (bilateral and regional) and dialogue. A 13

Extraterritorial is here used simply to indicate jurisdiction that catches conduct engaged in outside the territorial boundaries of the state. 14 See International Bar Association (IBA) discussing two views on extraterritorial jurisdiction as a matter of international law: IBA, ‘Report of the Task Force on Extraterritorial Jurisdiction’ (IBA 2008) 8–9. 15 Traditional or negative comity is a principle of deference – one country restrains itself so as not to allow its laws and law enforcement actions to harm or impede another country’s important interests. See discussion at text accompanying n 38. The object of positive comity is to allocate investigation and prosecution of conduct to the country in the best position to carry out those functions. Positive comity was recommended in the OECD’s non-binding 1973 ‘Recommendation Concerning a Consultation and Conciliation Procedure on Restrictive Business Practices Affecting International Trade’ (1973) C(73)99(Final) and was first adopted in a competition context in the agreement

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 351

jurisdictional rule alone is not sufficient to ensure a proper balance between the requirements of promoting domestic welfare and avoiding international conflict. Nevertheless it seems reasonable to suggest that some basic principles ought to inform domestic extraterritorial jurisdiction. The starting point is an examination of the US experience with extraterritoriality.

3. THE EXTRATERRITORIAL REACH OF US ANTITRUST LAW The extraterritorial reach of US antitrust law is complex. Complexity is difficult to avoid where the interests to be balanced are both important and antithetical. Thus the extraterritorial reach of US antitrust regulation recognizes critical distinctions, between US markets and foreign markets (even where the economic market is global) and between public and private enforcement. 3.1 Development of the Effects Doctrine In American Banana the US Supreme Court rejected the notion that US antitrust law applied to conduct occurring wholly outside the territorial boundaries of the US.16 According to Justice Holmes ‘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 was done’.17 If Justice Holmes meant that the effects of any impugned conduct are to be ignored in determining jurisdictional competence, then it is difficult to support such a strict interpretation. A state must have some right to protect itself against foreign-based actions that have a serious, adverse impact on the state.18 between the US and the European Commission: see Cooperation Agreement (EC) 27.4.95 Agreement between the Government of the United States of America and the Commission of the European Communities regarding the Application of their Competition Laws [1995] OJ L 95/47 (signed and entered into force 23 September 1991); corrected at [1995] OJ L 131/38. 16 American Banana Co. v United Fruit Co., 213 US 347 (1909). 17 Ibid 356. 18 International law recognizes extraterritorial criminal jurisdiction where the effects of criminal conduct are experienced within the state. This is called the objective territorial principle to distinguish it from the strict territorial principle and was applied by the Permanent Court of International Justice in SS ‘Lotus’ (France v Turkey) (1927) PCIJ Series A, No 9 (judgment of 7 September 1927).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

352 Comparative competition law

The US Court of Appeals applied a broader notion of jurisdiction in Alcoa.19 The US Department of Justice alleged that a Canadian producer had conspired with a number of European producers to control the production of aluminium, including exports to the US, in violation of § 1 of the Sherman Act.20 The alleged unlawful agreements were made outside the US and would clearly have been in breach of the Act if made within the US. The Court held that Congress intended the Sherman Act to catch conduct engaged in by a foreign national outside the borders of the US where that conduct had intended and actual effects on commerce in the US.21 The effects doctrine of extraterritorial antitrust reach has been applied in US cases since 1945 and remains the cornerstone of US policy to discipline foreign anti-competitive conduct that affects US consumers. Its application, however, has not always been consistent.22 The element of intent is sometimes largely ignored.23 The severity of the domestic effects required has attracted a variety of standards.24 Nevertheless, the generally accepted test is that applied by the Supreme Court in Hartford, that the relevant effects must be intended and substantial.25 19

US v Aluminium Co. of America, 148 F.2d 416, 433 (2d Cir 1945). 15 USC § 1. § 1 states that every contract, combination or conspiracy in restraint of trade or commerce among the several states or with foreign nations is illegal. 21 In previous US cases at least some part of the impugned conduct had occurred in the US: see, for example, US v American Tobacco Co. 221 US 106 (1911); US v Sisal Sales Corp. 274 US 268 (1927). For a thorough analysis of the evolution of the effects doctrine, see Zenith Radio Corp. v Matsushita Electric Industrial Co. 494 F Supp. 1161 (D C, Pa, 1980). 22 See Antitrust Modernization Commission (AMC), ‘Report and Recommendations’ (AMC Website) 226 (hereafter AMC Report). 23 Intent has either been totally ignored or, more commonly, given a de minimis standard: See respectively Sabre Shipping Co v American President Lines Ltd (1968) 285 F.Supp 949. For example, holding that ‘a low threshold of proof is all that is required for a jurisdictional determination’: Zenith Radio Corp v Matsushita Electric Indust. Co. (1980) 494 F Supp 1161; Fleischmann Distilling Corp v Distillers Co. (1975) 395 F.Supp 221. See, also Austen Parrish, ‘The Effects Test: Extraterritoriality’s Fifth Business’ 61 Vanderbilt Law Review 1455. 24 See AMC Report (n 22) 226. 25 See Hartford: ‘the Sherman Act applies to foreign conduct that was meant to produce and did in fact produce some substantial effect in the US’: Hartford Fire Insurance v California, 509 US 764, 796 (1993); Minn-Chem, Inc., v Agrium Inc. 683 F.3d 845, 855 (7th Cir 2012). 20

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 6 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 353

The Hartford test only applies to foreign conduct that affects US domestic or import commerce.26 Where the foreign conduct does not involve US domestic or import commerce, jurisdiction is governed by the Foreign Trade Antitrust Improvements Act (FTAIA)27 The FTAIA requires that effects be ‘direct, substantial and reasonably foreseeable’.28 3.2 International Responses to the Effects Doctrine The Alcoa effects doctrine drew some robust international responses.29 Some of the US’s major allies and trading partners expressed serious concerns. Whilst conflict was driven ultimately by different attitudes to economic policy and international law,30 tensions were clearly exacerbated by particular aspects of the US legal system, notably private antitrust actions, the right to treble damages and US pre-trial discovery

26

For a discussion on the meaning of ‘US import commerce’ see MinnChem, Inc., v Agrium Inc. 683 F 3d 845 (7th Cir 2012). 27 Foreign Trade Antitrust Improvements Act 15 U.S.C. § 6a (2000 & Supp IV) discussed below at text accompanying n 66. 28 For a discussion on the FTAIA standard see Florian Wagner-von Papp, ‘Competition Law, Extraterritoriality and Bilateral Agreements’ in Ariel Ezrachi (ed), Research Handbook on International Competition Law (Edward Elgar 2012). The author, for instance, argues that the requirement of ‘directness’ is misleading and that the true operative part of the test is the need for the effects to be ‘reasonably foreseeable’. 29 See, for example, Harold G Maier, ‘Interest Balancing and Extraterritorial Jurisdiction’ (1983) 31 American Journal of Comparative Law 579, 579: ‘The extraterritorial application of US law has been an irritant in US relationships with its allies and other trading partners since World War II.’ 30 See, for example, the British amicus brief in the Uranium case. The brief stated: [I]t has long been the position of the British Government that, in the context of the application of the US antitrust laws to conduct outside the US by non-US citizens, the “effects” test is inconsistent with international law’: Amicus Curiae Brief of the Government of the United Kingdom of August 1979 in Uranium Antitrust Litigation 617 F.2d 1248 (7th Cir 1980). See also in US v Nippon Paper Industries Co Ltd 109 F 3d 1 (1997) (Amicus Curiae Brief: Government of Japan). Japan also objected to US extraterritorialism in Zenith Radio Corporation v Matsushita Electric Industrial Co., 513 F Supp 1100 (E D Pa 1981) and in the lower court decision in F. Hoffmann-La Roche Ltd v Empagran S.A., 542 US 155 (2004) (Amicus Curiae Brief: Government of Japan). Objections to the US effects doctrine were also forcibly expressed by Australia, Canada and South Africa during the Uranium case. See Stephen Ramsey, ‘The US–Australian Antitrust Cooperation Agreement: A Step in the Right Direction’ (1983) 24 Virginia Journal of International Law 127, 133.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 9 /

Date: 14/10

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

354 Comparative competition law

procedures.31 So, for example, when the US sought to apply its anticartel rules extraterritorially in Uranium Antitrust Litigation,32 it directly confronted the economic policies of other states – Great Britain, Canada, France, Australia and South Africa – which actively encouraged certain cartels, or at least in the circumstances saw no reason to prohibit them.33 Not surprisingly, these states objected vigorously.34 Aside from diplomatic protests and filing objections with US courts (including the amicus curiae process), foreign states reacted legislatively by introducing laws that were designed to impede US extraterritoriality – evidence blocking laws,35 laws that prohibit local firms from complying with certain foreign awards36 and claw-back laws (which enable firms to claw back any damages paid in excess of compensatory damages).37 Although these legislative defences have not been widely used, they 31 See Rio Tinto Zinc Corporation v Westinghouse Electric Corporation [1978] 1 All ER 434. 32 Re Uranium Antitrust Litigation; Westinghouse Electric Corporation v Rio Algom Ltd, 617 F 2d 1248 (7th Cir 1980). 33 For a discussion of Uranium Antitrust Litigation see Ramsey (n 30) 133. Cartels were the norm in many European countries and in Japan prior to World War II. Although competition laws were introduced after the war, they were not accorded the same central significance to economic policy that the antitrust laws had in the US. Consequently, they were often ignored (Japan) or subject to numerous exceptions (United Kingdom and many European states). Other states took longer to develop workable competition rules. 34 The already heightened tensions were exacerbated by the comments of the US Court of Appeals in Re Uranium Antitrust Litigation; Westinghouse Electric Corporation v Rio Algom Ltd, 617 F 2d 1248, 1255-6 (7th Cir 1980). Judge William Campbell said: ‘In the present case, the [defendants objecting to US subject matter jurisdiction] have contumaciously refused to come into court and present evidence as to why the District Court should not exercise its jurisdiction. They have chosen instead to present their entire case through surrogates. Wholly owned subsidiaries of several defaulters have challenged the appropriateness of the injunctions, and shockingly to us, the governments of the defaulters have subserviently presented for them their case against the exercise of jurisdiction.’ 35 See, for example, Protection of Trading Interests Act 1980 (UK) c 11; Foreign Proceedings (Prohibition of Certain Evidence) Act 1976 (Cth) repealed and replaced by Foreign Proceedings (Excess of Jurisdiction) Act 1984 (Cth). 36 See, for example, Protection of Trading Interests Act 1980 (UK) c 11; Foreign Extraterritorial Measures Act, RSC 1985, c F-29, s 8 (Canada); Foreign Proceedings (Excess of Jurisdiction) Act 1984 (Cth). 37 See, for example, Foreign Proceedings (Excess of Jurisdiction) Act 1984 (Cth) s 10; Protection of Trading Interests Act 1980 (UK) c 11, s 6. See generally Deborah Senz and Hilary Charlesworth, ‘Building Blocks: Australia’s Response

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 355

served the purpose of highlighting the limitations surrounding a unilateral solution to an international problem. 3.3 The Role of Comity as Part of a Jurisdictional Test Comity is a principle of international relations: in relations between sovereign states there will be occasions when one state should as a matter of respect defer to the interests of another state.38 Its application often requires a balancing test in which the interests of the home state are weighed against the interests of the foreign state. In response to foreign state reactions to the effects doctrine the US antitrust agencies adopted the Antitrust Enforcement Guidelines for International Operations.39 The Guidelines provide that: ‘[I]n determining whether to assert jurisdiction to investigate or bring an action, or to seek particular remedies in a given case, each Agency takes into account whether significant interests of any foreign sovereign would be affected.’40 The Guidelines provide a list of criteria to be used in performing the comity analysis. The decision to proceed or to defer to the interests of a foreign state is made by the state.

to Foreign Extraterritorial Legislation’ (2001) 2 Melbourne Journal of International Law 69. 38 The origins of comity lie in the emergence of the notion of state sovereignty. Comity implies something less than an international obligation, something discretionary but nevertheless imbued with custom and expectations of reciprocity. The content of comity, however, remains elusive. Maier described comity as ‘an amorphous never-never land whose borders are marked by fuzzy lines of politics, courtesy, and good faith’: Harold G Maier, ‘Extraterritorial Jurisdiction at a Crossroads: The Intersection between Public and Private International Law’ (1982) 76 American Journal of International Law 280, 281. Nevertheless, despite its apparent indeterminacy, it is a useful doctrine (at least diplomatically, if not judicially) that encompasses all the reasons why one state or one legal system might defer to another state or legal system. For a discussion on the origins and uses of comity see Joel R Paul, ‘Comity in International Law’ (1991) 32 Harvard International Law Journal 1. See also B Pearce, ‘The Comity Doctrine as a Barrier to Judicial Jurisdiction: A US-EU Comparison’ (1994) Stanford Journal of International Law 525; Note, ‘Predictability and Comity: Toward Common Principles of Extraterritorial Jurisdiction’ (1985) 98 Harvard Law Review 1310. 39 Department of Justice, Antitrust Division, Antitrust Enforcement Guidelines for International Operations (2nd edn, 1995) [3.2] (hereafter ‘US International Guidelines’). 40 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

356 Comparative competition law

Judicially the initial response was equally accommodating. In Timberlane the Appellate Court held that Congress intended US antitrust law to apply to foreign conduct only if US interests were sufficiently compelling.41 This was to be determined by balancing the foreign interests involved against US interests.42 The Timberlane judicial balancing test was much criticized (including by the US antitrust authorities) for imposing on the courts an unworkable and costly standard and for usurping the role of the political branch of government.43 According to the Antitrust Enforcement Guidelines, where a case involves foreign conduct the decision whether to take government action is a political matter. The courts should accept the comity analysis of the government.44 Commentators have been critical of the whole notion of the courts performing a case-by-case balancing test.45 In 41 Timberlane Lumber Co. v Bank of America, 549 F 2d 597, 613 (9th Cir 1976) (hereafter ‘Timberlane’). 42 The court said: ‘The elements to be weighed include the degree of conflict with foreign law or policy, the nationality or allegiance of the parties and the locations or principal places of businesses or corporations, the extent to which enforcement by either state can be expected to achieve compliance, the relative significance of effects on the US as compared with those elsewhere, the extent to which there is explicit purpose to harm or affect American commerce, the foreseeability of such effect, and the relative importance to the violations charged of conduct within the US as compared with conduct abroad’: Timberlane (n 41) 613. This approach was supported by the minority judgment in Hartford Fire Insurance v California 509 US 764 (1993) (Justice Scalia dissenting). It also draws some support from the Restatement (Third) of the Foreign Relations Law of the US. A balancing-of-interests test was applied by the Third Circuit in Mannington Mills v Congoleum Corporation 595 F 2d 1287, 1298 (3rd Cir 1979), but was rejected by the Court of Appeals for the District of Columbia in Laker Airways v Sabena Belgian World Airlines 731 F 2d 909 (DC Cir 1984). In Timberlane the Court of Appeal referred to the international balancing-ofinterests test as a jurisdictional rule-of-reason test: Timberlane (n 41). 43 See for example Maier (n 29); Russell J Weintraub, ‘The Extraterritorial Application of Antitrust and Securities Laws: An Inquiry into the Utility of a “Choice-of-Law” Approach’ (1992) 70 Texas Law Review 1799; William S Dodge, ‘Extraterritoriality and Conflict-of-Laws Theory: An Argument for Judicial Unilateralism’ (1998) 39 Harvard International Law Journal 101. 44 ‘The Department does not believe that it is the role of the courts to “second-guess” the executive branch’s judgment as to the proper role of comity concerns under these circumstances’: US International Guidelines (n 39) s 3.2. 45 See for example Hovenkamp commenting at 6 that ‘the “jurisdictional rule of reason” that Timberlane adopted is cumbersome, often indiscriminate, conducive to lengthy and expensive discovery, and thus burdensome to both litigants and courts.’: Herbert Hovenkamp, ‘Extraterritorial Criminal Jurisdiction

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 12 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 357

Empagran, Justice Breyer, delivering the decision of the majority in the Supreme Court, also cast doubts on the advisability of case-by-case jurisdictional analysis.46 More generally, interest balancing has been criticized as leading in practice to a local law bias. In Hartford a majority in the US Supreme Court affirmed the effects test but rejected the Timberlane ‘balancing of interests’ approach.47 According to the majority, comity, if indeed it was relevant at all, could only preclude an exercise of jurisdiction where there was in fact a true conflict between US law and foreign law. A true conflict exists only where foreign law requires the defendant to act in the impugned manner. Most competition regimes do not require firms to act in any particular manner.48 Therefore, where subject-matter jurisdiction is based on the Hartford interpretation of the effects doctrine, US courts will generally have little discretion to defer to foreign interests.49 This appears to leave very little scope for any judicial comity analysis.

under the Antitrust Laws’ (2013) University of Iowa Legal Studies Research Paper No 13-14 Feb, . 46 F. Hoffmann-La Roche Ltd v Empagran S.A., 542 US 155, 168 (2004) per Breyer J arguing that case-by-case comity analysis is ‘too complex to prove workable’. See discussion in Hovenkamp commenting that the Timberlane court may have chosen to adopt a balancing test because (1) it did not apply a substantial effects test and (2) on just about any view US jurisdiction was not really justified: Hovenkamp (n 45). 47 Hartford Fire Insurance v California, 509 US 764 (1993). The decision in Hartford remains controversial. Both Souter J (for the majority) and Scalia J (in dissent) applied the Restatement (Third) of Foreign Relations Law of the US. It has been claimed that both applied the Restatement incorrectly, thus, leaving this area of the law unclear. See Russell J Weintraub, ‘Globalization’s Effect on Antitrust Law’ (1999) 34 New England Law Review 27, 29. See also A F Lowenfeld, ‘Conflict, Balancing of Interests, and the Exercise of Jurisdiction to Prescribe: Reflections on the Insurance Antitrust Case’ (1995) 89 American Journal of International Law 42, and the reply by L Kramer, ‘Extraterritorial Application of American Law after the Insurance Antitrust Case: A Reply to Professors Lowenfeld and Trimble’ (1995) 89 American Journal of International Law 750. 48 See, for example, In re Vitamin C Antitrust Litigation, 810 F Supp 2d 522 (EDNY Sep 06, 2011) holding that Chinese law did not compel the price-fixing conduct alleged. For China’s reaction see ‘MOFCOM Criticises US Court Decision on Vitamin C Price Fixing As Unfair and Inappropriate’ (2013) 25 China Competition Bulletin 8. 49 509 US 764, 542–4.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 13 /

Date: 13/10

JOBNAME: Duns PAGE: 14 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

358 Comparative competition law

In Hartford itself the British government appearing as amicus to the Supreme Court argued that US antitrust jurisdiction did not extend to conduct occurring in London because Great Britain had a regulatory scheme in place for the reinsurance industry and the impugned reinsurance agreements were exempt from Great Britain’s competition laws under that scheme. The agreements, though permitted under British law, were not compelled by British law. Therefore there was no true conflict between US antitrust law and British law. The majority applied the effects doctrine without any recourse to a balancing test. It is difficult to fault the actual outcome in Hartford. Although the reinsurance agreements were made in London, they were clearly designed to affect the US insurance market. The effects test was applied to a criminal prosecution in Nippon Paper Industries (1997).50 Nippon involved an alleged conspiracy by Japanese defendants to fix the price of thermal fax paper sold in North America. All meetings between the defendants were held in Japan. The defendants sold the thermal paper to third-party trading houses in Japan, which then imported and sold the paper in the US. Therefore, all of the defendants’ alleged unlawful conduct took place in Japan. On appeal, the Court held that even in a criminal prosecution ‘[S]ection One of the Sherman Act applies to wholly foreign conduct which has an intended and substantial effect in the US.’51 In Nippon the Court rejected an argument based on comity analysis. Nippon has been employed (at least implicitly) in many cases since. If Hartford is applied strictly, it is difficult to see that comity has any part to play in a case-by-case judicial jurisdictional test that goes beyond the act of state doctrine or the foreign state compulsion doctrine.52 If 50 US v Nippon Paper Industries Co Ltd, 109 F 3d 1 (1st Cir 1997); cert denied, 522 US 1044 (1998). For a discussion of the case and of the Japanese reaction see Jennifer Farlow, ‘Ego or Equity? Examining US Extension of the Sherman Act’ (1998) 11 Transnational Lawyer 175, 195–201. 51 US v Nippon Paper Industries Co Ltd, 109 F 3d 1, 9. See also Re TFT-LCD (flat panel) Antitrust Litigation 2011-1 Trade Cases P 77, 419. For an analysis of the ‘effects’ required in a criminal matter, see Hovenkamp (n 45). 52 As a matter of recognizing sovereignty most states exempt foreign sovereigns: see for example Foreign Sovereign Immunities Act of 1976, 28 U S C 1605(a); State Immunity Act 1978 (UK) c 33, s 3; State Immunity Act, [R.S. 1985, c. S-18] s 5 (Canada); Foreign States Immunities Act 1985 (Cth) s 11 (Australia). However, the trend is for these exemptions not to apply where the foreign state is acting commercially. In Republic of Argentina v Weltover, Inc., 504 US 607, (1992) the Court commented, ‘when a foreign sovereign acts, not as a regulator of a market, but in the manner of a private player within it, the

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 14 /

Date: 13/10

JOBNAME: Duns PAGE: 15 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 359

foreign conduct has an intended and substantial effect on US commerce then US antitrust laws will apply to that conduct unless the conduct was compelled (not just tolerated or even encouraged) by the foreign state.53 While this approach protects US welfare, it leaves the way open for conflict where another state considers its interests are equally significant. A more internationally acceptable approach would be one that somehow avoids the pitfalls of a Timberlane balancing test and the bias of the Hartford test. Rather than either conducting a cumbersome interestbalancing analysis or merely determining whether the foreign sovereign compelled the conduct, there is need for a workable test that answers the question whether the impugned conduct has a sufficient nexus with the US to displace the presumption against extraterritorial application.54 It is difficult, however, to envisage a concrete form for such a test.55 A test that relies on proof that the US effects were substantial and intended (in the sense of being reasonably foreseeable) may be the best fit that currently exists. Because US antitrust authorities have made it clear that they will continue to apply the Antitrust Enforcement Guidelines in determining what actions to bring, the practical significance of Hartford is mainly in the field of private actions, particularly hard-core cartel actions where the substantive rule to be applied is a per se one. 3.4 US Private Actions and Jurisdiction The US antitrust regime relies heavily on private enforcement.56 Plaintiffs are given powerful incentives such as treble damages, contingency fee foreign sovereign’s actions are “commercial” within the meaning of the Foreign Sovereign Immunities Act’. 53 See, for example, Trugman-Nash, Inc. v New Zealand Dairy Bd., Milk Products Holdings (North America) Inc., 954 F Supp 733 (1997). 54 The US Supreme Court has recently applied a similar approach to other areas of regulation. See Morrison v National Australia Bank, 130 S Ct 2869 (2010) (securities regulation under the Securities Exchange Act) and Kiobel v Royal Dutch Petroleum Co. 133 S Ct 1659 (2013) (foreign torts under the Alien Torts Act). 55 The issue is discussed by Hovenkamp (n 45). Professor Fox has suggested that a more appropriate effects doctrine could be based on the Restatement with some amendments. See Eleanor M Fox, ‘Modernization of Effects Jurisdiction: From Hands-Off to Hands-Linked’ (2009) 42 New York University Journal of International Law and Policy 159. 56 It has been estimated that private actions account for up to 90% of US antitrust cases. The existence of private actions is a political economy choice to

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 15 /

Date: 13/10

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

360 Comparative competition law

arrangements, costs rules that favour plaintiffs, joint and several liability (without any guarantee of contribution), liberal pre-trial discovery rules and class actions.57 Yet the main function of antitrust law is regulatory. Thus, while private actions clearly have an important compensatory role this is subordinate to the regulatory role. The problem with US private actions is that they lie beyond the control of the regulatory authorities.58 This restricts the use of political comity in extraterritorial cases. The issue therefore is how to contain private actions so that they assist international governance efforts and not impede them. Problems associated with the Hartford test have already been discussed. Not all foreign anti-competitive conduct, however, is subject to Hartford. Congress, in part recognizing the need to constrain US antitrust law as a matter of international comity, enacted the FTAIA.59 The FTAIA removes foreign anti-competitive conduct (other than US import trade and commerce) from the Sherman Act unless the conduct satisfies two conditions, first that the conduct had a ‘direct, substantial and reasonably foreseeable’ effect on US domestic or import commerce or on certain export commerce, and second that such effect gave rise to a Sherman Act claim.60

partially privatize enforcement. As such the US position is unlikely to change. The situation is slightly different in, for example, Australia where the executive retains some control over foreign damages suits. An action for damages based on extraterritorial jurisdiction of the competition law requires the consent of the relevant Minister. See discussion at text accompanying n 168. 57 See Douglas Ginsburg, ‘Comparing Antitrust Enforcement in the US and Europe’ (2005) 1(3) Journal of Competition Law and Economics 427, 436. See also Hannah Buxbaum, ‘The Private Attorney General in a Global Age: Public Interests in Private International Antitrust Litigation’ (2001) 26 Yale Journal of International Law 219. 58 This is not necessarily the case in other states where the needs of public enforcement are given priority. 59 Foreign Trade Antitrust Improvements Act 15 U.S.C. § 6a (2000 & Supp IV). 60 Ibid, § 6a(1). The distinction between ‘import commerce’ that falls outside the FTAIA (and is governed by Hartford) and import commerce that is governed by the FTAIA is discussed in Minn-Chem, Inc., v Agrium Inc. 683 F.3d 845 (7th Cir 2012). If the relationship between a foreign cartel participant and its US subsidiary is such that the conduct of the foreign parent is to be imputed to the US entity, then the matter involves domestic commerce not import commerce. See Carrier Corp., v Outokumpu OYJ, 673 F 3d 430 (6th Cir 2012).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 361

In Empagran the US Supreme Court dealt with the second condition.61 The Court had to determine whether the FTAIA entitled a foreign plaintiff injured outside the US by foreign cartel activities to bring an action for damages in US courts. The cartel’s activities in so far as they injured domestic US commerce were within the criminal and civil jurisdiction of the US courts.62 However, foreign losses which arose independently of the US harm were something different. The Supreme Court held that US antitrust law did not reach so far. The Court applied comity principles as a matter of statutory construction to restrict the extraterritorial scope of the US antitrust law.63 Justice Breyer, delivering judgment in Empagran, recognized that the US required a substantial reason to interfere with the internal ordering of foreign states. He said:64 No one denies that America’s antitrust laws, when applied to foreign conduct, can interfere with a foreign nation’s ability independently to regulate its own commercial affairs. But our courts have long held that application of our antitrust laws to foreign anti-competitive conduct is nonetheless reasonable, and hence consistent with principles of prescriptive comity, insofar as they reflect a legislative effort to redress domestic antitrust injury that foreign anti-competitive conduct has caused.

But why is it reasonable to apply those laws to foreign conduct insofar as that conduct causes independent foreign harm and that foreign harm alone gives rise to the plaintiff’s claim? Why should American law supplant, for example, Canada’s or Great Britain’s or Japan’s own 61 F. Hoffmann-La Roche Ltd v Empagran S.A., 542 US 155 (2004). The case arose as part of the global vitamins cartel. 62 ‘[A] purchaser in the US could bring a Sherman Act claim under the FTAIA based on domestic injury’ 542 US 155, 159. 63 The reaction of foreign states to the expansion of US private antitrust enforcement, as demonstrated in the amici briefs filed in Empagran, suggests that US expansion had gone too far. Although the states that filed briefs in Empagran emphasized the danger to their leniency programmes, underlying this functional objection was a more deep-seated objection to US interference in their internal affairs. See briefs filed by the UK and the Netherlands, Germany and Belgium, and Japan. 64 542 US 155, 164–5 (2004). A similar trend has emerged in other areas of regulation; see Morrison v National Australia Bank Ltd, 130 S Ct 2869 (2010) (securities regulation) and Kiobel v Royal Dutch Petroleum Co. 133 S Ct 1659 (2013) (Alien Torts Act). In Kiobel Roberts CJ, delivering the opinion of the majority, said: ‘And where the claims touch and concern the territory of the US, they must do so with sufficient force to displace the presumption against extraterritorial application.’

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 17 /

Date: 13/10

JOBNAME: Duns PAGE: 18 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

362 Comparative competition law

determination about how best to protect Canadian or British or Japanese customers from anti-competitive conduct engaged in in significant part by Canadian or British or Japanese or other foreign companies? Theoretical arguments based on an optimal level of deterrence were not sufficient to overcome these sovereignty concerns.65 For the purposes of its decision the Supreme Court assumed that the foreign harm was independent of US antitrust harm. The case was remitted to enable a hearing on this issue. The plaintiffs argued that their harm was not independent of US harm, that the foreign harm could only exist because of the cartel’s US effects. If the cartel did not produce the relevant effects in the US, arbitrage possibilities would have made the cartel impossible to maintain in other locations. Therefore, according to the foreign plaintiffs, US harm and the foreign harm were linked. In Empagran II the Court rejected this argument, and the Supreme Court refused certiorari.66 According to Empagran II, the FTAIA required the US effects to give rise directly to the foreign harm. An argument based on but-for reasoning was insufficient. The US effects must be the proximate cause of the foreign harm. Other courts have adopted the same interpretation.67 The Empagran decisions have significantly curtailed the extraterritorial scope of the antitrust laws.68 Viewing the matter from an international regulatory perspective, the Empagran I and II approach is to be welcomed. This approach was strongly supported by other states.69 65

The main argument put forward in favour of a broad approach to the extraterritorial application of the FTAIA was based on the need for optimum deterrence and the fact that this could only be achieved by exposing cartel participants to the danger of worldwide damages claims through US courts. In turn this argument is based on the claim that there is under-deterrence of cartel activities due to under-enforcement particularly in developing states. See amici briefs Empagran, S.A. v F. Hoffmann-La Roche Ltd., 2004 WL 533931-5; 2004 WL 542780; 2004 WL 531749. 66 Empagran, S.A. v F. Hoffmann-La Roche Ltd., 417 F.3d 1267 (DC Cir 2005), cert. denied 546 US 1092 (2006). 67 See, e.g., In re Monosodium Glutamate Antitrust Litigation, 477 F.3d 535 (8th Cir 2007); In re Dynamic Random Access Memory (DRAM) Antitrust Litigation, 546 F.3d 981 (9th Cir 2008); Boyd v AWB Ltd, 544 F Supp 2d 236 (S.D.N.Y., 2008). 68 It has been argued that the issue is still unclear. For this reason the Antitrust Modernization Commission recommended a legislative change to the following effect: ‘As a general principle purchases made outside the US from a seller outside the US should not be deemed to give rise to the requisite effects under the FTAIA.’ AMC Report (n 22) 228. 69 See amici briefs in F. Hoffmann-La Roche Ltd v Empagran S.A., 542 US 155.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 18 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 363

The meaning of ‘direct, substantial and reasonably foreseeable’ (the first condition in the FTAIA test) and the Hartford test (intended and substantial effect) were discussed in Minn-Chem, Inc. v Agrium Inc.70 Particular attention was paid to the meaning of ‘direct’. The Court denied that ‘direct’ meant ‘an immediate consequence’ and held that it served as a remoteness requirement. ‘[F]oreign activities … [must not be] too remote from the ultimate effects on US domestic or import commerce’.71 In other words, it is sufficient that there be a ‘reasonably proximate causal nexus’ between the conduct and the US effect.72 The facts of Minn-Chem demonstrate the respective uses of the Hartford test and the FTAIA. The potash cartel (all foreign producers) made up over 70 per cent of the worldwide market. The US was the second largest consumer of potash behind China. The cartel entered into sophisticated arrangements to raise the worldwide price of potash by up to 600 per cent. Some of the US plaintiffs bought potash directly from a cartel member. As these transactions were US import trade or commerce they came under the Hartford test.73 Those transactions that fell outside this class were subject to the FTAIA. The Court described one such situation in the following terms: ‘For example, Canpotex [a defendant] is the unified marketing and sales agent for Agrium, Mosaic and PCS [three defendants] in all markets except Canada and the US, yet its actions are an important part of the alleged scheme to set inflated benchmark prices.’74 Therefore, the Court had to be satisfied that Canpotex came within the exception to the FTAIA before determining whether Canpotex had breached US antitrust law. 3.5 Conclusion The US approach to the extraterritorial reach of the antitrust laws is complex and nuanced. A number of distinctions must be drawn, for instance, between government and private actions and between local and 70

Minn-Chem, Inc. v Agrium Inc., 683 F 3d 845 (7th Cir 2012); cert. denied 134 S.Ct. 23 (2013). 71 Ibid, 857. But see Motorola Mobility LLC v AU Optronics No. 14–8003 (7th Cir) where Judge Posner held that price fixing a component part sold in the main outside the US but later resold as part of a finished product to the US failed to satisfy the FTAIA’s requirement of a direct and foreseeable effect on US commerce. The case is currently on appeal. 72 See also Lotes Co., Ltd v Hon Hai Precision Industry, No. 13-2280 (2nd Cir 2014). 73 Ibid, 855. 74 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

364 Comparative competition law

foreign plaintiffs. Where action is being contemplated by a US antitrust agency the state is well placed to take into account the concerns of other states by having regard to comity considerations.75 Most obviously it can do this by refusing to proceed (even though it otherwise has jurisdiction). Government restraint may be exercised as a result of a negative comity agreement (of which a number exist76) or as a matter of unilateral government policy (for example, see the US Antitrust Enforcement Guidelines for International Operations discussed above). Private actions are less amenable to government influence. While the US Supreme Court has applied comity principles to restrict the interpretation of the FTAIA and thus limit the possibilities for foreign plaintiffs, the role of comity either as part of that test or as part of the Hartford test for import commerce is still unclear. If comity does have some part to play the criticisms expressed earlier must eventually be faced.

4. THE EXTRATERRITORIAL REACH OF EU COMPETITION LAW Extraterritorial application of Articles 101 and 102 of the TFEU77 depend on two doctrines, the implementation test and the single economic entity doctrine. The European Commission supports an effects test but this has not yet been judicially approved.78 4.1 Implementation Test In 1985 the European Commission (‘EC’) fined a number of foreign wood pulp producers (from Finland, Canada and the US) for engaging in cartel activities that were in breach of the competition laws of the 75 The Federal Trade Commission is a statutory authority and as such is not part of the US executive. Nevertheless, it is a signatory to the US International Guidelines (n 39). 76 For a list of the relevant US bilateral agreements, see Department of Justice, Antitrust Division (Department of Justice (US) 2013) at 15 October 2013. 77 Consolidated Version of the Treaty Concerning the Functioning of the European Union [2012] OJ C 326/88-9. 78 For a discussion on the EU see Damien Geradin, Marc Reysen and David Henry, ‘Extraterritoriality, Comity, and Cooperation in EU Competition Law’ in Andrew Guzman (ed), Cooperation, Comity and Competition Policy (Oxford University Press 2011) 21–9.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 20 /

Date: 13/10

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 365

European Economic Community (‘EEC’ now the EU).79 The US defendants were operating under a Webb-Pomerene exemption, the effect of which was to exempt them from US antitrust laws.80 Kraft Export Association (KEA), an American export association to which some of the defendant corporations belonged, was also held liable. KEA had not itself engaged in any trade with the EEC. Although the agreements were made outside the EEC, the EC nevertheless determined that the defendants had agreed to fix prices and other trading terms – for example, resale and re-export bans – in breach of Article 85 of the Rome Treaty.81 The Wood Pulp defendants appealed. The European Court of Justice (ECJ) rejected the defendants’ argument that the EEC did not have jurisdiction. According to the ECJ an illegal price-fixing agreement under Article 85 is made up of two acts, the agreement itself and the implementation of that agreement. If the implementation of the agreement occurred within the EEC, it is irrelevant that the agreement was made elsewhere. The act of selling products pursuant to the unlawful agreements would amount to implementing the price agreement. The ECJ held: ‘The producers in this case implemented their pricing agreement within the common market. It is immaterial in that respect whether or not they had recourse to subsidiaries, agents, sub-agents, or branches within the Community in order to make their contracts with purchasers within the Community.’82 As a critical element of the impugned conduct had occurred within the EEC, the territorial principle applied to give the EEC jurisdiction over the defendants other than KEA.83 The ECJ rejected the suggestion that an international rule of non-interference (requiring the EEC to reject jurisdiction) applied in the circumstances.

79 European Commission Decision 85/202/EEC of 19 December 1984 relating to a Proceeding under Article 85 of the EEC Treaty [1985] OJ L 85/1 (hereafter ‘Wood Pulp’). 80 Webb-Pomerene Act of 1918, 15 USC §§ 61–66 (2005). 81 Treaty Establishing the European Economic Community [1957] (opened for signature 25 March 1957) 298 UNTS 11, art 85 (entered into force 1 January 1958) now the Consolidated Version Treaty on the Functioning of the European Union [2012] OJ C 326 art 101 (hereafter TFEU). 82 Cases C-89, 104, 114, 116–17, 125–129/85 A Åhlström Osakeyhtiö and Others v Commission of the European Communities [1988] ECR 5193, 5243. 83 As KEA had not played a separate role in the implementation of the price agreements, the EC decision against it was void.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

366 Comparative competition law

The test from Wood Pulp was applied by the Court of First Instance in Gencor.84 This case involved a merger between two South African companies, one of which was a subsidiary of a UK company. If permitted, the merger would lead to a duopoly on the world platinum market. The South African competition authority did not object to the merger. The companies sold their product into the European Community, thus satisfying the Wood Pulp implementation test. The Court referred to the acceptability of the effects doctrine as a matter of international law,85 although in fact it applied a form of the implementation test. The European Community’s rules governing which mergers are examinable under the Merger Regulation are likely to ensure that the EU Merger Regulation will continue to produce potentially controversial cases. If the annual turnover of the merging businesses exceeds specified thresholds in terms of global and European sales, the proposed merger must be notified to the EC, which must then examine it.86 In many cases the Wood Pulp implementation test will undoubtedly deliver a similar result to the US effects doctrine – for example, the EC also applied Community law to fine the members of the lysine cartel, all of whom were non-EU firms. The ECJ recently upheld the EC’s decision to fine Japanese firms that took part in the Gas Insulated Switchgear global cartel: the Japanese firms agreed not to enter the European market and the European firms agreed not to enter the Japanese market.87 There may, however, be limited circumstances where a US effects doctrine would deliver jurisdiction, but not the implementation test. An example of the broader scope of the US effects doctrine is perhaps to be found in Wood Pulp itself. Under the US effects doctrine KEA would probably have been liable.88

84 Case T-102/96 Gencor Ltd v Commission [1999] ECR II-753, [1999] 4 CMLR 971. 85 ‘[A]pplication of the Merger Regulation is justified under public international law when it is foreseeable that a proposed concentration will have an immediate and substantial effect in the Community’: ibid 90. 86 Council Regulation (EC) 139/2004 of 20 January 2004 on the control of concentrations between undertakings (Merger Regulation) [2004] OJ L 24, 1-22, art 1. 87 Joined Cases C-239/11 P, C-489/11 P and C-498/11 P Siemens AG, Mitsubishi Electric Corp. and Toshiba Corp. v Commission OJ C 52/5. 88 As discussed above the EC, applying an effects test, held KEA liable: Wood Pulp (n 79).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 22 /

Date: 13/10

JOBNAME: Duns PAGE: 23 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 367

4.2 Single Economic Entity Doctrine The EC has long favoured an effects test. It applied such a test in the 1969 Dyestuff case and again 20 years later in Wood Pulp. In Dyestuff the Advocate General supported such a test. However the ECJ was able to avoid the issue by relying on the single economic entity doctrine.89 Firms outside the Community were within the Community’s jurisdiction if they had subsidiaries within the Community.90 The operation of this doctrine is not automatic – evidence that the subsidiary acts independently of the parent means the doctrine will not apply.91 In Hydroferm the Court explained the doctrine: ‘Community competition law recognizes that different companies belonging to the same group form an economic unit and therefore an undertaking within the meaning of Articles 81 and 82 of the Treaty if the companies concerned do not determine independently their own conduct on the market.’92 That the subsidiary ‘determined independently its own conduct on the market’ is often difficult to establish.93 Where the subsidiary is wholly owned by the parent, there is a rebuttable presumption that the parent controls the subsidiary so that for purposes of EU competition law they are to be regarded as a single unit.94 The single economic entity doctrine, when applied to claim jurisdiction over foreign conduct, relies on the active nationality principle of jurisdiction; states have jurisdiction over their own nationals even where they 89 The doctrine reflects the court’s construction of the expression ‘Undertaking’, which is used throughout the EU competition provisions: Imperial Chemical Industries Ltd v Commission [1972] ECR 619 (hereafter ‘Dyestuff ’). 90 Ibid. 91 In Cooper Tire v Shell Chemicals UK Ltd & Ors [2009] EWHC 2609 (Comm) the UK Court of Appeal commented that just because one subsidiary breached the competition laws did not necessarily justify claiming jurisdiction over another subsidiary of the same parent company. The Court of Appeal did not have to decide the matter but commented that if it had to it would have considered referring the matter to the ECJ. See also Roche Products Ltd v Provimi Ltd [2003] 2 All ER (Comm) 683. 92 Case T-203/01 Michelin v Commission [2003] ECR II-4071 [290] citing Case 170/83 Hydrotherm Gerätebau GmbH v Compact del Dott. Ing. Mario Andreoli & C. Sas. [1984] ECR 2999 [11]. See also Case C-97/08 P Akzo Nobel NV and Others v Commission of the European Communities [2009] ECR I-08237, [58] and the Commission Decision in Case COMP/F/38.899 Gas Insulated Switchgear C(2006) 6762 final of 24 January 2007 [334]. 93 See, for example, Case C-90/09 P General Química v Commission [2011] ECR I-1. 94 Case C-97/08 Akzo Nobel NV v Commission [2009] ECR I-8237.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

368 Comparative competition law

are acting outside the borders of the state. European reluctance to proclaim a full-blown effects doctrine may have something to do with the traditional hostility of some Members of the EU to US extraterritoriality.95 The UK, for example, has often reacted unfavourably to US extraterritorialism as have France and Germany. The single economic entity doctrine, however, is useless where the foreign conspirators do not have Community subsidiaries. This occurred in Wood Pulp. 4.3 Private Actions in the EU Private enforcement has played only a small part in European competition law.96 The situation is slowly changing, due in part to the ECJ decisions in Courage and Manfredi and in part to the enthusiastic advocacy of the European Commissioners responsible for competition policy.97 The EC under Mario Monti and later Neelie Kroes championed private enforcement.98 The EC is convinced that private actions are necessary to optimize deterrence. In other words, private enforcement is seen as a valuable regulatory tool. 95 DG Goyder, EC Competition Law (4th edn, Oxford University Press 2003) 499. 96 See Wouter PJ Wils, The Optimal Enforcement of EC Antitrust Law: Essays in Law & Economics (Wolters Kluwer 2002) 14, claiming that damages awards in the EU are statistically insignificant. For an overview of private enforcement in the EU see Bojana Vrcek, ‘Overview of Europe’ in Foer and Cuneo (n 9) 277–95. 97 Case C-453/99 Courage Ltd v Crehan [2001] ECR I-6297; Case C-295/04 Manfredi v Lloyd Adriatico Assicurazioni SpA [2006] ECR I-6619. For a discussion on the number of private competition cases see U Boge and K Ost, ‘Up and Running, Or Is It? Private Enforcement – the Situation in Germany and Policy Perspectives’ [2006] European Competition Law Review 197; Barry Rodger, ‘Private Enforcement of Competition Law, The Hidden Story: Competition Litigation Settlements in the United Kingdom’ [2008] European Competition Law Review 96; J Kortmann and C Swaak, ‘The EC White Paper on Antitrust Damages: Why the Member States are (Right to Be) Less than Enthusiastic’ [2009] European Competition Law Review 340. 98 Under Monti and Kroes’ leadership the EC commissioned the Ashurst Report, produced the Green Paper on Damages Actions (2005) and the White Paper (2008). See respectively Denis Waelbroeck, Donald Slater and Gil EvenShoshan, ‘Study on the Conditions of Claims for Damages in Case of Infringement of EC Competition Rules: Comparative Report’ (Ashurst 2004); European Commission, ‘Damages Actions for Breach of the EC Antitrust Rules’ (COM(2005) 672); European Commission, ‘White Paper on Damages Actions for Breach of the EC Antitrust Rules’ (COM(2008) 165).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 24 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 369

European law does not provide for damages actions in European courts; therefore, any such actions must be brought in the national courts of the member states. Until there is a Council Directive on the issue, national procedural rules apply. These rules vary considerably from member state to member state. An EC Draft Directive on procedures to facilitate private damages actions in competition cases appeared in 2009, but was later removed from the EC agenda.99 The Draft Directive was based on proposals made in the EC’s 2008 White Paper. The shelving of the Draft Directive could be due to widespread concern about the ‘excesses’ of US-style private litigation, particularly class actions. US-style litigation is a concern to many member states.100 A new EU Directive on damages became law in November 2014 and must be implemented by member states by December 2016.101 The Directive covers such procedural issues as access to evidence, whether national competition authority decisions constitute proof of infringement, limitation periods, passing on defence, quantum of damages and joint and several liability. It does not provide for treble damages. The EC has released a Recommendation on Collective Redress.102 Given the suspicion with which class actions are viewed in certain member states, it is not surprising that the EC recommends an opt-in system. Even in the UK, which has perhaps the most positive approach to private enforcement, the incentives to seek damages do not match those in the US – neither treble damages nor punitive damages are available, costs are based on the ‘costs follow the event’ rule, there is no scope for a cy-près order, contingency fees are not permitted. Stand-alone actions are permitted in the UK but normally a private action will follow on from 99

European Commission, ‘Proposal for a Directive on rules governing damages actions for infringements of Article 81 and 82 of the Treaty’ (2009). Articles 81 and 82 are now arts 101 and 102 of the TFEU. 100 Commissioner Almunia has also commented on the need ‘to avoid the excesses of the US system’. J Almunia, ‘Common standards for group claims across the EU’ (Speech delivered at University of Valladolid, School of Law, 15 October 2010). See also Kortmann and Swaak (n 97) 340 discussing the reaction of member states and the European Parliament’s Economic and Monetary Affairs Committee. 101 Directive of the European Parliament and of the Council on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union [2014] OJ L 349/1. 102 Commission recommendation (EC) 396/2013 on common principles for injunctive and compensatory collective redress mechanisms in the Member States concerning violations of rights granted under Union Law [2013] OJ L 201.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 25 /

Date: 13/10

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

370 Comparative competition law

a public investigation.103 In 2013 the UK government announced that it would introduce a new opt-out class action regime for competition cases. This is likely to make the UK an even more attractive choice for European damages actions. Jurisdiction within the Community is governed by Council Regulation 44/2001 (Brussels Regulation).104 Under the Brussels Regulation the defendant may be sued in its place of domicile105 or, where there is more than one defendant, in the place of domicile of one of them.106 In a tort action the defendant at the election of the claimant may be sued in the place where the harmful event occurred or may occur.107 To bring an action in the UK it is sufficient that the UK-domiciled defendant is part of an ‘undertaking’ (single economic unit) as understood in EU law.108 It is not necessary to plead that the UK-domiciled defendant knew of the cartel.109 The English judicial system is attractive to claimants for a number of reasons. These include pre-trial discovery and trial procedures (e.g. cross-examination of witnesses). The UK may also be more 103 Vincent Smith, Anthony Maton and Scott Campbell, ‘England and Wales’ in Foer and Cuneo (n 9) 297. 104 Council Regulation (EU) 44/2001 of 22 December 2000 on jurisdiction and the recognition and enforcement of judgments in civil and commercial matters, [2001] OJ L 12/1, arts 2(1), 5(3) (hereafter ‘Brussels Regulation’). See also Convention on Jurisdiction and the Enforcement of Judgments in Civil and Commercial Matters, opened for signature 16 September 1988, 1659 UNTS 202, art 5(3) (entered into force 1 January 1992) (‘Lugano Convention’), which covers non-EU European states. 105 Brussels Regulation (n 104) art 2(1). 106 ‘[P]rovided the claims are so closely connected that it is expedient to hear and determine them together to avoid the risk of irreconcilable judgments resulting from separate proceedings’: Brussels Regulation (n 104) art 6(1). 107 Brussels Regulation (n 104) art 5(3). See Deutsche Bahn AG & Ors v Morgan Advanced Materials plc & Ors [2013] EWCA Civ 1484. Other provisions may apply in relevant circumstances. 108 See Provimi Ltd v Roche Products Ltd [2003] EWHC 961 (UK courts had jurisdiction to hear a damages case where the anti-competitive cartel was organized outside the UK (but inside the Community) and the injury was sustained outside the UK by a German claimant. UK jurisdiction was based on the fact that a subsidiary of one of the cartelists had implemented the cartel in the UK by charging the cartel price. It was not necessary to impute knowledge of the cartel to the UK subsidiary. 109 See also Cooper Tire v Dow Deutschland [2010] EWCA Civ 864 (a case involving the rubber cartel). Thus the court rejected the notion that the UK version of the corporate veil applied. For a case in which the court rejected jurisdiction see Sandisk Corporation v Koninklijke Philips Electronics NV & Ors [2007] EWHC 332.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 27 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 371

attractive because of its specialist competition tribunal, the Competition Appeal Tribunal (CAT), and by the introduction in the near future of opt-in class actions. The issue of exterritorial enforcement in the member states of the EU is important where the impugned conduct does not have a Community dimension. In such cases jurisdiction will depend on local law. In Germany, the Law against Restraints on Competition (GWB) provides that the Act ‘applies to all restraints of competition having an effect within the territorial scope of this Act, even if they are caused outside the territorial scope of this Act’.110 The Act does not define what effects are required. According to Buxbaum, ‘[t]o generalize from the tests developed in different areas of [German] competition law, domestic effects must be direct, considerable, actual, and foreseeable in order to support legislative jurisdiction’.111 Where in asserting jurisdiction there would be actual or potential conflict with another state, the German courts look to international norms to determine whether jurisdiction ought to be claimed.112 In keeping with the decision of the ECJ in Courage v Crehan, the German legislature amended the GWB to facilitate private actions. German civil procedures, however, remain conservative.113 Treble damages, class actions and contingency fees are not available. Discovery rules are much stricter than those in the US. The extraterritorial situation in other EU member states is similar.114

5. EXTRATERRITORIAL REACH IN A RANGE OF OTHER STATES Extraterritorial reach is now quite common among competition regimes. Many are based on an effects doctrine. Others are more limited in scope, normally requiring that a defendant be incorporated within the state or 110 Gesetz Gegen Wettbewerbsbeschr-ankungen (Law against Restraints on Competition) 1957 discussed in Hannah L Buxbaum, ‘Territory, Territoriality, and the Resolution of Jurisdictional Conflict’ (2009) 57 American Journal of Comparative Law 631. 111 Ibid 641. 112 Ibid 652–61. 113 Hannah Buxbaum, ‘German Legal Culture and the Globalization of Competition Law: A Historical Perspective on the Expansion of Private Antitrust Enforcement’ (2005) 23 Berkeley Journal of International Law 474. 114 On France see IBA (n 14) Annex.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 27 /

Date: 13/10

JOBNAME: Duns PAGE: 28 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

372 Comparative competition law

carrying on business within the state.115 While extraterritoriality is common in law, it has tended to be less common in practice, though growing. Few states have utilized the potential extraterritorial reach of their laws in the same way that the US or even the EU have. This is due in part to an historical reluctance to act extraterritorially, in part to a lack of power and/or resources, in part to a lack of knowledge and in part to the lack of private actions. As previously discussed, even the EU is still wrestling with the appropriate framework to be applied to private enforcement. One area where extraterritorial considerations are becoming increasingly more important is mergers. An increasing number of mergers have international implications. As a consequence they are likely to be a suitable subject for competition examination by a number of states. Often the state will have to assume extraterritorial jurisdiction to conduct such an examination. A growing number of states are prepared to do this. 5.1 Japan Japan, which (at least in theory) has had a competition regime since 1948, has almost completely ignored the role of extraterritorial enforcement. Recently, however, in keeping with a more vigorous approach to competition regulation generally, there have been signs of a shift to a broader interpretation of the jurisdictional reach of the Antimonopoly Act (AMA).116 This is perhaps most notable in the case of mergers.117 For example, in 2008, the Japan Fair Trade Commission (JFTC) commenced an enquiry into the proposed merger between global resources giants, BHP Billiton and Rio Tinto, neither of which had a presence in Japan.118 As the merger did not proceed, no action was ultimately required. Due to the lack of cases it is not clear what principles underpin the extraterritorial reach of the AMA. If those principles reflect an effects 115

See discussion on Australia and New Zealand below at text accompanying

n 160. 116 Act Concerning Prohibition of Private Monopolization and Maintenance of Fair Trade, Act No 54 of 1974 (‘AMA’). See Naoki Ohkubo and Zenichi Shishido, ‘Cooperation, Comity and Competition Policy in Japan’ in Guzman (ed) (n 78) 88–9 discussing two cases which have been used by scholars to argue for an effects test in Japan. Neither case, however, ultimately relied on an effects test. 117 Foreign mergers are subject to the mandatory pre-merger notification requirements. 118 Ohkubo and Shishido (n 116) 88–9.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 28 /

Date: 13/10

JOBNAME: Duns PAGE: 29 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 373

doctrine,119 the effects necessary to attract jurisdiction are most likely those set out in the substantive provisions, that is, the foreign conduct must produce a ‘substantial restraint on competition’ in Japan or a ‘tendency to impede fair competition’ in Japan.120 Thus, it is probably not necessary to prove that the participants in the foreign conduct intended to contravene the AMA.121 Private actions are available in Japan, but for a variety of reasons have been rare when compared to the US but not so when compared to EU member countries.122 To conclude, the extraterritorial reach of the AMA (both in theory and practice) is still far from settled. However given Japan’s traditional preference for horizontal and vertical linkages between commercial entities, it should not be assumed that Japan will readily develop a taste for the aggressive pursuit of foreign conduct. 5.2 Korea Korean competition law was introduced in 1981 but only enforced vigorously from the 1990s onwards.123 Generally speaking Korea has not only shown a greater willingness to enforce its competition laws than has Japan, but has also shown a greater enthusiasm for enforcing those laws beyond Korea’s territorial boundaries. The Korea Fair Trade Commission (KFTC) adopted the effects doctrine and as a result has pursued a number 119

Ibid 88–9. See AMA (n 116), Articles 2(5) and (6) [‘substantial restraint of competition’] and 2(9) [‘tendency to impede fair competition’] discussed in Ohkubo and Shishido (n 116) 89. The JFTC has not published any guidelines on the issue of extraterritorial jurisdiction. 121 See Ohkubo and Shishido (n 116) 89–90. Contrast the position in the US discussed at text accompanying n 25 above. 122 Private actions may be brought under s 709 of the Civil Code or under s 25 of the AMA. The advantage of the Civil Code is that it allows for stand-alone actions, the disadvantage is that the courts tend to require strict proof. Action under the AMA, on the other hand, is only available after the issuance of an administrative order by the JFTC. If, however, an order has been issued, liability is strict. See discussion in Hiromitsu Miyakawa, ‘Japan’ in Foer and Cuneo (n 9) 533–41. For a statistical analysis of private actions in the US, EU countries and Japan see Simon Vande Walle, ‘What Keeps Plaintiffs Away from the Court? An Analysis of Antitrust Litigation in Japan, Europe and the US’ (SSRN February 2014) or . 123 Korean competition law is found in the Monopoly Regulation and Fair Trade Act (‘MRFTA’). See OECD, ‘Korea – Updated report on competition law and institutions’ (OECD 2004). 120

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 29 /

Date: 13/10

JOBNAME: Duns PAGE: 30 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

374 Comparative competition law

of prosecutions against foreign participants including not only global cartels, but also abuse of dominance cases.124 In 2004 the effects doctrine was given legislative approval. The Monopoly Regulation and Fair Trade Act (MRFTA) provides that the ‘Act shall apply to activities carried out overseas when they are deemed to have an effect on the domestic market’.125 The level of effect required is not defined. Nevertheless the Guidelines made pursuant to the MRFTA set out the types of effect required and importantly those effects that are to be disregarded.126 The KFTC has continued to apply the effects doctrine to foreign activities. For example, significant fines were imposed on the participants in the global LCD (liquid crystal display) cartel and the marine hose cartel among others. The Korean pre-merger notification rules as well as its merger rules apply to all relevant mergers even those between foreign corporations.127 The KFTC has indicated that it will continue to closely monitor overseas transactions that may potentially cause harm to domestic consumers.128 Private enforcement is being encouraged in Korea and is on the increase.129 So far all private actions have been against domestic enterprises, but this may change in the future. One factor restricting the use of private actions is the prohibition in Korea on class actions. Given the cost of private enforcement, particularly where defendants are located outside the jurisdiction, this is probably a significant disincentive.

124 The KFTC prosecuted participants in the vitamins cartel (2002) and the graphite electrodes cartel (2002). It also took action against US-based companies Microsoft Corporation (2006) and Intel (2008) for abuse of dominance. Microsoft was fined $32 million for unlawful bundling. Intel was fined $25 million for providing loyalty rebates to its customers (equipment manufacturers) on condition that they do not acquire a competing product from Intel’s competitor. The KFTC also applied conduct remedies. 125 Monopoly Regulation and Fair Trade Act, art 2.2. 126 See, for example, KFTC, ‘Guidelines for Concerted Practice Review’ (KFTC 2009) . 127 For example, the KFTC investigated the BHP Billiton and Rio Tinto proposed merger in cooperation with other competition agencies. 128 See among other things, KFTC, ‘Guidelines for Notification on Combination of Enterprises’ (the pre-merger filing guidelines) (KFTC 2014). 129 For a discussion on private enforcement in Korea, see Hwang Lee and Byung Geon Lee, ‘Korea’ in Foer and Cuneo (n 9) 542–52.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 30 /

Date: 8/9

JOBNAME: Duns PAGE: 31 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 375

5.3 China China introduced a comprehensive competition law for the first time in 2007.130 Article 2 of China’s Anti-Monopoly Law (AML) states: This law is applicable to monopolistic conduct outside the territory of the People’s Republic of China which has an eliminative or restrictive impact on competition in the domestic market.

To date s 2 has not been applied.131 However this is probably due to a lack of opportunity rather than willingness. Indications are that China is prepared to adopt an expansive approach to the territorial principle of subject-matter jurisdiction (the objective territorial principle which includes an effects doctrine rather than the strict territorial principle). Recently the National Development and Reform Commission (NDRC) applied the competition provisions of the 1998 Price Law extraterritorially to participants in the global LCD panels cartel.132 Two Korean firms (Samsung and LG) and four Taiwanese firms were fined for fixing prices in a series of meetings held in Korea and Taiwan.133 The Price Law was applied because the AML was not in force at the time the impugned cartel activity occurred.134 There is no reason why the same approach will not apply to the AML. In fact, unlike the AML, the Price Law contains no specific reference to extraterritorial application. China’s merger law applies to a merger between two offshore entities that affects market competition in China. This clearly envisages extraterritorial application based on an effects doctrine of some sort. China applied that doctrine recently in the Glencore/Xstrata merger. Glencore 130

The Anti-Monopoly Law (AML) was adopted on 30 August 2007 and became effective on 1 August 2008. 131 But see below discussion on mergers. See Michael Faure and Xinzhu Zhang, ‘Towards an Extraterritorial Application of the Chinese Anti-Monopoly Law that Avoids Trade Conflicts’ (2013) 45 George Washington International Law Review 101. 132 See Nate Bush, ‘NDRC penalises global LCD price cartel’ (China Law and Practice, March/April 2014) . 133 Zhan Hao, ‘The Price Law or the Anti-Monopoly Law: Observation of NDRC’s Antitrust Enforcement in China’ (China Law Vision, 22 January 2013) . 134 Ibid. Both laws still apply and the NDRC has used the Price Law even when the AML was available.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 31 /

Date: 13/10

JOBNAME: Duns PAGE: 32 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

376 Comparative competition law

International plc agreed to acquire all the shares in Xstrata plc.135 Neither company had a presence in China. However, the merger was important to China because of the significant role that Glencore and Xstrata played in the global supply of raw materials, particularly copper concentrate, zinc concentrate and lead concentrate. These raw materials were important for Chinese economic development.136 Following a long investigation the Ministry of Commerce of the People’s Republic of China (‘MOFCOM’) finally approved the merger in April 2013 subject to conditions.137 The conditions included both structural and conduct remedies. One of the conditions required divesture by September 2014 of Xstrata’s Las Bambas copper mine project in Peru to a party approved by MOFCOM.138 MOFCOM has imposed conditions on at least three mergers between non-Chinese companies that had been unconditionally approved by US and European regulators.139 It has also imposed conditions on a joint 135 See ‘MOFCOM Conditionally Clears Glencore’s Acquisition of Xstrata (MOFCOM, 16 April 2013)’ (2013) 25 China Competition Bulletin. 136 See also the Marubeni/Gavilon merger, which involved the supply of soy beans, another important Chinese import. It is notable that in both cases the market shares involved were quite low. 137 Previously China (like Japan and Korea) had shown a willingness to involve itself in the proposed merger of BHP Billiton and Rio Tinto. 138 Both the US and Australian competition authorities had unconditionally approved the merger. The EU approved the merger subject to conditions, albeit different conditions to those imposed by MOFCOM. See also discussion of MOFCOM’s conditional approval of a merger between two Russian companies: ‘Antitrust Alert: China Approves Merger between Russian Potash Producers but Requires They Continue to Supply the Chinese Market’ (Jones Day, 13 June 2011) . Also Christopher Bright and others, ‘China: China Conditionally Approves Thermo Fisher/Life Technologies Merger: MOFCOM Continues To March To Its Own Drumbeat’ (Mondaq 12 March 2014) . 139 See MOFCOM, ‘Announcement No. 25, 2012 of the Ministry of Commerce – Announcement of Approval with Additional Restrictive Conditions of the Acquisition of Motorola Mobility by Google’ (Ministry of Commerce PRC, 31 May 2012) ; the GM/Delphi merger is discussed in Margaret Wang, ‘China’s Current Approach to Vertical Arrangements Under the Anti-Monopoly Law’ [2012] Competition Policy International ; for the Seagate/Samsung transaction, which had been cleared by seven other competition

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 32 /

Date: 8/9

JOBNAME: Duns PAGE: 33 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 377

venture ‘established outside China between non-Chinese firms and with a competition concern (foreclosure) primarily related to a global IP portfolio of which 90% of its licensing revenue comes from outside of China’.140 Private actions are available in China to those injured by anticompetitive conduct.141 It is not necessary to wait for one of the enforcement agencies to take action first. According to commentators the number of private cases is gradually growing.142 All indications are that China will vigorously protect its economic interests. If those interests are threatened by foreign conduct, then China has the means and the will to pursue the participants. 5.4 India India has also recently overhauled its competition laws. The Competition Act 2002 provides that extraterritorial jurisdiction exists where foreign conduct produced an ‘appreciable adverse effect on competition in the relevant market in India’. This applies to all forms of anti-competitive activity as well as mergers. As with the majority of jurisdictions India now has a mandatory pre-merger reporting regime which applies to both domestic and foreign mergers. Entirely foreign transactions which are not likely to have a significant local nexus and effect on markets in India are ordinarily exempt. Presumably the nexus is established where the likely

authorities see Michael Han and Zhaofeng Zhou, ‘MOFCOM’S Approach to Merger Remedies: Distinctions from Other Competition Authorities’ [2012] Competition Policy International 5–6 . 140 David Stallibrass, ‘The Chinese MOFCOM Conditionally Clears an Off-shore Joint Venture Involving European Technology Groups (ARM/Giesecke & Devrient/Gemalto)’ (Institute of Competition Law, 6 December 2012) . 141 AML art 50. The PRC Supreme People’s Court released the ‘Provisions on Certain Issues Concerning the Application of Law in Hearing Cases Involving Civil Disputes Arising out of Monopolistic Behaviors’ in May 2012 after many years of discussion. The Provisions became effective on 1 June 2012. 142 For a discussion on private enforcement in China, see Susan Beth Farmer, ‘Recent Developments in Regulation and Competition Policy in China: Trends in Private Litigation’ in Michael Faure and Xinzhu Zhang (eds), The Chinese Anti-Monopoly Law: New Developments and Empirical Evidence (Edward Elgar 2013) 15–72.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 33 /

Date: 8/9

JOBNAME: Duns PAGE: 34 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

378 Comparative competition law

effect is an ‘appreciable adverse’ one on a local market. Private enforcement of local competition complaints, let alone extraterritorial actions, has been very rare in India. Commentators, however, are hopeful that this will change as the Competition Act settles in.143 5.5 Brazil Article 2 of Brazil’s competition law explicitly incorporates the ‘effects’ test into Brazilian competition law.144 This applies to cartels, single firm conduct and mergers. To date the scope of the Brazilian effects doctrine has not been judicially tested.145 However, the Brazilian antitrust agency, CADE,146 has applied the effects test to intervene in a number of mergers.147 Private actions, while permitted in Brazil, are very rare. 5.6 Canada The Competition Act makes no reference to the extraterritorial reach of its substantive provisions.148 This, therefore, is a matter of statutory construction. According to the Supreme Court of Canada, Canada has a ‘legitimate interest in prosecuting persons for activities that take place abroad but have an unlawful consequence’ in Canada.149 Not every consequence will be sufficient. Extraterritorial application of Canadian law depends on there being a ‘real and substantial connection’ between the subject matter and Canada.150 A ‘real and substantial connection’ means that a significant portion of the activities constituting the offence 143 See Pradeep Mehta and Cornelius Dube, ‘India’ in Foer and Cuneo (n 9) 507–13. 144 Federal Antitrust Law, Federal Law 8884 of 1994. 145 See Luciano Benetti Timm, ‘Jurisdiction, Cooperation, Comity and Competition Policy in Brazilian International Antitrust Law’ in Guzman (ed) (n 78) 70–2. 146 Conselho Administrativo de Defesa Econômica. 147 See Timm (n 145) 72. 148 The exception is s 46 of the Competition Act. Section 46 is specifically designed to capture international cartels by making any corporation carrying on business in Canada liable for a breach of Canadian law if it implements the cartel at the directive of a foreign firm (presumably the parent of the Canadian corporation) even though the Canadian corporation was not aware of the conspiracy. 149 Libman v The Queen [1985] 2 SCR 178, 209. 150 Ibid 212–13. A ‘real and substantial connection’ has been suggested as the overarching criterion for subject-matter jurisdiction: see S Coughlan, R Currie, H Kindred and T Scassa, ‘Global Reach, Local Grasp: Constructing

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 34 /

Date: 8/9

JOBNAME: Duns PAGE: 35 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 379

must take place in Canada.151 This provides scope for some form of an effects doctrine. Initially Canada adopted a restrained approach to the extraterritorial reach of its competition laws.152 However, in keeping with other states the restrictive reins have been somewhat loosened in recent years. Certainly the Competition Bureau has made it clear that it will pursue foreign participants of cartels formed overseas that adversely affect Canadian markets.153 The Canadian government has also indicated a more expansive view as demonstrated by its amicus curiae briefs filed in Hartford and Empagran.154 As with other states extraterritorial application is most likely to figure in relation to mergers. The Competition Bureau’s approach to multijurisdictional mergers is set out in the Bureau’s Merger Remedies in Canada.155 Where possible, Canada will seek to coordinate its remedies with other states. This may involve not taking any action where a foreign remedy is sufficient. However there are circumstances where the Bureau will take particular notice. According to the Bureau: While enforcement decisions are made on a case by case basis, the Bureau is more likely to formalize negotiated remedies within Canada when the matter raises Canada-specific issues, when the Canadian impact is particularly significant, when the asset(s) to be divested reside in Canada, or when it is critical to the enforcement of the terms of the settlement.156 These circumstances clearly contemplate an extraterritorial role.

Canada permits private actions for damages, including class actions. Class actions are opt-out and are becoming more popular. In 2013 plaintiffs settled a class action against Samsung Electronics Co. and other DRAM (dynamic random access memory) manufacturers for conspiracy to fix the prices for DRAM or products which contain DRAM in Extraterritorial Jurisdiction in the Age of Globalization’ (2007) 6 Canadian Journal of Law and Technology 29–60. 151 Libman v The Queen (n 149) 212–13. 152 See IBA (n 14) 49. 153 See Edward Iacobucci, ‘The International Reach of Canadian Competition Law’ in Guzman (ed) (n 78) 48–9 citing former Commissioner of Competition, Konrad von Finckenstein. 154 See George Addy, Chris Margison and Ryan Doig, ‘National Sovereignty and the Enforcement of Competition Law: Striking the Right Balance’ (Paper delivered at the Annual Fall Conference on Competition Law, Hull, Quebec, 23–24 September 2004) 6–7. 155 Competition Bureau, ‘Information Bulletin on Merger Remedies in Canada’ (Competition Bureau 2006). 156 Ibid 22.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 35 /

Date: 13/10

JOBNAME: Duns PAGE: 36 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

380 Comparative competition law

Canada.157 Recently, a class action against Mars Canada Inc., Cadbury Adams Canada Inc., Hershey Canada Inc. and Nestlé Canada Inc. for price fixing chocolate in Canada was also settled.158 In other respects (e.g. quantum of damages, costs), private enforcement is not as attractive as in the US. However, a recent decision of the Supreme Court permitting indirect purchaser actions while rejecting the passing on defence is more generous to plaintiffs than the position in the US.159 5.7 Australia Australia does not apply an effects doctrine. Mere effects, no matter how substantial, will not be sufficient to attract jurisdiction. Unlike the Canadian Competition Act, the Australian Competition and Consumer Act 2010 (Cth) (CCA) makes specific provision for its extraterritorial application.160 Section 5 of the CCA provides that the competition laws extend to the engaging in conduct outside Australia by bodies corporate incorporated or carrying on business within Australia or by Australian citizens or persons ordinarily resident within Australia.161 Where a cartel is hatched overseas but with effects in Australia, the issue will be whether the foreign participants can be said to have been carrying on business in Australia.162 There is no need for a firm to have a 157

See Competition Policy International, ‘Canada: Memory Chip Class Action Against Samsung Finalized’ (Competition Policy International, 3 June 2013) . 158 See King & Wood Mallesons, ‘Bitter Sweet Settlement for Chocolate Makers’ (In Competition Blog 26 September 2013) . 159 Pro-Sys Consultants Ltd. v Microsoft Corporation, 2013 SCC 57, SunRype Products Limited v Archer Daniels Midland Company, 2013 SCC 58 and Infineon Technologies AG v Option consommateurs, 2013 SCC 59. 160 Competition and Consumer Act 2010 (Cth). For New Zealand see s 4(1) of the Commerce Act 1986: see Poynter v Commerce Commission [2010] NZSC 38. 161 Section 5(2) extends extraterritorial application in cases of vertical dealing restraints and resale price maintenance to engaging in conduct outside Australia by any persons in relation to the supply by those persons of goods or services to persons within Australia. 162 See, for example, Norcast S.ár.L v Bradken Limited (No 2) [2013] FCA 235. A broader jurisdiction applies where the case is brought pursuant to one of the equivalent Australian state laws. Under state law the relevant provisions apply to: (a) persons carrying on business within this jurisdiction, or (b) bodies corporate incorporated or registered under the law of this jurisdiction, or (c)

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 36 /

Date: 8/9

JOBNAME: Duns PAGE: 37 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 381

place of business in Australia to be carrying on business in Australia. The foreign corporation may carry on business in Australia through an Australian agent (including a subsidiary). However, just because a foreign corporation has a subsidiary in Australia does not mean that the foreign corporation is carrying on business in Australia. Indeed, having regard to Australia’s laws concerning separate corporate identity, it will generally be very difficult to establish that a subsidiary is acting as an agent for its foreign parent.163 In Bray v F. Hoffmann-La Roche the Federal Court rejected the proposition that the European firms, Hoffmann-La Roche, BASF and Rhône Poulenc – the conspirators in the global vitamins cartel – were carrying on business in Australia through their Australian subsidiaries.164 Section 5 of the CCA does not apply the European economic unit doctrine.165 Nor is there a specific provision equivalent to Canada’s s 46. Although the extraterritorial provisions did not apply in Bray, the foreign conspirators were nevertheless held subject to the CCA because the relevant substantive provision made it unlawful to make or give effect to the cartel. The Court held that the European conspirators’ hands-on approach to implementing the cartel in Australia (e.g. communicating directly with local personnel working for the Australian subsidiaries and directing them in the implementation of the cartel regarding prices, quantities etc) amounted to giving effect to the cartel in Australia.166 This persons ordinarily resident in this jurisdiction, or (d) persons otherwise connected with this jurisdiction. See, for example, Competition Policy Reform (New South Wales) Act 1995 s 8 discussed and applied in ACCC v Prysmian Cavi E Sistemi Energia SRL (No 4) [2012] FCA 1323. 163 See generally IM Ramsay and DB Noakes, ‘Piercing the Corporate Veil in Australia’ (2001) 19 Company and Securities Law Journal 250. Merkel J said: ‘In my view something more than the indirect legal and commercial capacity of the parent companies to control and direct the subsidiaries, plus the parent’s involvement in implementing the cartel arrangement, is required to lift the corporate veil between the subsidiaries and their parents or to find that each of the subsidiaries is carrying on its business as agent for the parent. That is particularly so where it is contended (as it is in the present case) that the parent, rather than the subsidiary, is carrying on business in Australia or, put another way, the subsidiary is engaging in all of its commercial activities on behalf of, and therefore as agent for, the parent’: Bray v F. Hoffmann-La Roche Ltd [2002] FCA 243, [80]. 164 Bray v F. Hoffmann-La Roche Ltd [2002] FCA 243, [64]–[81]. 165 Section 44ZRC applies a group economic doctrine but from a jurisdictional point of view is subject to s 5. 166 Bray v F. Hoffmann-La Roche Ltd [2002] FCA 243, [158] upheld on appeal: Bray v F. Hoffmann-La Roche Ltd [2003] FCAFC 153 (appeal).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 37 /

Date: 13/10

JOBNAME: Duns PAGE: 38 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

382 Comparative competition law

was sufficient to bring the actions of the conspirators within the Court’s jurisdiction.167 Australia recognizes private actions. Major damages actions, however, have been fairly rare. The first class action involving damages for cartel activities (the vitamins cartel) was settled in 2006.168 While class actions suits are likely to be a growth area, the distribution of incentives suggests that such growth will not be explosive. Where the matter involves a claim for damages based on the extraterritorial application of the CCA, consent of the relevant Minister must first be obtained.169 Consent will be granted unless (1) the law of the country in which the conduct concerned was engaged in required or specifically authorized the engaging in of the conduct; and (2) it is not in Australia’s national interest that the consent be given.170 This provides scope for a comity analysis to be made by the relevant Minister. To conclude, Australia gives a more limited extraterritorial application to its competition laws than do either the US or the EU.171

6. CONCLUSION Some form of extraterritorial application of domestic competition law is essential in a globalizing world. States need to protect their citizens from conduct occurring offshore. However in venturing beyond territorial boundaries states are entering a space where others have a legitimate regulatory interest. To avoid the types of regulatory contest that occurred in the 1970s and 1980s states need to develop mechanisms to mediate these occasions of concurrent jurisdiction. A starting point, but only a starting point, is to develop some common understandings about the legitimate use of extraterritorial jurisdiction. This is not to say that states must have uniform jurisdictional rules, but rather that they must have a common understanding of extraterritorialism’s limits. Thus, states ought to recognize that they only have a 167 See also ACCC v Prysmian Cavi E Sistemi Energia SRL (No 4) [2012] FCA 1323. 168 Darwalla Milling Co Pty Ltd v F. Hoffmann-La Roche Ltd (No 2) [2006] FCA 1388. Other settlements were approved in Wright Rubber Products Pty Ltd v Bayer AG (No 3) [2011] FCA 1172 (rubber chemicals cartel) and Jarra Creek Central Packing Shed Pty Ltd v Amcor Limited [2011] FCA 671. 169 Competition and Consumer Act 2010 (Cth) s 5(3). 170 Ibid s 5(5). 171 In Bray v F. Hoffmann-La Roche Ltd [2002] FCA 243, [48] Merkel J suggested that it might be time for Australia to recognize an effects test.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 38 /

Date: 13/10

JOBNAME: Duns PAGE: 39 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

International governance of competition 383

legitimate interest in regulating foreign conduct if it causes substantial effects in a domestic market. In other words, there must be a sufficient nexus between the state asserting jurisdiction and the regulated conduct.172 Appropriate bases for such jurisdiction are already recognized in international law. The effects doctrine (perhaps more accurately characterized as the ‘substantial’ effects doctrine) represents a key element. In fact, there is a worldwide trend to adopt the effects doctrine. However, it is less clear that states have a uniform understanding of what that comprehends. However, harmonizing jurisdictional rules alone will never be sufficient. A convergence in competition standards will alleviate some of the pressures of concurrent jurisdiction. This will also be the case if states develop a better understanding of and respect for each other’s policies and needs.173 There is and will always be an important role for comity, particularly as an administrative tool, rather than as a judicial tool. An area of growing importance is mergers that have multijurisdictional effects. This trend will continue to highlight the importance of both a domestic jurisdictional rule that allows scope for extraterritorialism and a willingness to apply comity considerations to make that extraterritorialism function properly. Private enforcement raises concerns additional to those raised by public enforcement. Private actions are well entrenched in the US and increasing in other jurisdictions. Private actions present a challenge because they are more likely to be beyond administrative control. This reduces the likelihood of comity (as an administrative tool) playing a central role and increases the likelihood of courts being called upon to apply a form of judicial comity on a case-by-case basis. While the US Supreme Court has relied on comity to restrict the operation of a range of statutes (including antitrust provisions), it has not favoured comity as a tool for case-by-case judicial analysis.174

172

See IBA (n 14) 9. There are already a number of processes and institutions for advancing these goals. The development of negative and positive comity agreements, the work done by the Competition Law and Policy Committee of the OECD, the work done by the ICN on both standards and procedures, and the recent work on capacity building being done by UNCTAD have reduced policy conflicts and improved the regulation of international cartels and mergers. 174 See discussion above at text accompanying n 38. 173

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 16_Chapter13

/Pg. Position: 39 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

14. Private antitrust enforcement: Comparative and policy considerations Daniel A. Crane, Keith Klovers and Adam Speegle *

In the United States (US), there are approximately ten private antitrust lawsuits for every antitrust lawsuit by the principal public enforcers – the Justice Department’s Antitrust Division and the Federal Trade Commission.1 Until recently, the US was exceptional in this regard. In most other antitrust jurisdictions, most or all enforcement was public. In recent years, however, there has been lots of talk around the world about expanding private antitrust enforcement. Most prominently, the European Commission’s 2005 Green Paper and 2008 White Paper called for expansion of private antitrust enforcement of Articles 101 and 102 of the Treaty on the Functioning of the European Union (TFEU) within the European Union’s Member States. At this time, the issue remains under consideration. Beyond the EU, a number of growing antitrust jurisdictions are implementing or expanding private antitrust enforcement.2 This chapter introduces some of the key themes raised by private antitrust enforcement. Section 1 briefly summarizes historical and recent developments in the development of the private enforcement model. Section 2 considers whether private antitrust enforcement achieves its two primary assumed objectives – compensation and deterrence. Section 2 also considers the effects of prioritizing one goal or the other, and possible alternatives to compensation and deterrence as goals of private enforcement. Section 3 considers some of the effects of private antitrust * The views presented in this chapter are solely those of the authors, and not of their employers. 1 On the growth of private antitrust enforcement in the United States, see Daniel A Crane, The Institutional Structure of Antitrust Enforcement (Oxford University Press 2011). In particular chapters 3, 9, and 10. Portions of this chapter are based on those chapters. 2 See generally Albert A Foer and Jonathan W Cuneo (eds), The International Handbook of Private Enforcement of Competition Law (Edward Elgar 2010); Ilene K Gotts, The Private Competition Enforcement Review (3rd edn, Law Business Research 2010).

384

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 2 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 385

litigation on the overall system of antitrust enforcement, including possible negative or positive spillovers to public litigation and the creation of antitrust constituencies. Finally, Section 4 introduces some of the issues that arise when parties litigate antitrust cases across borders.

1. HISTORY AND RECENT DEVELOPMENTS 1.1 United States The Sherman Act of 1890 created a private right of action to enforce the newly enacted antitrust laws, including a right for the injured plaintiff to receive treble damages – three times its actual damages. However, prior to World War II, private antitrust litigation was relatively subdued, with the most important developments occurring in governmental antitrust cases. Following the war, private antitrust grew rapidly. The number of new filings more than doubled from the period 1945–1949 to the period 1950–1954. Private litigation surpassed government litigation (in pure numerical terms) in the 1940s. From the 1940s to about 1965, the ratio between private and public litigation was about 6:1.3 Thereafter, private litigation continued to grow at an astronomical rate, reaching a peak of 1,611 new private cases filed in 1977, with a 20:1 ratio in favor of private litigation.4 Beginning in the 1970s, the federal courts began to react negatively to the growing trends in private antitrust litigation, for reasons explored in Section 4.1 below. In part due to new antitrust doctrines – such as the limitation in indirect purchaser standing and the antitrust injury requirement – and in part due to more conservative attitudes in the courts on the substance of antitrust law, private antitrust enforcement declined during the late 1970s and into the 1980s and 1990s, eventually reaching the rough equilibrium ratio of 10:1 over public cases that generally holds today. Although private antitrust litigation has shrunk from its peak in the US, it continues to play a significant role in shaping US antitrust policy. For example, the US Supreme Court last heard an appeal in a government antitrust case in 1999.5 Since that time, it has decided 12 private antitrust 3 Steven C Salop and Lawrence J White, ‘Economic Analysis of Private Antitrust Litigation’ (1986) 74 Georgetown Law Journal 1001, 1002–03. 4 Ibid 1003. 5 California Dental Association v FTC, 526 US 756 (1999).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

386 Comparative competition law

cases. In Section 4.2 below, we consider ways in which private antitrust litigation has had important spillover effects on public litigation. 1.2 European Union Unlike the US, Europe has seen very little private antitrust litigation.6 An external study supported by the European Commission found that, between May 1, 2004 and the end of 2007, 96 damages actions were filed within the Union, concentrated in only ten Member States.7 The Commission estimates that the lack of an effective private enforcement mechanism may be costing European consumers and businesses more than 20 billion euros per year.8 The principal obstacles to private antitrust enforcement in Europe are more practical than dogmatic. The European Court of Justice has held that compensation should be available in Member State courts for violations of Articles 101 and 102 of the TFEU.9 However, litigants face many practical obstacles to initiating successful antitrust actions, such as the limited availability of claim aggregation and discovery and stringent requirements of proof of liability, causation, and damages. In light of these background conditions, the Commission has been considering ways to expand the availability of private litigation in Europe since at least 2005 when it issued its Green Paper,10 updated in 2008 with a White Paper.11 Some of the important elements of these papers are considered in Section 2. The Commission’s proposal stalled amid political controversy over the expansion of private litigation under a Europewide mandate. The Commission continues to study and consult on a number of issues presented by private antitrust enforcement, including claim aggregation and the computation of damages.

6

Bojana Vreck, ‘Overview of Europe’ in Foer and Cuneo (n 2) 277. Ibid. 8 Ibid. 9 Case C-453/99 Courage Ltd. v Crehan [2001] ECR I-6297; Joined Cases C-295-298/04 Manfredi v Lloyd Adriatico Assicurazioni [2006] ECR I-6619. 10 European Commission ‘Green Paper’ (European Commission Website, 2005) . 11 European Commission ‘White Paper’ (European Commission Website, 2008) . 7

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 387

1.3 South Africa, Asia, and South America Despite advances by public antitrust enforcement agencies around the world, private antitrust enforcement in many countries remains relatively weak. While reforms that might expand the availability of a private antitrust remedy are under discussion, relatively few private cases have been filed to date in Africa, Asia, and South America. South Africa has limited private enforcement experience. In 1999 the Competition Act created a right for private parties to pursue civil damages suits following a final determination by South African competition authorities.12 In 2006 the first follow-on damages case was brought against South African Airways for abuse of dominance related to loyalty rebates to travel agents.13 Since this case, however, little has developed beyond rumors of potential future actions.14 Similarly, private enforcement in South America is still in its infancy. In Peru, private actions are almost nonexistent due to requirements placed on plaintiffs by its antitrust enforcement regime.15 Argentina also has few private enforcement actions, not witnessing its first successful private damages suit until 2009.16 Brazil has established a system where private plaintiffs may obtain injunctive relief or damages through an administrative or judicial body, or both.17 Despite this comprehensive framework for relief, relatively few private antitrust actions have been brought. This has been attributed in part to a general lack of awareness, knowledge, and

12 Section 65, Competition Act, No 89 of 1998 (Nov. 30, 1998) . 13 Case No. 83/CR/Oct04. 14 Shawn van der Meulen, ‘The possibility of Class Action claims for damages against SAA for anti-competitive conduct’ (Webber Wentzel, May 14, 2012) . See also Kasturi Moodaliyar, ‘South Africa’ in Foer and Cuneo (n 2). 15 See Alfredo Bullard and Alejandro Falla, ‘Peru’ in Foer and Cuneo (n 2), describing restrictions such as a requirement for exhaustion of administrative remedies before an independent government agency and prohibition of direct private actions before ordinary courts. 16 Auto Gas SA c/ YPF SA y otro s/ ordinario, Commercial Court No 14, Clerk’s Office 27 (September 16, 2009). The law under which antitrust actions are brought became effective in 1999. Law No 25,156, BO 5 (September 16, 1999, effective September 28, 1999). 17 Law No 8,884/94 (June 11, 1994).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

388 Comparative competition law

expertise in antitrust.18 There is, however, a government proposal to encourage more private actions, including a proposal to permit double damages.19 Private antitrust enforcement in many Asian countries has grown in recent years. In Korea, 2004 amendments to its Monopoly Regulation and Fair Trade Act have led to an increase in private damages actions.20 These amendments, aimed at encouraging private enforcement, abolished the requirement that private damages actions follow final determinations by the Korean Fair Trade Commission, and alleviated the burden on plaintiffs for proving damages.21 Japan, which provides for both monetary and injunctive relief through private civil actions under its AntiMonopoly Act (‘AMA’), has also experienced growth in private enforcement.22 Though private actions seeking injunctive relief have a very low success rate,23 2009 amendments to the AMA enhancing discovery in injunction cases may change this.24 China, Asia’s largest economy, passed its Anti-Monopoly Law in 2007, which included

18

See Leonardo Maniqlia Duarte and others, ‘Brazil: Private Antitrust Enforcement’ in ‘The Antitrust Review of the Americas 2012’ (Global Competition Review, 2012) . 19 See Barbara Rosenberg and others, ‘Private Enforcement in Brazil: Not Quite There Yet’ (2012) 1 International Antitrust Bulletin 6, . 20 Article 57, Monopoly Regulation and Fair Trade Act, Law No 3320 (December 31, 1980, as amended by Law No 7315, December 31, 2004) (hereafter MRFTA) . see also Hwang Lee and Byung Geon Lee, ‘Korea’ in Foer and Cuneo (n 2). 21 Article 57 MRFTA (n 20); see Sang-Seung Yi and Youngjin Jung, ‘A New Kid on the Block: Korean Competition Law, Policy, and Economics’ (2007) 3 Competition Policy International 153, 178. 22 Ch VII, The Antimonopoly Act, Act No 54 (April 14, 1947, as amended) ; see also Simon Vande Walle, ‘Private Enforcement of Antitrust Law in Japan: An Empirical Analysis’ (2011) 8 Competition Law Review 7. 23 Vande Walle (n 22). 24 Act for Partial Revision of the Antimonopoly Act, Act No 51 (June 3, 2009, effective June 3, 2010); see Japan Fair Trade Commission, Summary of the Amendment to the Antimonopoly Act (June 2009) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 389

provisions permitting private parties to bring civil damages actions.25 With 61 actions already filed by the end of 2011, private enforcement activity is expected to expand even more due to a series of recent rules governing private antitrust litigation.26 These rules, which took effect in June 2012, shift the burden of proof to the defendants in many antitrust actions and ease evidentiary burdens, thereby providing a much more plaintiff-friendly environment.27 While many Asian countries experience growth in private enforcement, others falter. India, for example, continues to experience nearly nonexistent private antitrust enforcement due in large part to limitations in that country’s antitrust regime, which tie private actions closely to public enforcement.28 Thus, even though there are promising advancements, private antitrust enforcement in many parts of the world leaves significant room for development.

2. OBJECTIVES OF PRIVATE ENFORCEMENT 2.1 Deterrence and Compensation The US Supreme Court has repeatedly stated that the private right of action for treble damages under the antitrust laws serves two purposes: compensating injured victims of unlawful conduct and attracting enforcement resources to supplement the government’s deterrence-oriented efforts.29 Most antitrust experts probably agree that these are the two

25

‘Anti-Monopoly Law’ (promulgated by the Standing Committee National People’s Congress on August 30, 2007, and effective August 1, 2008) Art 50 . 26 ‘Press Release of the Supreme People’s Court’ (Supreme People’s Court Website, May 8, 2012) . 27 Ibid; see also John D Graubert and others, ‘E-Alert, Chinese Court Issues Rules Governing Private Antitrust Litigation’ (Covington and Burling LLP, May 15, 2012) . 28 See Pradeep S Mehta and Cornelius Dube, ‘India’ in Foer and Cuneo (n 2). 29 See for example, Pfizer, Inc. v Government of India, 434 US 308, 314 (1978); American Society of Mechanical Engineers, Inc. v Hydrolevel Corp., 456 US 556, 572 No 10 (1982).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 6 /

Date: 13/10

JOBNAME: Duns PAGE: 7 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

390 Comparative competition law

appropriate goals.30 Unfortunately, it is often far from clear that private antitrust litigation serves either of these goals very well. 2.1.1 Compensation Antitrust injuries are conventionally classified into two categories: static and dynamic. Static injuries concern the effects on consumers from an increase in price, attendant to the exercise of monopoly power.31 Dynamic injuries arise when a dominant firm stifles innovation or new product development.32 Private antitrust litigation could, in theory, attempt to compensate injured victims for both types of losses, but it fails at both tasks. In a static model of competitive injury, the clearest social cost of an antitrust violation falls upon a class of people who will often not be able to sue – those who stopped purchasing from the defendant because the price increased too much. When a firm acquires market power through collusion or exclusion of rivals, it typically will raise its price to a monopoly level. Two classes of consumers suffer harm. First, some consumers who would otherwise have purchased the product are no longer willing to purchase it at the higher price. Hence, they substitute to a second-best preference. These consumers’ loss is a ‘deadweight’ loss – a class of foregone transactions that should, in a competitive market, have taken place.33 Economists identify deadweight loss as the primary social cost of anti-competitive behavior.34 It is a social cost because it results in a misallocation of scarce social resources. Customers must buy their second-best, rather than their first-best, preferences. The second class of harm accrues to consumers who continue to purchase the product at the higher price. Their injury is a wealth transfer from consumers to producers.35 This is not necessarily economically inefficient, since it merely transfers money from one person to another 30 See, for example, Phillip Areeda, ‘Antitrust Violations Without Damage Recoveries’ (1976) 89 Harvard Law Review 1127, 1127. 31 See generally Herbert Hovenkamp, Federal Antitrust Policy: The Law of Competition and Its Practice (3rd edn, Hornbook 2005) 19–20. 32 See generally Herbert Hovenkamp, ‘Restraints on Innovation’ (2007) 29 Cardozo Law Review 247. 33 Herbert Hovenkamp, ‘Antitrust’s Protected Classes’ (1989) 88 Michigan Law Review 1, 30. 34 Kenneth G Elzinga and William Breit, The Antitrust Penalties: A Study in Law and Economics (Yale University Press 1976) 6–7; William M Landes, ‘Optimal Sanctions for Antitrust Violations’ (1983) 50 University of Chicago Law Review 652, 653. 35 Hovenkamp (n 33) 9–11.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 391

and does not directly affect the deployment of scarce resources.36 There is nothing that says a priori that the money is better off in the hands of consumers than the producers. For example, if the price-fixed product is a luxury good, then price-fixing may act as a progressive tax on rich consumers for the benefit of poor artisans. Private parties who sue antitrust defendants typically do not sue to vindicate the interests of the consumers who stopped buying the goods because they were too expensive. Instead, private parties sue only for the purchasers who did buy and actually incurred an overcharge.37 For example, imagine the difficulty of representing a class of purchasers who stopped buying vitamins because of a vitamins cartel or, even worse, who never started buying vitamins because of excessive prices. How could one prove who the consumers were or quantify their injury? It is much easier to recover damages on behalf of purchasers who made the purchase but paid too much. In this case, the formula is simply to posit a but-for price and use the difference between the actual price and the but-for price as damages. Assume, however, that compensating for wealth transfer is every bit as important a goal of antitrust as preventing deadweight losses. How well does private antitrust enforcement do on this score? As to consumers, the answer is ‘not very well.’ The widely distributed nature of the overcharge and the difficulty in locating the real economic victims work against this goal. Suppose that the supplier of a part used to manufacture large trucks enters into a price-fixing conspiracy. It charges a monopoly price to the truck manufacturers, who then pass on a portion of that overcharge to their dealers, who pass on a portion of that overcharge to trucking companies, who pass on a portion to shippers they service, who pass on a portion to their customers, and so on down the line. A monopoly overcharge often produces endless ripples in the economy.38 The law can give a practical answer to the above scenario: just as a person can be damaged without legal injury, so also can a person be legally injured without actual damage. Under the current US system of 36 To the extent that would-be monopolists expend resources trying to effectuate wealth transfers, such expenditures may be inefficient: Richard A Posner, Antitrust Law (2nd edn, University of Chicago Press 2001). However, private enforcers do not sue to recover such wasted expenditures. 37 See Hovenkamp (n 33) 36. 38 See Robert G Harris and Lawrence A Sullivan, ‘Passing on the Monopoly Overcharge: A Comprehensive Policy Analysis’ (1979) 128 University of Pennsylvania Law Review 269, 276: noting that ‘in a multiple-level chain of distribution, passing on monopoly overcharges is not the exception: it is the rule’.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

392 Comparative competition law

standing, a private plaintiff who is a direct purchaser from an overcharging monopolist has a right of action to recover the full extent of the overcharge, even if the plaintiff passes on the entire overcharge downstream and therefore suffers no economic damage.39 Conversely, an indirect purchaser bearing the real economic brunt may recover nothing.40 Figuring out which, if any, of these groups suffered the real economic harm is often difficult or impossible given the complexity and variability of the relevant economic relationships. This brings us to the second problem with pursuing a compensation goal. Even if the real economic victims of an overcharge as a class could be identified, identifying the actual people who suffered injury and issuing them a check is often so expensive that administrative costs swallow the whole recovery.41 The US experience on this is instructive. Realizing that the costs of issuing checks to the members of the injured class were often prohibitive, creative lawyers invented other ways in which the injured consumers could be compensated. For example, the settling lawyers would have the defendants agree to issue coupons to class members. These coupons could be used for future purchases from the defendants. The coupons often go unclaimed and unused, however.42 Further, the issuance of coupons may simply encourage the defendants to raise the price of their products, thus wiping away any compensatory benefit, or it may cause inefficient overconsumption by the compensated class.43 There exists one class of interests to which the foregoing observations do not apply with equal force. When competitors are injured by an exclusionary practice, they are not overcharged. Instead, they lose profits. But there are several problems with justifying a compensation-oriented 39

Hanover Shoe, Inc. v United Shoe Mach. Corp., 392 US 481 (1968). Illinois Brick Co. v Illinois, 431 US 720 (1977). It bears noting that the Illinois Brick and Hanover Shoe rules only apply in federal lawsuits. In California v ARC America Corp., 490 US 93 (1989) the Supreme Court held that federal antitrust law does not preempt state antitrust laws that allow indirect purchaser suits. A number of states have allowed such indirect purchaser suits. 41 See Richard A Posner, ‘Oligopoly and the Antitrust Laws: A Suggested Approach’ (1969) 21 Stanford Law Review 1562, 1590. 42 See Christopher R Leslie, ‘The Need to Study Coupon Settlements in Class Action Litigation’ (2005) 18 Georgetown Journal of Legal Ethics 1395, 1396–97: reporting redemption rate of 26.3 percent. 43 See A Mitchell Polinsky and Daniel L Rubinfeld, ‘A Damage-Revelation Model of Coupon Remedies’ (2007) 23 Journal of Law, Economics & Organization 653, 653. 40

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 393

private remedy on the grounds that competitors are good candidates for compensation. First, there is a widespread and probably legitimate concern that competitor plaintiffs frequently use antitrust law for anti-competitive purposes.44 If anything, private competitor lawsuits make it harder for consumers to receive full compensation, since courts often react to competitor lawsuits by contracting liability norms in a way that stymies consumer suits as well. Second, competitor harm – or at least the kind of harm recognized by antitrust law – does not occur in many antitrust cases. If a cartel raises prices, a firm that does not join the cartel nonetheless benefits from it. If an anti-competitive merger harms a rival firm because in the but-for world it would have faced less competition, the rival firm has not suffered ‘antitrust injury’ and may not recover damages. On the other hand, it is also possible that antitrust violations harm competitors without harming consumers, as is true in cases of attempted monopolization, where a dominant firm harms its rival but fails to obtain monopoly power. But in those cases, the ordinary justification for allowing the competitor suit is a prophylactic one: namely, that it is better to prevent antitrust violations before they occur. This is not a compensation justification; rather, it is a deterrence argument in which the rival’s lawsuit is merely a means to achieve the end of consumer welfare. Third, it is awkward to justify a compensation goal with reference to competitor suits. In the US and the EU at least, it is generally recognized that antitrust law exists to protect consumers, not competitors. Competitors are only instrumental to the goal of consumer welfare. Turning now to dynamic injuries, economists and antitrust scholars increasingly view static consumer injuries as far less significant than dynamic injuries.45 The major dispute concerning competition policy and innovation is not whether innovation creates more consumer welfare than 44 See William J Baumol and Janusz A Ordover, ‘Use of Antitrust to Subvert Competition’ (1985) 28 Journal of Law and Economics 247; Frank H Easterbrook, ‘The Limits of Antitrust’ (1984) 63 Texas Law Review 1; Daniel A Crane, ‘The Paradox of Predatory Pricing’ (2005) 91 Cornell Law Review 1, 5–32; R Preston McAfee and Nicholas V Vakkur, ‘The Strategic Abuse of the Antitrust Laws’ (2005) 2 Journal of Strategic Management and Education 37, 37–38; Edward A Snyder and Thomas E Kauper, ‘Misuse of the Antitrust Laws: The Competitor Plaintiff’ (1991) 90 Michigan Law Review 551. 45 See Hovenkamp (n 32) 253; Michael E Porter, ‘Competition and Antitrust: A Productivity-Based Approach’ in Charles D Weller (ed), Unique Value: Competition Based on Innovation: Creating Unique Value for Antitrust, the Economy, Education and Beyond (Innovation Press 2004) 154, 156–57.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

394 Comparative competition law

static efficiency but whether antitrust enforcement can meaningfully advance dynamic efficiency. For those who follow the tradition of Joseph Schumpeter, all significant competition is dynamic, and static efficiency is of dubious importance.46 Because the path of innovation is unpredictable, some economists in the Schumpeterian tradition doubt whether antitrust law does much to advance consumer welfare.47 On the other side, those who follow the tradition of Kenneth Arrow contend that antitrust law is critical to fostering consumer welfare, since oligopolistic markets tend to foster lower rates of innovation than more competitive ones.48 Still, whether one follows Schumpeter or Arrow, there is general agreement that innovation and technological progress are vital to the advancement of consumer welfare, and that they are probably far more significant than short-term pricing and output effects.49 Private antitrust litigation is no better at compensating consumers for dynamic injuries than for static injuries. Indeed, it is probably worse. With respect to static injuries, it is possible for economists to estimate the amount of the overcharge, though the estimate may sometimes be speculative. In contrast, measuring the amount of dynamic efficiency loss and quantifying this loss in dollars often amounts to little more than guesswork. Dynamic efficiency losses suffer from an incommensurability problem. For example, suppose that Microsoft undertakes a campaign to quash the development of a multiplatform Java programming system in order to prevent Java from developing a program that could compete with Microsoft’s Windows operating system. As a result, Microsoft retains its operating system dominance and overcharges consumers for Windows. The overcharge portion of the static harm, although not the deadweight loss, can be quantified and reduced to a damages award in litigation. But the much larger harm may be the loss of an unknown amount and variety 46 Joseph Schumpeter, Capitalism, Socialism, and Democracy (Harper and Brothers 1942) 84–85. 47 See Harold Demsetz, ‘How Many Cheers for Antitrust’s 100 Years?’ (1992) 30 Economic Inquiry 207; Michael L Katz and Howard A Shelanski, ‘Mergers and Innovation’ (2007) 74 Antitrust Law Journal 1, 3. 48 Kenneth J Arrow, ‘Economic Welfare and the Allocation of Resources for Invention’, in Richard Nelson (ed), The Rate and Direction of Inventive Activity: Economic and Social Factors (National Bureau of Economic Research 1962) 619–22. See generally Jonathan B Baker, ‘Beyond Schumpeter vs. Arrow: How Antitrust Fosters Innovation’ (2007) 74 Antitrust Law Journal 575. 49 See, for example, Thomas O Barnett, ‘Maximizing Welfare Through Technological Innovation’ (2007–2008) 15 George Mason Law Review 1191, 1194.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 395

of technological innovation that would have followed had Microsoft not stymied Java programmers in creating a competitive alternative. Since operating systems function as a platform to other applications, technological innovations in the operating systems field could have led to untold innovations in applications. Unsurprisingly, very few antitrust plaintiffs seek compensation for dynamic injuries. Even when they do seek such compensation, courts are inhospitable to their claims. Proving the but-for world of competitive innovation is far too speculative an endeavor to meet the evidentiary burden required by civil litigation. Thus, courts have rejected claims against Microsoft for stifled innovation. For example, the US Court of Appeals for the Fourth Circuit held that: It would be entirely speculative and beyond the competence of a judicial proceeding to create in hindsight a technological universe that never came into existence. … It would be even more speculative to determine the relevant benefits and detriments that non-Microsoft products would have brought to the market and the relative monetary value … to a diffuse population of end users.50

As the Fourth Circuit observed, the problems of proof were not limited to the specific facts of the Microsoft case: ‘At bottom, the harms that the plaintiffs have alleged with respect to the loss of competitive technologies are so diffuse that they could not possibly be adequately measured. The problem is not one of discovery and specific evidence, but of the nature of the injury claimed.’51 Damages for these sorts of harms are almost never recovered in private litigation. 2.1.2 Deterrence Those who believe that antitrust law exists primarily to promote economic efficiency argue that deterrence, and not compensation, should be the primary purpose of private antitrust enforcement.52 In their view, antitrust damages supplement criminal fines by removing any expected profitability from antitrust violations.53 Decision makers in dominant 50

Kloth v Microsoft, 444 F 3d 312, 324 (4th Cir. 2006) quoting In re Microsoft, 127 F. Supp. 2d 702, 711 (2001). 51 Ibid. 52 See, for example, William H Page, ‘The Scope of Liability for Antitrust Violations’ (1985) 37 Stanford Law Review 145, 145; Posner (n 36) 266. 53 See generally Gary S Becker, ‘Crime and Punishment: An Economic Approach’ (1968) 76 Journal of Political Economy 169. As Bill Landes has explained, ‘[t]he optimal penalty should equal the net harm to persons other than

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 12 /

Date: 8/9

JOBNAME: Duns PAGE: 13 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

396 Comparative competition law

firms will perceive that they are better off by not violating antitrust law, given the likelihood of detection multiplied by the penalty. Unlike other fields of law in which the deterrent value of private enforcement has been questioned,54 the deterrent success of private antitrust enforcement is largely taken for granted.55 In order to evaluate this deterrence claim, one must ascertain who is being deterred. The primary class of relevant decision makers is corporate managers, although one must also consider the possibility that vigilant shareholders will rein in managers who fail to respond to antitrust incentives. With respect to these managers, the argument that private antitrust litigation provides effective deterrence is increasingly doubtful. Two converging trends – the increasing length of antitrust proceedings and the increasing shortness of managerial tenure – make it likely that corporate managers severely discount the threat of future damages litigation. First, the time gap between the planning of antitrust violation (which is presumably the moment at which deterrence should take root) and antitrust judgment day is increasingly long. The average private antitrust case takes many years, usually over five and often approaching ten, from the date of the challenged acts to the date of a civil judgment in favor of the plaintiff. Then there are appeals, which may take several years to run their course. Two prominent monopolization cases in the US are instructive. In LePage’s Inc. v 3M,56 the allegedly anti-competitive bundled rebates were put in place in 1992, and LePage’s filed suit in 1997 but did not prevail until 2004, when the Supreme Court denied certiorari. In Conwood Co. v US Tobacco Co.,57 the plan to eliminate Conwood was hatched in 1990, Conwood sued in 1998, and the Supreme Court denied certiorari in 2003. the offender, adjusted upward if the probability of apprehension and conviction is less than one’: Landes (n 34) 678. 54 See, for example, John C Coffee, Jr, ‘Reforming the Securities Class Action: An Essay on Deterrence and Its Implementation’ (2006) 106 Columbia Law Review 1534, 1535–36: ‘As presently constituted, securities class actions produce wealth transfers among shareholders that neither compensate nor deter.’ 55 Some commentators have questioned whether treble damages are adequate to achieve optimal deterrence. See, for example, Robert H Lande, ‘Five Myths About Antitrust Damages’ (2005–2006) 40 University of San Francisco Law Review 651. However, the general view is that the prospect of antitrust damages is a successfully conveyed message to the firm, even if the exact size of the multiplier should be higher. 56 LePage’s Inc. v 3M, 324 F.3d 141, 144 (3d Cir 2003). 57 Conwood Co. v US Tobacco Co., 290 F.3d 768, 772, 775 (6th Cir 2002).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 397

This time lag should be paired with the fact that the managers who put into place anti-competitive schemes are increasingly unlikely to be around to internalize their effects at judgment day. During the 1980s, the turnover rate among senior managers in large corporations was just above 10 percent.58 By all accounts, the turnover rate has increased significantly, and perhaps even doubled, in the 1990s and 2000s, as various capital market factors accentuated shareholder demand for short-term performance.59 Today, the average CEO in the US holds her job for about six years.60 Mid-level executives, such as divisional managers, typically hold their jobs for an even shorter period, perhaps less than four years.61 Thus, most of the executives responsible for an antitrust violation will no longer be with the firm by the time a damages award is entered against the company. High managerial turnover rates might not thwart the deterrence objective if managers were to internalize some of the detrimental effects of antitrust judgments rendered after they leave the defendant firm. In particular, managers might incur a reputational cost in lost future employment opportunities or take a prestige hit in the business community by virtue of their past roles in a later-adjudicated antitrust

58

See Anne T Coughlan and Ronald M Schmidt, ‘Executive Compensation, Management Turnover, and Firm Performance: An Empirical Investigation’ (1985) 7 Journal of Accounting & Economics 43, 52 tbl.1: documenting an average CEO turnover of 12.7 percent for all firms listed in Forbes compensation survey from 1978 to 1980; Jerold B Warner, Ross L Watts and Karen H Wruck, ‘Stock Prices and Top Management Changes’ (1988) 20 Journal of Financial Economics 461, 461 tbl. 1: calculating average annualized rate of 11.5 percent for arrivals or departures by top management including CEO, chairman, or president positions based on Wall Street Journal reports from 1963 to 1978. 59 Kathleen A Farrell and David A Whidbee, ‘Impact of Firm Performance Expectations on CEO Turnover and Replacement Decisions’ (2003) 36 Journal of Accounting & Economics 165, 173 tbl.1: recording a 9 percent turnover rate from 1986 to 1997. Steven N Kaplan and Bernadette A Minton, ‘How Has CEO Turnover Changed? Increasingly Performance Sensitive Boards and Increasingly Uneasy CEOs’ (2006) National Bureau of Economic Research, Working Paper No 12465, 2: reporting that in the period 1998–2005, CEO turnover increased to 16.5 percent, implying an average CEO tenure of just over six years. 60 Kaplan and Minton (n 59) 2. 61 RM Bushman, RJ Indjejikian and A Smith, ‘Compensation Contracts in Hierarchical Organizations: Determinants of Incentive Pay for Business Unit Managers’ (1994) University of Chicago Working Paper: reporting 3.7 year average tenure for division managers.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

398 Comparative competition law

violation.62 But there is scant evidence suggesting that individual managers’ reputations are much affected by antitrust judgments against their former employers. Individual managers are not often named as co-defendants in private antitrust cases and usually do not appear in any public pronouncement of liability. Liability in complex antitrust cases seldom turns on the culpability of a single manager but rather on a cluster of managerial decisions over time, making it difficult to pinpoint blame. In light of these facts, it is difficult to see how the threat of a future damages judgment disciplines managerial decision making. When managers plan conduct that brings immediate large profits but only potential liability at some future date, the extent to which the future liability deters them from choosing immediate profits is a function of their implicit discount rate for the potential damages award. The longer the perceived time until judgment day, the more likely it is that managers will discount the threat of damages. If managers believe that they are unlikely to be employed by the firm at the distant judgment day, they will tend to disregard the threat of future liability altogether. Yet, the story is not so simple. One must consider the possibility that private antitrust litigation forces managers and shareholders to internalize the costs of an antitrust violation long before any legal judgment day. Efficient capital markets may punish the firm and, by extension, its managers, as soon as a lawsuit is threatened or filed. According to one study, the filing of a securities class action lawsuit by investors more than doubled the likelihood of CEO turnover.63 It is unlikely, however, that the mere filing of a private antitrust suit brings about severe consequences for a defendant firm’s managers. Unlike securities lawsuits, which may impair shareholder confidence in the integrity of insiders, antitrust lawsuits may instead communicate that company managers are acting aggressively to expand market share and increase profitability. This is especially true in the case of private antitrust lawsuits, which many shareholders may rightly regard as signaling that rival firms are merely disaffected by aggressive, but ultimately lawful, competition. While empirical work suggests that the filing of an antitrust action by the Department of Justice or Federal Trade Commission has an immediate 62

See Christopher R Leslie, ‘Antitrust Amnesty, Game Theory, and Cartel Stability’ (2006) 31 Journal of Corporation Law 453, 459–60. 63 See Philip E Strahan, ‘Securities Class Actions, Corporate Governance and Managerial Agency Problems’ (SSRN, June 1998) 22 http://papers.ssrn.com/ absract=104356 reporting that filing of a securities class action lawsuit by investors increased probability of CEO turnover from 9.8 percent to 23.4 percent.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 15 /

Date: 13/10

JOBNAME: Duns PAGE: 16 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 399

and significant negative effect on a defendant firm’s share price, the filing of a private antitrust lawsuit has only about a tenth of the effect of a public suit. Empirical studies have found that defendants lost, on average, 6 percent of their share value upon the filing of a government antitrust lawsuit64 but only about 0.6 percent of their share value – with an average loss of $4 million – upon the filing of a private lawsuit.65 A half-percent drop in market capitalization is unlikely to engender ruinous consequences to most managers, particularly if the gains from the challenged behavior were large. If managers cannot be expected to respond reliably to the threat of distant and unpredictable liability judgments against their firms, what about shareholders? Two related factors suggest that shareholders do not have strong incentives or capabilities, either. First, the kind of industrial behavior that gives rise to antitrust claims creates the prospect of substantial long-run costs if an antitrust challenge is successful but offers significant short-run profits in the meantime. From an outsider’s perspective, it is very hard to evaluate the risk that firm behavior will produce eventual antitrust liability. Whether conduct of the kind adjudged under antitrust’s rule of reason and monopolization law’s amorphous standards is likely to result in antitrust liability is difficult for antitrust experts to predict, let alone the average institutional investor. It is unlikely that many shareholders will try to curb behavior that increases a firm’s market share and profits but that might eventually lead to antitrust liability. Even assuming that a large institutional investor with access to corporate insiders was aware that conduct was risky from an antitrust perspective, given the expected length of time until judgment day, it will often be a rational strategy to approve tacitly the risky conduct, hold the stock for a few years while its value increases, and then sell before an antitrust suit commences. The initial filing of the private lawsuit does relatively little damage to the issuer’s share price, so savvy shareholders will have plenty of time to reap their profits and then exit. When we stop treating the corporation as a black box and instead introduce realistic assumptions about managerial incentives, shareholder monitoring difficulties, and other substantial agency costs, and then pair those concerns with managerial turnover rates and the lengthy duration of 64

Kenneth D Garbade and others, ‘Market Reaction to the Filing of Antitrust Suits: An Aggregate and Cross-Sectional Analysis’ (1982) 64 Review of Economics & Statistics 686, 686–71. 65 John M Bizjak and Jeffrey L Coles, ‘The Effect of Private Antitrust Litigation on the Stock-Market Valuation of the Firm’ (1995) 85 Journal of American Economic Review 436, 437.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

400 Comparative competition law

private antitrust litigation, the assumption that private antitrust litigation provides significant deterrence needs to be subjected to more careful scrutiny. Of course, conditions may be different in other jurisdictions or circumstances. In the case of closely held companies with shareholder managers, litigation systems where the time lapse between anticompetitive conduct and legal judgment day is much shorter, or countries or industries where managers enjoy much longer tenure, the observations may be different. These are questions that jurisdictions seeking to add deterrence through private antitrust litigation should ask. 2.2 Consequences of the Prioritization of Compensation or Deterrence In the previous section, we presented some reasons for skepticism that either compensation or deterrence is necessarily furthered in most private antitrust litigation. Be that as it may, most antitrust regimes that adopt private enforcement will point to deterrence, compensation, or both as primary objectives of private enforcement. Although many people involved in setting up new antitrust regimes may not think it terribly important to decide whether compensation or deterrence comes first, the choice is actually critically important to many decisions about structuring a private enforcement system. The ongoing European experience is instructive. As noted above, in 2008, the European Commission released a White Paper – essentially a blueprint for further action by the Commission – calling for expanded private antitrust enforcement within the European Union.66 Two dominant themes emerge from the White Paper: (1) Europe needs more private enforcement but (2) not in the US style. On virtually every major aspect of private litigation – including standing, discovery, claim aggregation, damages – the White Paper argues for an approach that essentially rejects the US position. A glimpse back three years before the issuance of the White Paper will illustrate the origin and nature of the White Paper’s orientation. In a 2005 Green Paper – a kind of publicly disseminated draft of the White Paper – the Commission had taken the view that deterrence should be a primary goal of an expanded private enforcement regime.67 But between 2005 and 66 European Commission ‘European Communities White Paper on Damages Actions for Breach of the EC Antitrust Rules’ (COM (2008) 165 final, European Commission April 2, 2008) (hereafter ‘EU White Paper’). 67 European Commission ‘Damages Actions for Breach of the EC Antitrust Rules: Green Paper’ (European Commission, December 19, 2005) : asserting that private enforcement should ‘contribut[e] significantly to the maintenance of effective competition in the Community (deterrence).’ 68 Ibid 3. 69 Ibid 4. 70 Ibid 8. 71 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 18 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

402 Comparative competition law

on the overcharge.72 Conversely, the ICN study concluded that the vast majority of surveyed jurisdictions would probably allow suits by indirect purchasers.73 To the extent that jurisdictions favor a compensation goal, they will presumably opt for indirect purchaser standing and the availability of a pass-on defense. The effects of a compensation objective do not end with standing principles. The White Paper calls for ‘full compensation of the real value of the loss suffered.’74 The White Paper goes on to make clear that this includes not only ‘actual loss’ but also lost profits and interest.75 While this all sounds generous to plaintiffs, there is a highly significant subtext: treble damages or other damages multipliers are inconsistent with a compensatory regime, since they award plaintiffs a windfall far in excess of their actual damages. Two more procedural aspects of compensation-oriented regimes are significant. First, class action procedures are rare in continental systems. The White Paper recognizes that the parties bearing the economic burden of antitrust violations have often ‘suffered scattered and relatively lowvalue damages,’ and that some form of claim aggregation is therefore necessary for private enforcement.76 Still, the White Paper shies away from US-style class actions, calling instead for ‘representative actions’ by ‘qualified entities’ such as state bodies or trade associations or ‘opt-in collective actions.’77 The ICN reports that class actions are available in only five of the jurisdictions it surveyed.78 Another procedural issue raised by the White Paper and commonly discussed in jurisdictions considering private antitrust enforcement is discovery. Here, again, the White Paper rejects the US model. Although the White Paper recognizes the need for some information disclosure in litigation, it seems far more concerned about ‘the negative effects of overly broad and burdensome disclosure obligations, including the risk of abuses.’79 It proposes that any discovery should be court ordered (rather than party initiated as in the US) and only available when the plaintiff 72 Cartels Working Group ‘Report: Interaction of Public and Private Enforcement in Cartel Case’ (International Competition Network Annual Conference, Moscow, May 2007) 12. 73 Ibid. 74 EU White Paper (n 66) 7. 75 Ibid. 76 Ibid 4. 77 Ibid. 78 Cartels Working Group (n 72) 4. 79 EU White Paper (n 66) 5.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 403

has pleaded facts plausibly showing the existence of an antitrust violation; satisfied the court that he cannot reasonably obtain the facts except through discovery; specified the precise categories of information to be disclosed; and demonstrated that the discovery requests are relevant, necessary, and proportionate. In sum, the proposed EU system, which is also characteristic of other emerging antitrust jurisdictions, has a compensation objective and, consequently, indirect purchaser standing, single damages, and restrictive rules on claim aggregation and discovery. By contrast, the US Supreme Court has nominally recognized deterrence and compensation as co-equal objectives but has seemed to favor deterrence over compensation when the two interests diverged.80 As previously noted, the US generally follows a ‘direct purchaser’ rule, in which the direct purchasers of the overpriced goods have standing to sue, even though they may pass on the overcharge by reselling the good at a marked-up price and thereby avoid economic injury and deny standing to indirect purchasers. The US also allows automatic treble damages by statute and liberal party-initiated discovery by rule. Consider now how effective the European model is likely to be at achieving compensation or in encouraging private antitrust litigation at all. If European courts chase the harm downstream to the ultimately injured party, they will find themselves with thousands or even millions of parties to compensate in the ordinary antitrust case. Not only will those parties be widely dispersed with many small injuries, they will also be denied the two features of the US system, generous claim aggregation and treble damages, that provide hope that a suit will be brought and monies recovered. Further, if the purchasers sue, they will face sharp limitations on discovery. Discovery stinginess is particularly problematic in a system that gives standing to downstream purchasers who are remote from the defendant. Unlike the direct purchasers, who are often sophisticated contractual parties of the defendant, the indirect purchasers may know very little about the defendant’s business and need generous amounts of discovery to make their case. Ironically, a primarily deterrence-oriented system, like that of the US, has less need of the very features of the US civil litigation system that continental systems often reject. It has less need for claim aggregation, since the direct purchasers are relatively fewer in number and have more 80

See Barak D Richman and Christopher R Murray, ‘Rebuilding Illinois Brick: A Functionalist Approach to the Indirect Purchaser Rule’ (2007) 81 Southern California Law Review 69, 90.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 20 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

404 Comparative competition law

concentrated injury. It has less need for treble damages, since the direct purchasers often have a sufficient incentive to sue for even single damages. Finally, it has less need for discovery, since the direct purchasers often know much more about the defendant’s business practices and conduct than do the downstream purchasers.

3. CONSEQUENCES OF PRIVATE ENFORCEMENT Proponents of private enforcement often argue that putting more ‘boots on the ground’ strengthens antitrust enforcement. We have already shared some skepticism that private antitrust enforcement increases deterrence. Beyond merely failing to achieve its deterrence objectives, private antitrust enforcement can trigger a judicial backlash against antitrust enforcement altogether, as the US experience shows. This backlash can spill over to public enforcement and impair antitrust enforcement by the public agencies. At the same time, however, expanding the private right of action can expand the constituency in favor of antitrust law and hence bolster the overall antitrust enterprise. 3.1 Judicial Backlash We have already noted that private antitrust enforcement boomed in the US in the post-World War II era and then contracted beginning in the late 1970s. One of the major explanations for the contraction is that judges began to perceive private enforcement as out of control and counterproductive and reacted by imposing new restrictions, substantive and procedural, on antitrust cases. One empirical study in the mid-1980s shed light on the realities of private antitrust enforcement. The Georgetown Private Antitrust Litigation Project studied 2,350 private antitrust cases filed in federal court in five judicial districts between 1973 and 1983.81 The survey results provided some interesting insights into what kinds of cases were being filed and who was filing them. For one, complaints of vertical misconduct outnumbered horizontal complaints.82 Second, most private cases were brought by businesses vertically related to the defendant, such as dealers, licensees, and franchisees or by competitors.83 Competitors brought 35.5 percent of all cases, dealers 27.3 percent, and suppliers 5.6 81 82 83

Columns Design XML Ltd

/

Salop and White (n 3) 1021. Ibid 2005. Ibid.

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 405

percent and franchisees and licensees cumulatively around 3 percent.84 Customers – the ostensible beneficiaries of antitrust law – played a comparatively minor role. Final or end customers brought 8.7 percent of cases.85 But customer companies filed only 12.5 percent,86 hence all customer complaints accounted for only about a fifth of antitrust filings. Third, around 88 percent of the cases settled; only about 5 percent of cases went to trial.87 Fourth, judges apparently used dismissal sparingly, dismissing fewer than 25 percent of all cases.88 Fifth, the average settlement was either $547,000 or $1,244,000, depending on whether one chose to include a $50 million outlier settlement.89 Either way, the numbers seemed significant. Finally, the combined cost (in 1984 dollars) of litigating the cases was estimated to be around $200,000–$250,000.90 One could draw many different lessons from the Georgetown study, but one set of conclusions could easily feed into the rising ‘Chicago School’ critique that was beginning to prevail in the courts.91 Private antitrust litigation was fueled primarily by disgruntled competitors, dealers, franchisees, and licensees whose interests were adverse to those of consumers. The bulk of private cases challenged vertical practices that the Chicago School considered likely to be benign, rather than horizontal practices that the Chicago School might have conceded could be more problematic. The cases were easy to bring, costly to defend, unlikely to be involuntarily dismissed by the courts, and hence likely to coerce the defendant to pay a substantial settlement to be rid of the case. The Georgetown study generated a large amount of scholarship on private litigation, much of it critical. An important theme that emerged from the study was that much private antitrust litigation was motivated by businesses seeking a strategic advantage in the marketplace rather than 84

Ibid 1008, Table 5. Illinois Brick Co. v Illinois, 431 US 720 (1977). 86 Salop and White (n 3) 1008, Table 5. 87 Ibid 1011. 88 Ibid. 89 Ibid 1012. 90 Ibid 1015–16. 91 The Chicago School is identified with scholars at the University of Chicago who began to make a systematic critique of interventionist antitrust law during the 1950s and 1960s. The Chicago School was ultimately successful in contributing to a dramatic transformation of antitrust law in the United States in the 1970s and 1980s: Daniel A Crane, ‘Chicago, Post-Chicago, and NeoChicago’ (2009) 64 University of Chicago Law Review 1911, reviewing Robert Pitofsky (ed), How the Chicago School Overshot the Mark (Oxford University Press 2008). 85

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

406 Comparative competition law

by publicly minded private attorneys general. Kenneth Elzinga and William Wood’s work with the Georgetown data led them to the conclusion that ‘only a minority of the cases in the data set, and presumably in private antitrust cases generally, have any connection with improving or maintaining efficiency.’92 Elzinga and Wood concluded that many private antitrust lawsuits were strategically motivated. Edward Snyder and Tom Kauper similarly analyzed the Georgetown data and found evidence that the private antitrust remedy was often used to subvert, rather than promote, competition.93 Although an important source of information and fount of valuable scholarship on private antitrust litigation, the Georgetown study was not the triggering event in judicial retrenchment of private antitrust litigation. For example, the idea that plaintiffs strategically misused antitrust law to subvert wholesome competition had been proposed in a number of books and articles that preceded the Georgetown study.94 Rather, the Georgetown study provided empirical evidence of intuitions or suspicions that lots of lawyers, economists, and judges had begun to hold about private antitrust enforcement since its boom in the post-war era. By the time the Georgetown study was released in the mid-1980s, the courts had already been cutting back on private antitrust cases for many of the reasons that Georgetown subsequently made explicit. Many of the ‘backlash’ cases that contributed to the fall in private antitrust litigation rates during the late 1970s and 1980s involved questions unique to private litigation – particularly standing and the use of procedural mechanisms to screen out unmeritorious cases. But, during the same time frame as the courts were cutting back private enforcement specifically, they were also cutting back on the substantive norms of illegality in antitrust cases, most of which happened to be private cases. The courts sharply constricted liability norms for all intra-brand vertical

92 Kenneth G Elzinga and William C Wood, ‘The Costs of the Legal System in Private Antitrust Enforcement’ in Lawrence J White (ed), Private Antitrust Litigation: New Evidence, New Learning (Cambridge University Press 1988). 93 Snyder and Kauper (n 44) 44. 94 See, for example, Richard A Posner, Antitrust Law: An Economic Perspective (University of Chicago Press 1976) 228; Frank H Easterbrook, ‘The Limits of Antitrust’ (1984) 63 Texas Law Review 1, 37; William J Baumol and Janusz A Ordover, ‘Use of Antitrust to Subvert Competition’ (1985) 28 Journal of Law & Economics 247, 256–59.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 24 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 407

restraints,95 predatory pricing,96 refusals to deal,97 tying,98 and boycotts.99 Although these decisions were mostly made in private cases, they had significant effects on public enforcement as well. 3.2 Spillovers to Public Enforcement It is difficult to contain judicial backlash against private enforcement at the boundaries of private litigation. Once the courts make it harder for private antitrust plaintiffs to win by establishing exacting liability norms, how can they apply less restrictive liability determinants in public cases? In the US experience, as the courts created increasingly restrictive substantive rules in private lawsuits, the government increasingly found itself stuck with those precedents as well. Consider two examples. First, predatory pricing. During the 1980s and 1990s, in a series of private cases, the federal courts sharply constricted the right of action for predatory pricing.100 The courts justified restrictive predation liability rules by claiming that opportunistic private plaintiffs could chill rivals’ aggressive pricing by bringing predatory pricing lawsuits for treble damages.101 During the years that the courts were developing these restrictive liability norms, neither the Federal Trade Commission (FTC) nor the Department of Justice brought any predatory pricing lawsuits. The liability rules were created with the institutional limitations of private litigation in mind. Then, in 1999, the Justice Department brought its first predatory pricing lawsuit in decades, against American Airlines.102 The government lost the case on summary judgment in the district court and then again in the court of appeals, largely because the courts applied off-the-rack predatory pricing liability rules designed to avoid abusive private litigation. If the law of predatory pricing had developed with the institutional parameters of public enforcement in mind, it is doubtful that the resulting liability rules would have been so deferential to pricing decisions by dominant firms. 95 96

Leegin Creative Leather Prods., Inc. v PSKS, Inc., 551 US 877 (2007). Brooke Group Ltd. v Brown & Williamson Tobacco Corp., 509 US 209

(1993). 97

Verizon Comms., Inc. v Law Offices of Curtis V Trinko, LLP, 540 US 398

(2004). 98

Jefferson Parish Hospital No 2 v Hyde, 466 US 2 (1984). NYNEX Corp. v Discon, Inc., 525 US 128 (1998). 100 See generally Daniel A Crane, ‘The Paradox of Predatory Pricing’ (2005) 91 Cornell Law Review 1, 3. 101 Ibid. 102 US v AMR Corp., 335 F 3d 1109 (10th Cir 2003). 99

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 24 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

408 Comparative competition law

A second example comes from the controversial issue of patent litigation settlements between branded drug companies and generic challengers. In these so-called ‘reverse settlement’ or ‘pay for delay’ settlements, the branded firm has a patent on a pioneer drug and sues the generic for patent infringement. Before a court has resolved the case, the branded firm then settles with the generic by having the generic agree not to market its drug for some period of time in return for a payment from the branded firm – which is sometimes in the hundreds of millions of dollars. These settlements have been challenged as market division agreements.103 The challenges have met with varying degrees of success in the courts. Consider the treatment these challenges have received in the US Court of Appeals for the Eleventh Circuit. The first reverse payment challenge to reach the court was a private challenge.104 In that case, the court rejected the private challenge and held that a patent settlement is not unlawful unless the settlement exceeds the exclusionary potential of the patent. Two years later, the Federal Trade Commission challenged a similar reverse payment settlement and the case was heard by the Eleventh Circuit.105 The court relied on the rule established in its earlier precedent in the private challenge to reject the FTC’s challenge. Copying antitrust rules established in private cases in subsequent public cases only makes sense if one is blind to the very different institutional contests of private and public enforcement. Unlike the private case, the FTC was not suing for treble damages or to impose a fine, but rather to prohibit similar settlements in the future. There should have been no concern in the FTC about the abusive potential of class actions or ill-motivated plaintiffs’ lawyers, incompetent juries, or various other forms of private litigation baggage. Whether or not the settlement should have been upheld in the FTC case, it would have made sense to ask some different questions in that case than in the private case. Unfortunately, it is politically and legally difficult to hold private litigants and the government to different legal standards. The ‘rule of law’ appears to require that legal norms be applied impartially and equally to all litigants. Although in the last few years in the US we have been discussing the possibility of applying different substantive rules in 103

See Daniel A Crane, ‘Exit Payments in Settlement of Patent Infringement Lawsuits: Antitrust Rules and Economic Implications’ (2002) 54 Florida Law Review 747. 104 Valley Drug Co. v Geneva Pharmaceuticals, Inc., 344 F 3d 1294 (11th Cir 2003). 105 Schering-Plough Corp. v FTC, 402 F 3d 1056 (11th Cir 2005).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 409

FTC cases than in others, this possibility has thus far failed to gain much traction. For the time being at least, liability rules created in private litigation continue to dominate US public enforcement. It is uncertain whether this pattern will repeat itself in other jurisdictions that expand the availability of private enforcement. Some of these effects may be idiosyncratic to the US civil litigation apparatus, particularly such features as contingency fees, civil juries, treble damages, and liberal discovery rules. Perhaps jurisdictions that strictly limit the availability of private enforcement will find these effects less acute. Time will tell. 3.3 Expansion of the Antitrust Constituency Thus far we have presented a largely negative view about the systemic effects of private antitrust enforcement. However, if carefully managed, private enforcement has the potential to provide a helpful complement to public enforcement. In particular, private enforcement expands the antitrust ‘constituency’ by creating new groups with an economic incentive to shape and enforce competition law.106 First, private enforcement may increase the volume of competition cases, requiring additional attorneys, expert witnesses, and other staff to handle the workload. Second, private enforcement creates an entirely new (and pro-enforcement) constituency in the plaintiff’s bar.107 Third, private enforcement may draw interest from consumer advocacy or industry advocacy groups, either as parties in litigation or as participants in policy debates concerning the proper scope and role of antitrust and private enforcement.108

106 See, for example, A Douglas Melamed, ‘International Antitrust in an Age of International Deregulation’ (Address before the George Mason Law Review Symposium on Antitrust in the Global Economy, October 10, 1997) 10: ‘In any event, it is rare to find constituents urging sound antitrust policy, without regard to the particular outcome they desire.’ 107 Compare Albert A Foer and Robert H Lande, ‘The Evolution of United States Antitrust Law’ (1999) American Antitrust Institute Working Paper, 7: noting the link between private enforcement and plaintiff bar advocacy and arguing that the US antitrust plaintiffs’ bar has been reduced in influence since the 1970s due to legal changes that ‘have made it more difficult for a private plaintiff to succeed.’ 108 See Melamed (n 106) 8–11: discussing lobbying by private groups related to the Boeing/McDonnell Douglas and Bell Atlantic/NYNEX mergers.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 27 SESS: 7 OUTPUT: Wed Oct 21 12:16:51 2015

410 Comparative competition law

4. ANTITRUST LITIGATION ACROSS BORDERS 4.1 Multiple-jurisdiction Litigation One question for the future is how firms will respond strategically to the growing availability of private antitrust claims in multiple jurisdictions around the world. The days in which litigants flocked automatically to the US courts are waning. Savvy global firms are learning that the antitrust game is often best played on a global scale by a combination of private lawsuits and complaints to the agencies in sympathetic jurisdictions. If US courts continue their recent trend of being less sympathetic to antitrust claims – particularly those involving unilateral exclusionary conduct – than European and other courts, global corporate antagonists may find it advantageous to place more emphasis on other venues. These global strategies are becoming increasingly important in hightech markets, where corporate giants like Microsoft, Intel, AMD, Apple, Samsung, Nokia, and Google wage their legal wars simultaneously on a variety of fronts. The example of the Intel/AMD wars provides a preview of coming attractions. From the early 2000s and continuing to some degree until the present, AMD and Intel waged a global antitrust war over the legal treatment of Intel’s loyalty discounts. To summarize the headlines briefly, AMD secured early decisions against Intel in Japan and Korea, a favorable decision and a €1.06 billion (almost US$1.5 billion) fine against Intel from the European Commission, a $1.25 billion settlement payment from Intel, and a complaint from the FTC that Intel quickly settled. In reaching these results, AMD played a strategic global game, pressing public and private complaints in various venues around the world. Coordination between public and private enforcement was not always smooth. In pressing its case before the European Commission, AMD was set back by the lack of a discovery process for obtaining documents to substantiate its complaints. So it shifted the discovery part of its case to the US, where it filed a private action in federal district court in California seeking to obtain documents from Intel that it could then use in Europe. The case ultimately went to the US Supreme Court, which held that such a use of US discovery in aid of European litigation was available if the party seeking discovery could satisfy a four-part test.109 Although AMD was ultimately unable to satisfy the test and obtain the discovery it sought, the 109

Columns Design XML Ltd

/

Intel Corp. v Advanced Micro Devices, Inc., 542 US 241 (2004).

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 27 /

Date: 8/9

JOBNAME: Duns PAGE: 28 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

Private antitrust enforcement 411

Intel decision leaves open the possibility that complaining parties can pick and choose between jurisdictions – invoking procedural and substantive rules separately in the jurisdictions where the relevant standards are the friendliest. 4.2 Jurisdiction and Extraterritoriality Private enforcement has also tested the jurisdictional limits of the antitrust courts. Traditionally, US and European courts have asserted broad ‘effects-based’ jurisdiction over public antitrust claims.110 Historically, some national legislatures have curtailed foreign jurisdiction by enacting ‘blocking’ or ‘clawback’ statutes to limit process or damages from cases litigated abroad.111 Courts have generally, but not uniformly, declined to grant similarly broad extraterritorial jurisdiction to private enforcement suits. In the United States, the Supreme Court recently declined to assert US jurisdiction over claims by foreign purchasers against foreign defendants in F. Hoffmann-La Roche Ltd. v Empagran S.A.112 Earlier lower court decisions also typically declined to assert extraterritorial jurisdiction over foreign harms by foreign parties.113 In contrast to recent American precedent, some British courts have allowed private antitrust plaintiffs to 110

See, for example, Hartford Fire Ins. Co. v California, 509 US 764 (1993) (case brought by the State of California); United States v Aluminum Co. of America (Alcoa), 148 F 2d 416 (2d Cir 1945): establishing the principle of effects-based jurisdiction in the US in a case brought by the US Department of Justice. Cases C-89, 104, 114, 116–117, 125–129/85 Åhlström Osakeyhtiö v Commission (Wood Pulp) [1988] ECR 5193: affirming effects-based jurisdiction in the European Union. 111 See, for example, Protection of Trading Interests Act of 1980, ch. 11 (UK): enacting Britain’s blocking and clawback statute; Laker Airways Ltd. v Sabena, Belgian World Airlines, 731 F 2d 909 (DC Cir 1984): noting Britain’s blocking and clawback statute but allowing suit to proceed irrespective of the difficulties imposed by the statute. 112 542 US 155 (2004). 113 See In re Banco Santander Securities-Optimal Litigation, 732 F Supp 2d 1305 (SD Fla 2010): dismissing foreign antitrust suit for forum non conveniens; In re Urethane Antitrust Litigation, 683 F Supp 2d 1214, 1221–25 (D Kan 2010): dismissing claims arising under EU antitrust law for forum non conveniens; In re Air Cargo Shipping Servs. Antitrust Litigation, No MD 06-1775, 2008 WL 5958061 (EDNY 2008): report and recommendation affirmed in part, 2009 WL 3443405 (EDNY 21 August 2009) (same); Info Resources, Inc. v Dun & Bradstreet Corp., 127 F Supp 2d 411 (SDNY 2000) (same). But see Multi-Juice, SA v Snapple Beverage Corp., No 02 Civ 4635, 2003 WL 1961636 (SDNY 25

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 28 /

Date: 13/10

JOBNAME: Duns PAGE: 29 SESS: 9 OUTPUT: Wed Oct 21 12:16:51 2015

412 Comparative competition law

sue in British courts for damages related to injuries abroad as long as those injuries have some nexus to Britain.114 Yet the scope of extraterritorial jurisdiction in Britain varies depending on the type of claims brought.115

5. CONCLUSION For better or worse, private remedies for antitrust violations are likely to expand in many emerging antitrust jurisdictions. The US experience with private enforcement suggests that these moves should be made with considerable caution. Private antitrust litigation often fails to live up to its ostensible goals and can produce counter-effects that may undermine antitrust enforcement as a whole. Still, each jurisdiction’s experience will be different, and it remains to be seen how expansion of private enforcement plays out globally.

April 2003): dismissing a case claiming violations of Greek and EU antitrust laws for failure to state a claim only after applying ordinary US procedural rules to the case. 114 See Provimi, Ltd. v Aventis Animal Nutrition SA [2003] EWHC 961 (Comm): first asserting such extraterritorial jurisdiction by allowing German firm to sue for injuries suffered in Germany. More recent cases have affirmed the extraterritoriality principle in Provimi, but only so long as the claims bear some relationship to Britain. See, for example, Cooper Tire & Rubber Co. Europe Ltd. v. Dow Deutschland Inc. [2010] EWCA Civ 864: allowing claims related to an EU-wide cartel because the pleadings alleged the participation of a British firm; SanDisk Corp. v Koninklijke Philips Electronics NV [2007] EWHC 332 (Ch): dismissing claims failing to allege a connection to Britain; Emerson Electric v Morgan Crucible [2011] CAT 4 judgment of 21 March 2011: dismissing private claims against a British subsidiary on the grounds that the British subsidiary itself was not mentioned in the European Commission’s public enforcement action as a cartel member. 115 See generally Peter Scott and Mark Simpson, ‘England and Wales’ in Gotts (n 2): discussing the scope of British courts’ extraterritorial jurisdiction under various subsections of the EU regulations.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 17_Chapter14

/Pg. Position: 29 /

Date: 13/10

JOBNAME: Duns PAGE: 1 SESS: 6 OUTPUT: Wed Oct 21 12:16:51 2015

PART IV COMPETITION LAW IN SELECTED JURISDICTIONS

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 1 /

Date: 21/10

JOBNAME: Duns PAGE: 2 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

15. Competition law in Japan Simon Vande Walle and Tadashi Shiraishi *

This chapter gives an overview of competition law in Japan, with a particular focus on recent case law and developments.1 Brick-sized handbooks have been written about Japan’s competition law in Japanese and condensing that scholarship in a single chapter inevitably means that some issues do not get the attention they deserve. We have nonetheless attempted to touch upon most major aspects of competition law in Japan, including its: (1) historical background, (2) substantive rules, and (3) enforcement mechanisms. We also discuss (4) the application of Japanese competition law to cross-border cases. Finally, we (5) reflect on the role and awareness of competition law in Japan and conclude (6).

1. HISTORICAL BACKGROUND Competition law was introduced in Japan at the behest of the United States (US) occupation authorities after the Second World War. The 1947 Antimonopoly Act has been amended repeatedly but is still the relevant statute.2 Initially, competition law faced skepticism and hostility and played only a minor role in Japanese society. In the 1950s and 1960s, it conflicted with Japan’s industrial policy, which favored cartels as a means to rationalize production. Enforcement in that period was anemic.3

* This work was supported by grant-in-aid for JSPS fellows number 23-01009 and JSPS KAKENHI grant number 25380070. This chapter expresses the authors’ own views, not those of any organization. 1 The chapter takes into account case law and developments up to 1 July 2015. 2 Shiteki dokusen no kinshi oyobi kōsei torihiki no kakuho ni kansuru hōritsu [Act on Prohibition of Private Monopolization and Maintenance of Fair Trade] Law No 54 of 1947, as amended (hereafter ‘Antimonopoly Act’). 3 Yoshio Kanazawa, ‘The Regulation of Corporate Enterprise: The Law of Unfair Competition and the Control of Monopoly Power’ in Arthur Taylor Von Mehren (ed), Law in Japan: The Legal Order in a Changing Society (Harvard Law Review Association 1963) 480, 505.

415

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 1 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

416 Comparative competition law

There was a revival in the 1970s, when the oil crisis wreaked havoc on the Japanese economy, but the momentum was lost again in the 1980s. At the end of the 1980s, however, Japan came under intense pressure from abroad to enforce its competition laws more strictly. The US government in particular viewed weak enforcement as a barrier that prevented American firms from entering the Japanese market. Hence, the issue was taken up in the Structural Impediments Initiative (SII) negotiations, a round of bilateral trade negotiations aimed at curbing the US’s large trade deficit with Japan. At the same time, competition law started gaining increased acceptance within Japan itself, as policymakers saw it as a potential tool to revive a stagnating economy. The convergence of external pressure and greater internal support set the stage for a revival of competition law that started in the 1990s. Since then, the law’s enforcement mechanisms have been significantly strengthened.

2. SUBSTANTIVE RULES After a brief discussion of the goals of competition law in Japan, we turn to the three key rules of the Antimonopoly Act: (1) the prohibition on unreasonable restraints of trade; (2) the prohibition on private monopolization; and (3) the prohibition on unfair trade practices. Within the latter category, we discuss a specific type of unfair trade practice, namely abuse of a superior bargaining position, separately and in some detail. Next we discuss the Antimonopoly Act’s regulation on mergers. The Antimonopoly Act also contains a provision that allows the Japan Fair Trade Commission (JFTC) to break up monopolistic situations, regardless of any conduct.4 The JFTC has never used the provision and it seems unlikely that it will be revived in the near future, so we do not discuss it here. 2.1 The Goals of Competition Law Japan has never subscribed to the view, now commonplace in the US, that competition law’s sole concern should be economic efficiency or consumer welfare. Competition law was originally introduced in Japan to achieve the political goal of economic democratization, i.e. to ensure equal opportunity for all individuals to engage in economic activity and

4 Antimonopoly Act (n 2) Art 2(7), in combination with Art 8-4 (this provision was introduced in 1977).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 5 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 417

to avoid excessive concentration of economic power.5 Indeed, at the time of the Antimonopoly Act’s enactment, one of the major concerns of the US occupation authorities was to prevent the re-emergence of the zaibatsu, the large family-owned conglomerates that had oligopolized the Japanese economy immediately prior to and during the war. Although the concern with the zaibatsu has faded to the background, the goals of competition law in Japan have always encompassed a mix of economic and non-economic elements.6 This diversity is reflected in the Antimonopoly Act’s opening provision, which lists a variety of goals, including the promotion of fair and free competition, the democratic and wholesome development of the national economy and the safeguarding of consumer interests.7 2.2 Horizontal Restraints The Antimonopoly Act’s prohibition on unreasonable restraints of trade8 is comparable to Section 1 of the Sherman Act and Article 101 of the Treaty on the Functioning of the European Union (TFEU). However, an early case held that it only applies to agreements between companies that are in a competitive relationship.9 Hence, the prohibition applies only to horizontal restraints such as price-fixing and bid-rigging, not vertical restraints. The latter are mostly governed by the prohibition on unfair trade practices, which is discussed below. An agreement will only constitute an unreasonable restraint of trade when it causes a ‘substantial restraint of competition.’ This is a crucial concept in the Antimonopoly Act, as it is not only a constitutive element of an unreasonable restraint of trade, but also of private monopolization, and the key test to assess whether mergers are lawful or not. A substantial restraint of competition exists when one company or several companies 5 See Hiroshi Iyori, ‘Competition Policy and Government Intervention in Developing Countries: An Examination of Japanese Economic Development’ (2002) 1 Washington University Global Studies Law Review 35, 39–40. 6 Shūya Hayashi, ‘The Goals of Japanese Competition Law’ in Josef Drexl, Laurence Idot and Joël Monéger (eds), Economic Theory and Competition Law (Edward Elgar 2009). 7 Antimonopoly Act (n 2) Art 1. 8 Antimonopoly Act (n 2) Art 2(6) (definition) and Art 3 (prohibition). 9 Tokyo Kōtō Saibansho [Tokyo High Ct] 9 March 1953, 2 Hanrei jihō 8 (K K Asahi Shinbunsha et al v Japan Fair Trade Commission). But see Tokyo Kōtō Saibansho [Tokyo High Ct] 14 December 1993, 840 Hanrei Taimuzu 81, 88 (Japan v Toppan mūa et al) (expressing doubts in obiter dictum as to the validity of this view).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 6 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

418 Comparative competition law

together have the power to determine price, quality, output or other market conditions; in short, when they have market power. Hence, a substantial restraint of competition requires a finding that the practice in question creates, maintains, or strengthens market power.10 Since an unreasonable restraint of trade requires the existence of a substantial restraint of competition, which, in turn, requires market power, it follows that the Antimonopoly Act contains no per se rule against price-fixing. For instance, a cartel between companies that together have no market power would not be caught by the ban on unreasonable restraints of trade. Of course, such a cartel would be unable to raise prices in the first place and, therefore, the fact that it does not fall under the prohibition is unproblematic. However, the requirement does mean that, in each case, the JFTC must delineate the market and establish the existence of a substantial restraint of competition. A side-effect of this is that the JFTC tends to define markets narrowly, to avoid any challenge that the cartel did not lead to a substantial restraint of competition.11 Another important requirement for conduct to be caught by the prohibition on unreasonable restraints of trade is that the entrepreneurs must have acted ‘in concert.’12 Such concerted conduct can result from an explicit or tacit agreement to collude, and indirect or circumstantial evidence can be sufficient to prove the agreement.13 This requirement is particularly important in assessing the lawfulness of parallel price increases, which often occur in oligopolistic markets. A parallel price increase as such is not unlawful, but when coupled with an exchange of information, it may constitute sufficient evidence of concerted action.14 10

Tokyo Kōtō Saibansho [Tokyo High Ct] 19 September 1951, 4(14) Kōminshū 497, 518 (in the context of a merger between Tōhō and Subaru); Saikō Saibansho [Sup Ct] 17 December 2010, 64(8) Minshū 117, 130 (NTT East v Japan Fair Trade Commission) (under point 5, second sentence; in the context of an exclusion-type private monopolization case); Saikō Saibansho [Sup Ct] 20 February 2012, 66(2) Minshū 796, 810 (in the context of a hard-core cartel). 11 Toshiaki Takigawa, ‘Competition Law and Policy of Japan’ (2009) 54(3) Antitrust Bulletin 435, 455. 12 Antimonopoly Act (n 2) Art 2(6). This requirement has been re-formulated by case law as a requirement that there be an ishi no renraku, a term whose English translation is so elusive that no two authors seem to have ever translated it in the same way. Literally the term means ‘contact of intentions.’ 13 Tokyo Kōtō Saibansho [Tokyo High Ct] 25 September 1995, 42 Shinketsushū 393, 417 (Tōshiba Chemical case). 14 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 4 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 419

In Japan, the prohibition on unreasonable restraints of trade has frequently been applied to collusive tendering, also known as bid-rigging. In this area, Japanese case law makes a distinction between the general or basic agreement between companies to engage in bid-rigging and the specific arrangement to rig a particular tender. The distinction is best illustrated with an example. Assume that construction companies in Kyoto agree to engage in bid-rigging for tenders issued by the city. This would constitute the basic agreement. The city subsequently issues a tender for the construction of certain road works and the companies agree that company X should be awarded the contract and they submit their bids accordingly. This is the specific arrangement. If the JFTC can establish the existence of a basic agreement, this is normally sufficient to establish an unreasonable restraint of trade.15 Proof of such a basic agreement also creates a presumption that the agreement was implemented in the specific tenders that occurred while the agreement was in effect. This is important because the fine is calculated as a percentage of the unlawful turnover, i.e. the turnover derived from construction works for which there was bid-rigging.16 Normally, the fine will be based on the turnover of all projects that fell under the basic agreement. That this is not always the case, however, is illustrated by the Tsuchiya kigyō. In that case, 14 construction companies had been invited to submit a tender for road works.17 All 14 companies had been party to a basic bid-rigging agreement that had functioned for years.18 However, in the case of the road works tender at issue, only 13 of the 14 companies managed to agree on a pre-determined winner. One company – Tsuchiya – had not agreed and wanted the works for itself. When the bids were submitted, 12 companies submitted artificially high bids, to make sure the 13th company – the pre-determined winner – would secure the tender. Tsuchiya, however, slightly underbid the predetermined winner and won the tender.19 The question was whether Tsuchiya could be subject to a fine. According to the Tokyo High Court, any restraint of competition in this case had been caused by the 13 companies that had made the specific arrangement, not Tsuchiya. Since

15 Saikō Saibansho [Sup Ct] 20 February 2012, Heisei 22 (gyō-hi) no 278, under point 4(2). 16 Antimonopoly Act (n 2) Art 7-2(1). 17 Tokyo Kōtō Saibansho [Tokyo High Ct] 20 February 2004, 50 Shinketsushū 708, 711 (Tsuchiya kigyō v Japan Fair Trade Commission). 18 Ibid 709–710. 19 Ibid 711.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 5 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

420 Comparative competition law

Tsuchiya did not take part in the specific arrangement, the court held it could not be fined.20 2.3 Private Monopolization The prohibition on private monopolization21 can be compared with the Sherman Act’s ban on monopolization and the EU’s prohibition on abuse of a dominant position, although there are some notable differences. First, unlike in the US or the EU, there is no statutory requirement of monopoly power or a dominant position.22 However, private monopolization does require a ‘substantial restraint of competition’ and, hence, a certain degree of market power. Moreover, the JFTC’s enforcement policy is to target companies with a market share in excess of 50 percent.23 Second, there are two types of private monopolization: exclusion-type and control-type. Exclusion-type monopolization covers practices that are aimed at excluding competitors, such as predatory pricing or refusals to deal. Control-type monopolization, on the other hand, refers to conduct by which one company, e.g. a majority shareholder, causes another company to follow its will. Monopolization cases have been scarce. Rather than invoking the private monopolization prohibition, the JFTC and private litigants often prefer to invoke the prohibition on unfair trade practices, which, contrary to private monopolization, does not require proof of a substantial restraint of competition. Moreover, amendments in 2005 and 2009 have made private monopolization subject to fines and companies are therefore more

20

Ibid 714–717. Antimonopoly Act (n 2) Art 2(5) (definition) and Art 3 (prohibition). 22 Tadashi Shiraishi, ‘Dokusen kinshi hō’ (Competition Law of Japan) 290 (2nd edn, 2009); Noboru Kawahama, Dai 2jō5 – Teigi – Shiteki dokusen [Article 2(5) – Definition – Private Monopolization], in Akira Negishi (ed) ‘Chūshaku dokusen kinshihō’ [Commentary on the Antimonopoly Law of Japan] (2009) 25, 26 (contrasting the provision in the Antimonopoly Act with Section 2 of the Sherman Act). 23 Japan Fair Trade Commission, Haijogata shiteki dokusen ni kakaru dokusen kinshi hōjō no shishin [The Guidelines for Exclusionary Private Monopolization Under the Antimonopoly Act] 3 (28 October 2009). In case of tying, the market share is assessed on the market for the tying product and in case of a refusal to deal, it is assessed on the upstream market: ibid 4 (note 2). 21

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 6 /

Date: 13/10

JOBNAME: Duns PAGE: 9 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 421

likely to resist a challenge based on private monopolization than one based on unfair trade practices.24 In recent years, the JFTC has nonetheless brought some monopolization cases. In the NTT East case, the JFTC found that the incumbent telecom operator had tried to exclude competitors from the market for FTTH, i.e. ‘fiber-to-the-home’ high-speed internet access. The prices NTT East was charging its own customers were lower than the prices it was charging its competitors for access to NTT’s network, which they needed to offer high-speed internet access to their customers. Hence, it would be almost impossible for competitors to compete or enter the market. The JFTC’s finding that NTT East engaged in exclusion-type private monopolization was ultimately upheld by the Supreme Court.25 In JASRAC, the JFTC issued a cease-and-desist order based on exclusion-type private monopolization by Japan’s largest music copyright collecting society.26 It found that JASRAC had excluded competitors by granting blanket licenses to broadcasters in return for a royalty that did not depend on how much music the broadcasters used. However, the cease-and-desist order was subsequently annulled after a review by a panel within the JFTC.27 The case has resulted in a protracted battle, as the JFTC panel’s decision was in turn appealed by a third party and annulled by the courts.28 As a result, the case is back with the JFTC, which has to decide whether or not to uphold the initial cease-and-desist order. The most recent monopolization case, Fukui Prefectural Federation of Agricultural Cooperatives, involved the control-type of monopolization.29

24 The different levels of fines for private monopolization and unfair trade practices are discussed below, in the section ‘Fines (surcharges)’. The two monopolization cases discussed in this section (NTT East and JASRAC) were handled before the relevant 2009 amendment took effect and, hence, no fines were imposed. 25 Saikō Saibansho [Sup Ct] 17 December 2010, Heisei 21 (gyō-hi) no 348, 57(2) Shinketsushū 215. 26 Japan Fair Trade Commission, Cease-and-desist order of 27 February 2009, Heisei 21 (so) no 2, 55 Shinketsushū 712. 27 Japan Fair Trade Commission, Decision of 12 June 2012, Heisei 21 (han) no 17. 28 Tokyo Kōtō Saibansho [Tokyo High Ct] 1 November 2013, available at www.jftc.go.jp. A further appeal against this judgment was dismissed by the Supreme Court: Saikō Saibansho [Sup Ct] 28 April 2015, Heisei 26 (gyō-hi) no 75, available at www.jftc.go.jp. 29 Japan Fair Trade Commission, Cease-and-desist order of 16 January 2015, Heisei 27 (so) no 2.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

422 Comparative competition law

Agricultural cooperatives in Fukui prefecture needed silos to dry their rice and instructed a local federation of which they were members to procure these silos on their behalf. The local federation was obliged by law to procure the silos through competitive tenders. Instead, for each silo, the federation dictated which construction company would build it and at what price, and then instructed the construction companies to submit their bids accordingly. The JFTC found that the federation had controlled the business activity of the silo construction companies and substantially restrained competition in the market for the construction of silos. This conduct constituted private monopolization and the JFTC issued a cease-and-desist order against the federation. The JFTC did not fine the federation because, among other reasons, the federation did not sell anything to the silo construction companies. 2.4 Unfair Trade Practices: General The Antimonopoly Act prohibits ‘unfair trade practices,’ which is an umbrella term for around a dozen specific types of unlawful conduct.30 Most types of unlawful conduct are specified in the Antimonopoly Act31 although some – those that are not subject to fines – are listed in a notice issued by the JFTC.32 Originally, the Antimonopoly Act did not prohibit ‘unfair trade practices’ but ‘unfair methods of competition,’ a term drawn from Section 5 of the American Federal Trade Commission Act (FTC Act), which also prohibits unfair methods of competition.33 In the US, the prohibition on unfair methods of competition is a catch-all provision that covers, among others, all Sherman Act and Clayton Act offenses. Hence, conduct that violates the Sherman Act’s ban on restraints of trade and monopolization also violates the FTC Act’s prohibition on unfair methods of competition. This overlap makes some sense in the US because the Federal Trade Commission’s jurisdiction is partly defined by reference to the concept of 30

Antimonopoly Act (n 2) Art 2(9) (definition) and Art 19 (prohibition). The exact number depends on whether one treats several closely related types of conduct as one specific type or not. 31 Antimonopoly Act (n 2) Art 2(9)(i)–(v). 32 Antimonopoly Act (n 2) Art 2(9)(vi), in combination with Japan Fair Trade Commission, Fukōseina torihiki hōhō [Designation of Unfair Trade Practices] (18 June 1982, as amended in 2009) (hereinafter ‘Designation of Unfair Trade Practices’). 33 15 USC § 45(a)(1). Since 1938, Section 5 also prohibits ‘unfair or deceptive acts or practices’ and this prohibition forms the basis for much of the FTC’s consumer protection work.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 423

unfair methods of competition. In Japan, however, a single agency is responsible for the enforcement of competition law and hence, it would have been absurd to interpret the ban on unfair methods of competition in the Antimonopoly Act in a similarly broad fashion. Instead, Japanese legislators and courts have given their own interpretation to the prohibition. In 1953, the legislature changed the term ‘unfair methods of competition’ to ‘unfair trade practices’ and modified the types of conduct covered by the prohibition. In its present form, the prohibition is aimed at a catalogue of mostly vertical restraints, such as refusals to deal, vertical price-fixing, tie-in sales, exclusive dealing, selling at unjustly low prices, discriminatory pricing and abuse of a superior bargaining position. The prohibition of unfair trade practices reflects a strong concern for fair competition rather than efficiency or consumer welfare. Some effect on the market is required to find an unfair trade practice but the effect is a ‘light version’ of the anti-competitive effect required in cases of unreasonable restraints of trade or private monopolization. The latter two violations require a ‘substantial restraint of competition,’ which in turn entails an inquiry into whether the suspected violators have market power. By contrast, for unfair trade practices cases, it is sufficient that the conduct ‘tends to impede fair competition’,34 a test that does not require market power. This means that practices such as selling at unjustly low prices, discriminatory pricing or tie-in sales can be unlawful, even when implemented by non-dominant firms. 2.5 Unfair Trade Practices: Abuse of a Superior Bargaining Position Abuse of a superior bargaining position35 is one type of unfair trade practices. It has taken on renewed importance since a 2009 amendment made it subject to penalties. The term in Japanese, yūetsuteki chii no ranyō, is sometimes also translated as abuse of a dominant bargaining position. The prohibition is aimed at curtailing exploitative abuses in vertical transactions between businesses. It has frequently been applied to largescale retailers exploiting suppliers or to powerful franchisors exploiting 34 Antimonopoly Act (n 2) Art 2(9)(vi) (as a result of a hastily drafted amendment in 2009, this requirement is now explicitly mentioned only for the unfair trade practices listed under (vi) but there is little doubt that it also applies, as before, to the unfair trade practices listed under (i) to (v)). 35 Antimonopoly Act (n 2) Art 2(9)(v) and Art 2(9)(vi)(e).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 9 /

Date: 13/10

JOBNAME: Duns PAGE: 12 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

424 Comparative competition law

franchisees. In what follows, we survey several recent abuse cases decided after the ban became subject to penalties. Next, we compare Japan’s prohibition on abuse of a superior bargaining position with the EU’s prohibition on abuse of a dominant position. Finally, we discuss the Subcontract Act, which supplements the prohibition on abuse of a superior bargaining position, and query why rules on exploitative abuse are more significant in Japanese competition law than in, for instance, US antitrust law. The first abuse case in which the JFTC imposed fines was Sanyō Marunaka.36 The case involved a supermarket chain in the Okayama area, where the chain was the largest supermarket and where it was the most important business partner for many local suppliers. The chain had abused its superior bargaining position by (1) forcing its suppliers to send thousands of employees to its stores, to remodel and embellish them; (2) forcing its suppliers to contribute financially to events organized by the store that produced no benefit to the suppliers; (3) unilaterally returning unsold goods to its suppliers; (4) unilaterally discounting the prices of products it bought from its suppliers; and (5) forcing suppliers to buy Christmas cakes and the like. Some of the same types of abusive conduct were at issue in Toys’R’Us.37 In that case the JFTC fined a major toy retailer because it had unilaterally returned unsold toys to its suppliers and accorded itself discounts. The three most recent cases brought by the JFTC illustrate that a breach of the prohibition on abuse of a superior bargaining position can lead to substantial penalties. In Edion,38 the JFTC imposed a fine in excess of 4 billion yen (more than 40 million US dollars at the time the fine was imposed) on a nationwide retailer of electrical appliances because it had forced its suppliers to send more than 10,000 employees to its retail stores for work on displays and remodeling. In Ralse, the JFTC 36 Japan Fair Trade Commission, Cease-and-desist Order and Surcharge Order of 22 June 2011, Heisei 23 (so) no 5 and Heisei 23 (nō) no 87, appeal pending (fining KK Sanyō Marunaka the equivalent of 2.7 million US dollars). 37 Japan Fair Trade Commission, Cease-and-desist Order and Surcharge order of 13 December 2011, Heisei 23 (so) no 13 and Heisei 23 (nō) no 262. These two orders were appealed and, on appeal, a JFTC panel partly annulled the cease-and-desist order and surcharge order, finding that Toys’R’Us did not engage in an abuse of a superior bargaining position towards certain companies. As a result, the amount of the fine to be paid by Toys’R’Us was reduced: Japan Fair Trade Commission, Decision of 4 June 2015, Heisei 24 (han) nos 6 and 7. 38 Japan Fair Trade Commission, Cease-and-desist Order and Surcharge order of 16 February 2012, Heisei 24 (so) no 6 and Heisei 24 (nō) no 10, appeal pending.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 10 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 425

imposed a fine of close to 1.3 billion yen on a supermarket chain in Hokkaido39 and in Direx it inflicted a fine of similar magnitude on a chain of discount stores mostly active in Kyushu.40 From a comparative perspective, the prohibition on abuse of a superior bargaining position is often presented as a unique feature of competition law in Japan. In reality, however, several other jurisdictions have similar rules.41 Moreover, the Japanese prohibition’s difference from the EU’s prohibition on abuse of a dominant position, laid down in Article 102 of the TFEU, should not be overstated.42 Although Article 102 of the TFEU is most frequently applied to tackle exclusionary abuse, it also targets exploitative abuse. Nonetheless, commentators stress the difference between the Japanese concept of ‘superior bargaining position’ and the European Union (EU) concept of ‘dominant position.’ Indeed, the guidelines of the JFTC make it clear that a company with a superior bargaining position ‘does not need to have a market-dominant position.’43 Instead it only ‘needs to have a relatively superior bargaining position compared to the counterparty.’ Hence, the conventional view is that the Japanese concept centers on the superior company’s power vis-à-vis its trading partner, whereas the EU concept focuses on the dominant company’s market power. In fact, an analysis of the JFTC’s Guidelines Concerning Abuse of a Superior Bargaining Position under the Antimonopoly Act and the relevant case law suggests that this conceptual divide may be less wide than often suggested. The key element for finding a superior bargaining position is that the inferior party has no other choice than to accept the

39 Japan Fair Trade Commission, Cease-and-desist Order and Surcharge Order of 3 July 2013, Heisei 25 (so) no 9 and Heisei 24 (nō) no 31, appeal pending. 40 Japan Fair Trade Commission, Cease-and-desist Order and Surcharge Order of 5 June 2014, Heisei 26 (so) no 10 and Heisei 26 (nō) no 113, appeal pending. 41 See International Competition Network (hereafter ‘ICN’) – Task Force for Abuse of Superior Bargaining Position, ‘Report on Abuse of Superior Bargaining Position’ (ICN 2008) . Austria, France, Germany, Italy, Japan, Korea and Slovakia have a specific provision on abuse of a superior bargaining position. 42 Tadashi Shiraishi, ‘A Baseline for Analyzing Exploitative Abuse of a Dominant/Superior Position’ (2013) 5 University of Tokyo Soft Law Review 1, 7. 43 Japan Fair Trade Commission, Yūetsuteki chii no ranyō ni kansuru dokusen kinshi hō jō no kangaekata [Guidelines Concerning Abuse of a Superior Bargaining Position Under the Antimonopoly Act] (30 November 2010) (under II.1).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 11 /

Date: 13/10

JOBNAME: Duns PAGE: 14 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

426 Comparative competition law

superior party’s unreasonable request.44 This is tantamount to saying that the inferior party needs to continue the business relationship, i.e. it cannot switch to another trading partner or can do so only at great cost. In line with this criterion, many abuse cases concern captive or locked-in parties. These cases can be conceptualized as cases in which the relevant market should be defined by reference to these captive parties. On this narrow market, the superior party has a dominant position and this dominant position allows the superior party to exploit the inferior party. The prohibition on abuse of a superior bargaining position is supplemented by the Subcontract Act,45 a specific law prohibiting various forms of exploitative abuse of subcontractors. This law is also enforced by the JFTC and can be regarded as a more detailed version of the prohibition on abuse of a superior bargaining position, specifically applicable in the relationship between subcontractors and the companies relying on them. The concern of Japanese competition law with exploitative conduct is sometimes linked with the idea that close, long-term and restrictive vertical relationships are a distinctive feature of the Japanese economy.46 Major Japanese manufacturers, for instance, are said to exercise a particularly large degree of control over their suppliers and distributors, which often form a vertical or distribution keiretsu or network, organized around the main manufacturer (e.g. Toyota keiretsu or Matsushita keiretsu). Likewise, most of Japan’s ubiquitous convenience stores are also operated on the basis of long-term vertical agreements between franchisors and franchisees, who are strongly dependent on the franchisor and rarely switch. Close-knit vertical relations such as these may reduce transaction costs but they also place smaller suppliers, subcontractors, franchisees or retailers in a position of dependency, creating the potential for abuse. Whether close vertical ties are unraveling in recent years is the subject of debate in the economic literature but their prevalence in the past helps explain why Japan has developed relatively elaborate rules on exploitative abuse.

44

Ibid (under II.1). Shitauke daikin shiharai chien tō bōshi hō [Act Against Delay in Payment of Subcontract Proceeds, Etc. to Subcontractors] Law No 120 of 1956. 46 See, for example, Mitsuo Matsushita, ‘The Antimonopoly Law of Japan’ in Edward M Graham and J David Richardson (eds), Global Competition Policy (Institute for International Economics 1997) 192–193. 45

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 12 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 427

2.6 Merger Control The JFTC will block mergers when they ‘may substantially restrain competition in a particular field of trade.’47 The reference to a particular field of trade means the JFTC must delineate the relevant product and geographic market. Merger Guidelines specify how the JFTC assesses whether the merger will substantially restrain competition in a particular market, with specific rules for horizontal, vertical and conglomerate mergers.48 In general, the JFTC uses an economics-based approach, in line with the approach taken by US and EU competition authorities. The JFTC has never formally blocked a merger and difficult cases are generally approved after the parties offer remedies. In 2012, for instance, the JFTC had to assess the merger between the Tokyo and Osaka stock exchanges, which are Japan’s two largest stock markets. The JFTC defined the relevant geographic market as Japan. In some product markets, the combined entity had market shares of up to 95 percent. In spite of these high levels of concentration, the JFTC approved the merger, after the merging parties offered behavioral remedies. Another remarkable case was the merger between Nippon Steel and Sumitomo Metal Industries in 2012, which resulted in the formation of the world’s second largest steel maker. In this case too, the JFTC defined the geographical market as Japan. Although the merger reduced the number of competitors from three to two in some product markets, the JFTC nonetheless approved the merger. It accepted the behavioral remedies proposed by the parties for some product markets and found that, in other product markets, the merger would not lead to a substantial restraint of competition, among others because any attempt by the merged entity to increase prices would probably be countered by increased imports from Korea and China.49 Apart from prohibiting mergers that would substantially restrain competition in a specific market, the Antimonopoly Act also prohibits

47 Antimonopoly Act (n 2) Art 10(1), Art 13, Art 14, Art 15(1)(i), Art 15-2(1)(i); Art 15-3(1)(i); Art 16 (1). 48 Japan Fair Trade Commission, Kigyō ketsugo shinsa ni kansuru dokusen kinshi hō no unyō shishin [Guidelines on the Application of the Anti-Monopoly Act to Reviewing Business Combinations] (31 May 2004, as amended). 49 Taking import pressure into account in the markets for hot-rolled steel sheets and H-section steel: Japan Fair Trade Commission, ‘Press Release: Results of Investigation into the Proposed Merger Between Nippon Steel Corporation and Sumitomo Metal Industries, Ltd’ (Japan Fair Trade Commission Website, 14 December 2011) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

428 Comparative competition law

companies from becoming so large that this would result in an excessive concentration of economic power in the Japanese economy in general. These anti-conglomerate or ‘anti-bigness’ provisions are the remnants of provisions originally designed to prevent the re-emergence of the zaibatsu. For instance, the Act prohibits the creation of a company with a substantial position in more than five interrelated business fields if this would have ‘large effects on the national economy and impede fair and free competition.’50 Although such prohibition may sound far-reaching, its scope is interpreted narrowly to avoid excessive interference with the market and no cases are recorded in which the JFTC blocked a transaction based on it. Another ‘anti-bigness’ provision that existed for a long time was the blanket ban on holding companies. That ban was abolished in 1997 and what remains is a provision that prohibits banks and insurance companies from holding more than 5 and 10 percent respectively of the voting stock in a nonfinancial company, subject to some exceptions.51 The prohibition is generally well complied with and, hence, it rarely gives rise to any problems.

3. ENFORCEMENT The Antimonopoly Act is primarily enforced by the JFTC through administrative proceedings. Private enforcement also occurs regularly and in case of price-fixing and bid-rigging, criminal proceedings have also been brought. For mergers, enforcement occurs through a system of ex ante control. 3.1 Administrative Enforcement This section of the chapter discusses administrative enforcement by the JFTC. We first discuss how the JFTC investigates possible violations. Next, we examine the hearing process that the JFTC must organize before it can issue a formal order in response to a violation. We then discuss the two kinds of orders the JFTC can issue: cease-and-desist orders and surcharge orders. We treat the latter in some detail because 50 Antimonopoly Act (n 2) Art 9 in combination with Japan Fair Trade Commission, Jigyō shihairyoku ga kado ni shūchū suru koto to naru kaisha no kangaekata [Guidelines Concerning Companies Which Constitute an Excessive Concentration of Economic Power] (12 November 2002) (under point 2). 51 Antimonopoly Act (n 2) Art 11.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 429

they constitute the most important sanction for violations of competition law in Japan. Next we discuss Japan’s leniency system. We then turn to the procedure by which the JFTC’s orders can be appealed, which was recently revised. Finally, we briefly survey the JFTC’s enforcement record. 3.1.1 Investigation Investigations by the JFTC are triggered by complaints, leniency applications or started ex officio.52 The JFTC typically gathers evidence through on-site inspections, requests for information, document production orders and witness statements.53 Persons that refuse to comply with the JFTC’s investigative measures or provide false information are subject to criminal penalties.54 The threat of criminal penalties means the JFTC can indirectly compel persons to be interrogated, since not doing so would expose them to criminal penalties. In practice, however, the JFTC rarely uses its formal powers to interrogate witnesses subject to penalties. Instead, the JFTC gathers witness statements through interviews that are technically voluntary and, in that case, the witnesses are not subject to penalties for providing false information. A notable feature of antitrust investigations in Japan is that there is no attorney–client privilege. Hence, the JFTC can seize correspondence between the company under investigation and its attorneys. Moreover, company employees who are interviewed, whether on a voluntary basis or pursuant to the JFTC’s legal powers, do not have the right to be assisted by an attorney during the interrogation. Businesses have long complained that these rules violate their rights to due process and, in the wake of a 2013 amendment to the Antimonopoly Act, the Japanese government set up an advisory panel to consider whether the rules should be changed. The panel consisted of academics and representatives of both businesses and consumer organizations and ultimately concluded that an attorney–client privilege should not be introduced, among other reasons because of concerns that such a privilege would impede the fact-finding

52

Ibid Art 45(1)–(3) (relating to investigations after complaints by third parties) and Art 45(4) (allowing the JFTC to start investigations ex officio, which also allows the JFTC to start an investigation after a leniency application). 53 The Japan Fair Trade Commission’s powers to take these and other investigative measures are laid down in the Antimonopoly Act (n 2) Art 47. 54 Antimonopoly Act (n 2) Art 94.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 18 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

430 Comparative competition law

ability of the JFTC.55 The panel also did not support the introduction of a rule allowing an attorney to be present during interrogations.56 3.1.2 Right to be heard before the JFTC issues an order Before the JFTC can issue any formal order against a violation, the companies under investigation have the right to be heard. This right has been significantly strengthened by a 2013 amendment to the Antimonopoly Act, which entered into force on 1 April 2015. Under the new rules, any cease-and-desist order or surcharge order is preceded by a hearing process, overseen by a JFTC official who has not worked on the case.57 At the start of this process, the companies under investigation are informed of the JFTC’s findings of fact, legal analysis and the content of the proposed cease-and-desist order.58 In the event that the JFTC intends to levy a fine, i.e. if it intends to issue a surcharge order, it must also inform the companies of the amount of the proposed fine and how it was calculated.59 The companies also obtain access to the evidence on which the JFTC’s findings of fact are based. More specifically, they have the right to consult the JFTC’s evidence at a predetermined place and, for evidence submitted by the company itself or its employees, make copies.60 Subsequently, a hearing is held at which the companies can present their views, submit evidence and put questions to the case team.61 The JFTC official presiding over the hearing drafts a record of the hearing, containing a summary of the points made by the companies under investigation,62 and a separate document listing the issues in dispute.63 Both documents are submitted to the JFTC, which must duly consider them when taking a final decision on whether to issue an order.64 55

Report Issued by the Advisory Panel on Administrative Investigation Procedures under the Anti-Monopoly Act (Summary), 24 December 2014, at 2 . 56 Ibid. 57 Antimonopoly Act (n 2) (as amended in 2013) Art 49 (for cease-anddesist orders); Art 62(4) (for surcharge orders); Art 53 (requirements for the official that will oversee the hearing process). 58 Ibid Art 50(1)(i)–(ii). 59 Ibid Art 62(4). 60 Ibid Art 52(1). 61 Ibid Art 54(2). 62 Ibid Art 58 (1). 63 Ibid Art 58(4). 64 Ibid Art 60.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 19 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 431

3.1.3 Cease-and-desist order If, after the hearing process, the JFTC is still convinced that a violation took place, it will usually order the violating company to cease the conduct in a cease-and-desist order.65 The JFTC may also order ‘any other measures necessary to eliminate’ the violation. For instance, the JFTC has frequently ordered cartel members to send letters to customers confirming that cartel activities have been terminated and to organize competition law training for their employees. 3.1.4 Fines (surcharges) Surcharges are the fines imposed by the JFTC. Most but not all violations of the Antimonopoly Act are subject to these fines and they are imposed through a surcharge order, which is normally issued together with a cease-and-desist order. The term surcharge reflects the fact that the fine is calculated as a percentage, fixed by law, of the violating firm’s sales of the product or service to which the violation relates. The surcharge rate for cartels is 10 percent, so if the JFTC uncovers a widget cartel, it will order each cartel member to pay 10 percent of its turnover for widgets in the relevant geographic market. The turnover to be taken into account is that which was generated while the violation lasted, but with a maximum of three years.66 The conception of fines as ‘surcharges’ has two important consequences. First, they are imposed mechanically, meaning the JFTC has no discretion in whether to impose them or not. If the JFTC finds a violation, it must apply the surcharge rate and impose the resulting fine. Second, because the fine is calculated as a percentage of the turnover to which the violation relates, a firm that participates in the violation but does not generate any turnover from it cannot be fined. Thus, in bid-rigging cases for instance, only the company that was awarded the tender has to pay a fine. The other companies, even if they colluded to ensure the predetermined winner would win, are not fined. Those companies derive no turnover from the violation and therefore pay no fine because 10 percent of zero is still zero. Likewise, if the relevant geographic market is Japan and a foreign company agrees with its Japanese rivals that it will stay out of the Japanese market, the foreign company will be in violation of the Antimonopoly Act but it will not be fined, since it derives no turnover from the violation in the relevant market. The Marine Hose case is an

65 66

Columns Design XML Ltd

Ibid Art 7 and Art 20. Ibid Art 7-2(1).

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 17 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

432 Comparative competition law

illustration of this approach. In that case, both foreign and Japanese companies were party to an unlawful market sharing agreement but only the Japanese company was fined, because the foreign companies were not selling in Japan.67 The surcharge rate is different for each type of violation. Hard-core horizontal restraints such as cartels and bid-rigging are subject to a 10 percent rate, while horizontal restraints that fall outside of this category are not subject to any fine.68 Exclusion-type private monopolization is subject to a 6 percent rate,69 while control-type monopolization is sanctioned with a 10 percent surcharge.70 Most unfair trade practices are only subject to fines in case of a repeat offense and in that case they attract a surcharge of 3 percent of the relevant turnover.71 Abuse of a superior bargaining position, however, is subject to a surcharge even in the case of a first-time violation, with the surcharge calculated as 1 percent of the entire turnover between the party with a superior bargaining position and the exploited parties.72 The other unfair trade practices are not subject to any fines. The application of these surcharge rates leads to fines that are substantial, but nonetheless significantly lower than the administrative fines imposed by the European Commission, which start from a base rate of up to 30 percent of the relevant turnover and are multiplied by the number of years during which the violation lasted, without limit. They are also lower than the criminal fines imposed on cartels by the US Department of Justice, which start from a base fine of 20 percent of the volume of affected commerce. The current rates are a multiple of the rates applicable when the surcharge system was introduced in 1977, with significant increases in 1991 and 2005. As a result, from a historical perspective, the current fines are at record high levels, as shown in Figure 15.1.73 67 Japan Fair Trade Commission, Cease-and-desist Order and Surcharge Order of 20 February 2008, Heisei 20 (so) no 2 and Heisei 20 (nō) no 10, 54 Shinketsushū 512, 623. 68 Ibid Art 7-2(1) (the Antimonopoly Act does not use the term hard-core horizontal restraint but restraints that relate to or affect ‘the consideration (taika)’ of goods and services). 69 Ibid Art 7-2(4). 70 Ibid Art 7-2(2). 71 Ibid Art 20-2 to 20-5, in combination with Art 2(9)(i) to Art 2(9)(iv). 72 Ibid Art 20-6, in combination with Art 2(9)(v). 73 The amount in 2010 is particularly high but this is partly the result of a change in the enforcement procedure and the way the JFTC allocates surcharges to a specific year. In fact, the figure for 2010 includes not only the surcharge

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 18 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 433 ¥80

Billions of yen

¥70 ¥60 ¥50 ¥40 ¥30 ¥20 ¥10 ¥0 99 00 01 02 03 04 05 06 07 08 09 10 11 12 13 Source: Japan Fair Trade Commission, Jimusōchō teireikaiken kiroku [Minutes of the Secretary General’s Regular Press Conference], ‘Sūji de miru kōsei torihiki iinkai no rekishi ni tsuite’ [The History of the Japan Fair Trade Commission in Numbers] 3 (Japan Fair Trade Commission Website, 16 July 2014) .

Figure 15.1 Evolution of surcharges imposed by the JFTC

3.1.5 Leniency system for cartels A 2005 amendment introduced a leniency system that allows companies to obtain full or partial immunity from fines, in return for information and evidence about cartels. If a leniency application is made before the JFTC has started investigating the violation, full immunity will be granted to the first leniency applicant.74 The second applicant in that pre-investigation stage will benefit from a 50 percent reduction, while the third, fourth and fifth applicants will obtain a 30 percent reduction.75

imposed by surcharge orders issued in 2010 but also the surcharges imposed in previous years and confirmed in decisions (shinketsu) rendered in 2010 under the administrative procedure that was in force prior to a 2005 amendment of the Antimonopoly Act. Under that old procedure, surcharge orders were suspended as long as the administrative appeal before the Commission was pending. About half of the amount allocated to 2010 corresponds to surcharges imposed by orders issued in previous years. 74 Antimonopoly Act (n 2) Art 7-2(10). 75 Ibid Art 7-2(11).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

434 Comparative competition law

The reductions are much less generous once an investigation has started. Up to three applicants can receive leniency in the postinvestigation stage but only a 30 percent reduction will be granted at that stage.76 However, the combined number of leniency applicants in the preand post-investigation stage can never be more than five, so if, for instance, four companies have already applied for leniency in the pre-investigation stage, only one company will be able to obtain leniency in the post-investigation stage. Successful leniency applicants obtain full or partial immunity from fines and, to some extent, criminal sanctions. The immunity from fines flows from the Antimonopoly Act itself.77 The immunity from criminal sanctions is based on a policy statement issued by the JFTC. Pursuant to that statement, the JFTC will not seek criminal charges against the first leniency applicant in the pre-investigation stage.78 The statement does not mention subsequent leniency applicants in the pre-investigation stage and the JFTC therefore retains the discretion to bring criminal charges against these applicants. Moreover, leniency does not offer any protection from civil liability and successful leniency applicants can be sued for single damages and pre-judgment interest, just like any other cartel member. 3.1.6 Appeals Companies that are subject to a cease-and-desist order or a surcharge order can appeal the order and have it reviewed. This review procedure was fundamentally revised by an amendment to the Antimonopoly Act that was passed in 2013 and entered into force on 1 April 2015. The amendment abolished the JFTC’s internal administrative appeal system, known in Japanese as the shinpan seido. Under that system, companies first had to go through a trial-like procedure before a panel within the JFTC. This procedure would lead to a decision by the JFTC confirming or annulling the initial cease-and-desist or surcharge order. The JFTC’s decision could then be appealed to the Tokyo High Court but that court was bound by the substantial evidence rule, meaning it had to accept the JFTC’s findings of fact if the JFTC could reasonably have made its factual findings based on the available evidence. Businesses had criticized this procedure, arguing that the JFTC’s panel was biased in 76

Ibid Art 7-2(12). Ibid Art 7-2(10) to Art 7-2(12). 78 Japan Fair Trade Commission, Dokusenkinshihō ihan ni taisuru keiji kokuhatsu oyobi hansoku jiken no chōsa ni kansuru kōseitorihikiiinkai no hōshin [Policy on Criminal Accusation and Investigation of Violations of the Antimonopoly Act] (7 October 2005, as amended in 2009). 77

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 20 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 435

favor of the JFTC, and that the substantial evidence rule before the Tokyo High Court prevented adequate judicial review. Under the new procedure, companies can appeal cease-and-desist and surcharge orders directly to the Tokyo District Court.79 The court will not be bound by the substantial evidence rule. After the Tokyo District Court, a further appeal is possible to the Tokyo High Court and ultimately the Supreme Court. 3.1.7 The JFTC’s enforcement record Bid-rigging has long been one of the principal targets of the JFTC. In 2013, for instance, out of the 18 cases in which the JFTC took formal measures, nine concerned bid-rigging.80 Eight cases concerned pricefixing cartels other than bid-rigging and the remaining case concerned an unfair trade practice, namely abuse of a superior bargaining position. In 2013 no private monopolization cases were brought and such cases have generally been rare.81 The JFTC also spends significant resources on the more fairnessoriented provisions of the Antimonopoly Act. Small businesses often complain about low prices by large players in the retail markets for liquor, gas and home appliances, alleging that this constitutes an unfair trade practice, namely selling at unjustly low prices.82 In response, each

79

Antimonopoly Act (n 2) (as amended in 2013) Art 85 (exclusive jurisdiction for the Tokyo District Court). 80 Japan Fair Trade Commission, Heisei 25 nendo ni okeru dokusen kinshi hō ihan jiken no shori jōtai ni tsuite [Enforcement of the Antimonopoly Act in 2013] 28 May 2014, at 1 (the figures relate to the Japanese fiscal year, which runs from 1 April to 31 March; cases in which formal measures were taken are cases in which either a cease-and-desist order or a surcharge order were imposed). 81 Japan Fair Trade Commission, Jimusōchō teireikaiken kiroku [Minutes of the Secretary General’s Regular Press Conference] – Sūji de miru kōsei torihiki iinkai no rekishi ni tsuite [The History of the Japan Fair Trade Commission in Numbers] 1 (16 July 2014) at (only one private monopolization case between 2007 and 2013). 82 Antimonopoly Act (n 2) Art 2(9)(iii): prohibition laid down in the Antimonopoly Act subject to penalties and which can be applicable when the sales price is ‘excessively below the cost required for the supply,’ which is interpreted as meaning below average avoidable cost; Designation of Unfair Trade Practices, No 6: prohibition laid down in a notice of the Japan Fair Trade Commission, not subject to penalties and which can be applicable when the sales price is ‘unduly low,’ which is interpreted as meaning below average total (variable and fixed) cost.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 21 /

Date: 13/10

JOBNAME: Duns PAGE: 24 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

436 Comparative competition law

year, the JFTC issues more than a thousand cautions (chūi) to those price-cutters.83 These do not, however, result in any sanction. The JFTC is also actively enforcing the Subcontract Act, which was discussed in the section on abuse of a superior bargaining position. In 2013 the JFTC issued ten recommendation decisions against companies that had violated the Act, for example by unilaterally reducing payments to subcontractors.84 The recommendation decisions ensure that the amounts due are paid by the infringers, but do not impose a sanction. Each year the JFTC also conducts a survey to check compliance with the Subcontract Act, covering several hundreds of thousands of businesses. 3.2 Private Enforcement Private parties contribute to the enforcement of competition law in various ways. They can trigger public enforcement by providing information to the JFTC, for instance by filing a formal complaint.85 But they also enforce competition law more directly by raising violations of competition law in civil litigation and seeking nullity, damages, or injunctive relief. Contracts or contractual clauses that violate the Antimonopoly Act are not automatically null and void,86 but courts have frequently found that such contracts violate public policy and are therefore null and void based on the Civil Code.87 Violations of the Antimonopoly Act can also form the basis of an action for damages or injunctive relief. For a long time such actions were rare but since the 1990s the number of cases has increased significantly, as shown in Figure 15.2, with more than 200 cases filed in the past two decades. Many damages actions have been filed in the wake of bidrigging uncovered by the JFTC. Initially, local governments who fell victim to bid-rigging for public works were hesitant to seek damages but local residents sued bid-riggers on behalf of their local government through so-called residents’ lawsuits.88 Damages actions in bid-rigging

83

Designation of Unfair Trade Practices, 8, 13. Japan Fair Trade Commission, Heisei 25 nendo nenji hōkoku1 [Annual Report 2013] 14 (October 2014). 85 Antimonopoly Act (n 2) Art 45(1). 86 See Saikō Saibansho [Sup Ct] 20 June 1977, Shōwa 48 (o) no 1113, 31(4) Minshū 449, 459 (KK Miyagawa v Gifu Shōkō Shinyōkumiai). 87 Minpō [Civil Code] Art 90. 88 Simon Vande Walle, Private Antitrust Litigation in the European Union and Japan (Maklu Publishers 2013) 131. 84

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 437

cases have resulted in significant damages awards and, in the aggregate, tens of billions of yen (hundreds of millions of US dollars) have been recovered by plaintiffs. Since 2001 the Antimonopoly Act has also allowed private parties to seek injunctive relief against unfair trade practices.89 In a recent case, for instance, independent taxi drivers were being excluded from the lucrative taxi stand in front of a train station in Kobe. They sought and obtained an injunction against the taxi company that was blocking their entry into the line-up.90 Competition law violations have also been invoked in derivative suits, filed by activist shareholders alleging that directors were negligent in failing to prevent violations from occurring or in filing a timely leniency application.91 24 22 20 18 16 14 12 10 8 6 4 2 0

71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 00 01 02 03 04 05 06 07 08 09 10

Source: Simon Vande Walle, Private Antitrust Litigation in the European Union and Japan (Maklu Publishers 2013) 131 (updated with more recent data).

Figure 15.2 Evolution of private enforcement: The number of damages actions and actions for injunctive relief filed per year on the basis of violations of competition law

89

Antimonopoly Act (n 2) Art 24. Osaka Kōtō Saibansho [Osaka High Ct] 31 October 2014, Heisei 26 (ne) no 471 (injunction awarded based on the unfair trade practice of ‘interference with a competitor’s transactions,’ Designation of Unfair Trade Practices, No 14.; the court also awarded damages). 91 Vande Walle (n 88) 123–125. 90

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 23 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

438 Comparative competition law

3.3 Criminal Enforcement Individuals and corporations that engage in unreasonable restraints of trade or private monopolization can be criminally prosecuted and be subjected to criminal fines and, where individuals are involved, also imprisonment.92 The JFTC has formally announced that it will specifically target two types of cases: (1) those involving pernicious infringements such as cartels, bid-rigging and private monopolization that have a wide impact on people’s lives, and (2) those involving repeat offenders, i.e. companies that had already been punished administratively in the past but committed an infringement again.93 In reality, all criminal cases so far have involved price-fixing and bid-rigging, not private monopolization. The JFTC has the exclusive right to initiate criminal prosecutions,94 but it has to work in tandem with the Prosecutor’s Office.95 The JFTC files the accusation (kokuhatsu) with the prosecutor, but the prosecutor then formally brings the criminal case before the court (kiso). The track record of the JFTC and the Prosecutor’s Office is stellar: almost all prosecutions have resulted in convictions, although courts have always imposed suspended prison sentences so no individual has actually served time in jail. The high conviction rate is not specific to competition law but a general characteristic of the criminal justice system in Japan. The conviction rate for general criminal cases brought to trial is over 99 percent.96 Foreign observers sometimes conclude that this must mean that the rights of defense are flouted but, in fact, the high conviction rate mainly results from the fact that prosecutors only bring those cases for which they are almost entirely certain to secure a conviction. The drawback is of course that few cases are brought. On average, around one criminal case per year is brought in relation to competition law violations.

92

Antimonopoly Act (n 2) Art 89 (individuals) and Art 95 (corporations). Japan Fair Trade Commission, Dokusenkinshihō ihan ni taisuru keiji kokuhatsu oyobi hansoku jiken no chōsa ni kansuru kōseitorihikiiinkai no hōshin [Policy on Criminal Accusation and Investigation of Violations of the Antimonopoly Act] (7 October 2005, as amended in 2009). 94 Antimonopoly Act (n 2) Art 96(1). 95 Antimonopoly Act (n 2) Art 74. 96 Hiroshi Oda, Japanese Law (3rd edn, Oxford University Press 2009) 439: ‘the acquittal rate at trial is less than 0.1 per cent.’ 93

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 24 /

Date: 13/10

JOBNAME: Duns PAGE: 27 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 439

3.4 Mergers: Ex Ante Control Mergers that meet certain thresholds must be notified to the JFTC prior to being implemented. The JFTC first assesses the legality of the merger in what is called a phase one review, which can last up to 30 days. If the merger raises concerns, the JFTC opens a phase two review and will request additional information from the parties. Once the JFTC has received all the information it requested, it has up to 90 days to conclude the review. Although the notification system is the main channel of enforcement, in fact the JFTC has the right to prohibit any merger that may substantially restrain competition, even when it is not notified. In 2008, for example, mining giant BHP Billiton announced its intention to acquire Rio Tinto. At the time, stock acquisitions did not have to be notified ex ante, which was probably one of the reasons why BHP Billiton had not notified the transaction. The JFTC nonetheless investigated the matter on its own motion and started assessing whether the acquisition would substantially restrain competition in Japan. Ultimately, the acquisition was abandoned before the JFTC reached a conclusion.97

4. APPLICATION OF COMPETITION LAW TO CROSS-BORDER CASES Foreign companies long stayed clear of any fines imposed by the JFTC but this has changed recently, leading to debates about the extent of the JFTC’s jurisdiction. In Marine Hose, a cartel involving foreign companies, the foreign companies were subject to a cease-and-desist order but not fined, because they were not selling in Japan.98 The case concerned a market-sharing agreement between Japanese and European companies selling marine hoses and the European companies had agreed not to sell to Japanese customers.

97

Japan Fair Trade Commission, ‘Press Release: The JFTC Closes Its Investigation into BHP Billiton’s Proposed Acquisition of Rio Tinto’s Shares’ (Japan Fair Trade Commission Website, 3 December 2008) . 98 Japan Fair Trade Commission, Cease-and-desist order and Surcharge Order of 20 February 2008, Heisei 20 (so) 2 and Heisei 20 (nō) 10, 54 Shinketsushū 512, 623.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 28 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

440 Comparative competition law

In the Cathode Ray Tubes cartel case99 11 manufacturers had fixed the prices of cathode ray tubes, the main component of traditional television sets. Ten of the 11 cartel members were foreign companies and six of them were fined by the JFTC. Remarkably, the cartel members were not selling directly to the Japanese market but to companies in South-East Asia. The link with Japan was that the South-East Asian companies that were buying the tubes were subsidiaries of Japanese companies and those Japanese parent companies had negotiated with the tube manufacturers. Whether such a tenuous link is sufficient for the JFTC to exercise jurisdiction is the subject of significant debate. Several members of the cartel have appealed the JFTC decisions and at the time of writing these appeals are pending before the courts.

5. AWARENESS AND ROLE OF COMPETITION LAW It is often said that competition law plays a less important role in Japanese society than in the US or Europe. Most observers seem to agree that, until the 1990s, the Antimonopoly Act was not considered very important and enforcement was weak. Opinions diverge as to why this was so. Some link the limited role of competition law with the idea that Japanese or Asian societies place greater value on cooperation and harmony and less on cut-throat competition.100 Others emphasize the strong influence of business interests on the policies of the Liberal

99 Japan Fair Trade Commission, Decisions of 22 May 2015, Heisei 22 (han) no 2–5, no 6, and no 7. These three decisions were rendered after appeals by some addressees of the JFTC’s prior cease-and-desist and surcharge orders: Japan Fair Trade Commission, Surcharge orders of 7 October 2009, Heisei 21 (nō) nos 62–66, 56(2) Shinketsushū 173 (fining MT Pictures Display (Malaysia), PT MT Picture Display Indonesia, MT Picture Display (Thailand), Samsung SDI (Malaysia) and LG Philips Displays Korea); Japan Fair Trade Commission, Surcharge order of 12 February 2010, Heisei 22 (nō) no 24 (fining PT LP Displays Indonesia). After service of process could not be effected on Samsung SDI (Malaysia), the Japan Fair Trade Commission reissued a surcharge order against that addressee and effected service of process through public notification. Japan Fair Trade Commission, Surcharge Order of 12 February 2010, Heisei 22 (nō) no 23. 100 Several sources that offer such an explanation are reviewed in Douglas E Rosenthal and Mitsuo Matsushita, ‘Competition in Japan and the West: Can the Approaches Be Reconciled?’ in Edward M Graham and J David Richardson (eds), Global Competition Policy (Institute for International Economics 1997) 313, 313–314.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in Japan 441

Democratic Party, which was in power for most of the postwar period.101 Still others point to the influence of bureaucrats, particularly those of Japan’s Ministry of International Trade and Industry, who favored industrial policy over competition policy and supported cartels.102 There is no empirical basis to evaluate these claims and it is likely impossible to come to any firm conclusions. It is, however, clear that competition law’s standing has risen considerably in the past two decades. Hence, the view that competition law plays only a minor role in regulating Japanese business behavior no longer seems warranted. There is now greater awareness of the law’s principles and goals, both in the legal community and in society at large. This awareness is illustrated by the fact that the general press regularly covers the JFTC’s enforcement actions and is generally positively predisposed towards those actions. Competition law or ‘economic law’ as it is often referred to in Japan is also one of the elective topics of the renewed bar exam that was introduced in 2004. This ensures that virtually all law schools offer a course in competition law and that, each year, a small but significant segment of law students enters the legal profession with a decent knowledge of competition law.

6. CONCLUSION Although Japan’s Antimonopoly Act, enacted in 1947, is one of the oldest competition laws in the world, for a long period of time it was of little relevance for society and legal practice. In the past two decades, however, tremendous change has taken place. The JFTC and the courts have gradually fleshed out the widely formulated prohibitions on unreasonable restraints of trade and private monopolization. In very broad strokes, the resulting rules serve a function that is similar to the rules on horizontal anti-competitive agreements and abuse of dominance in the EU and US. The Antimonopoly Act also contains a ban on unfair trade practices, which reflects Japanese competition law’s concern for fair competition.

101

J Mark Ramseyer, ‘The Antitrust Pork Barrel in Japan’ (1992) Antitrust 40, 41; see also Michael L Beeman, Public Policy and Economic Competition in Japan: Change and Continuity in Antimonopoly Policy, 1973–1995 (Routledge 2002) 8–10: combining this explanation with other elements. 102 See, for example, Chalmers Johnson, MITI and the Japanese Miracle (Stanford University Press 1982).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 27 /

Date: 8/9

JOBNAME: Duns PAGE: 30 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

442 Comparative competition law

The most drastic change, however, has been the increase in the level and effectiveness of enforcement. The fines imposed by the JFTC are now a multiple of the fines imposed a decade ago. The amounts recovered through damages actions have also increased. In spite of this increase, however, sanctions are still at levels significantly below the sanctions imposed through fines in the EU or the combination of private damages actions and fines in the US.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 18_Chapter15

/Pg. Position: 28 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

16. Competition law in China Wentong Zheng

1. COMPETITION LAW IN CHINA: AN OVERVIEW Formally a socialist country that is still transitioning to a market economy, China has an economic and political history distinct from that of the countries where modern competition law originated. Prior to the commencement of economic reforms in the late 1970s, China’s economy operated under a centrally planned system, whereby the government’s planning agencies made production plans and set prices for the country’s vast number of state-owned enterprises (SOEs). Since the late 1970s, China has adopted a series of market-oriented reform measures, under which the government abolished or relaxed price controls, promoted foreign and private investment, and converted SOEs to commercial entities.1 These reform measures catapulted China onto the world stage, making China the second largest economy, next only to the United States (US), as of 2010. Concurrent with these sea changes in China’s economy, the Chinese government itself has seen its role redefined, dramatically so in certain cases. A number of government ministries overseeing specific economic sectors in the planned economy era were abolished or streamlined, and in their place emerged a regulatory structure aimed at separating the government’s ownership role from its regulatory role.2 Despite these liberalization efforts, the state still retains a significant degree of presence in China’s economy. The state still maintains price control over key products and services,3 owns the largest SOEs, which 1 For detailed discussions of China’s economic reforms, see Bruce M Owen, Su Sun and Wentong Zheng, ‘Antitrust in China: The Problem of Incentive Compatibility’ (2005) 1 Journal of Competition Law and Economics 123, 127–28; Wentong Zheng, ‘Transplanting Antitrust in China: Economic Transition, Market Structure, and State Control’ (2010) 31 University of Pennsylvania Journal of International Law 643, 652–71 (hereafter Zheng, ‘Transplanting Antitrust’). 2 Zheng, ‘Transplanting Antitrust’ (n 1) 667–70. 3 Ibid 654.

443

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

444 Comparative competition law

account for the majority of output in key sectors,4 and maintains strict market entry restrictions against private enterprises in key sectors.5 The state also retains the ability to subject private enterprises to state control through the so-called industrial associations or chambers of commerce that are converted from government agencies and essentially exercise government functions.6 It is against this backdrop that competition and competition law began to take hold in China. In 1993 China adopted the Anti-Unfair Competition Law (AUCL), China’s first legislation regulating competition.7 In addition to prohibiting practices such as bribery, deceptive advertising, coercive sales, and appropriation of trade secrets, the AUCL also outlaws tie-in sales, price fixing, and bid rigging – practices that are typically considered antitrust violations in a market economy.8 In 1997 China adopted another law aimed at regulating firms’ competitive behavior: the Price Law.9 The Price Law prohibits certain ‘improper pricing practices’ that are typically considered antitrust violations in a market economy, including price fixing, predatory pricing, and price discrimination.10 The Price Law also allows the government to maintain price control or set guidance prices for certain key products or services.11 The Procurement and Bidding Law, adopted in 1999, also prohibits bid rigging.12 4

In 2006 SOEs accounted for almost all of the production of petroleum, natural gas, and ethylene, provided all of the basic telecommunication services, generated approximately 55% of electricity, and flew about 82% of passengers and cargo through the country’s air transportation system. See ibid 665–66. 5 Dominated by SOEs, these sectors include public utilities and infrastructure, social services, financial services, national defense, electricity, telecommunications, railroads, airlines, and petroleum: ibid 660–61. 6 Ibid 669–70. 7 ‘Anti-Unfair Competition Law of the People’s Republic of China’ (promulgated by the Standing Committee of the National People’s Congress on 2 September 1993 and effective 1 December 1993). 8 See Bruce M Owen, Su Sun and Wentong Zheng, ‘China’s Competition Policy Reforms: The Antimonopoly Law and Beyond’ (2008) 75 Antitrust Law Journal 231, 233. 9 See ‘The Price Law of the People’s Republic of China’ (promulgated by the Standing Committee of the National People’s Congress on 29 December 1997 and effective 1 May 1998) (hereafter Price Law). 10 Ibid Art 14. 11 Ibid Art 18. 12 See ‘Procurement and Bidding Law’ (promulgated by the Standing Committee of the National People’s Congress on 30 August 1999 and effective 1 January 2000) Art 32.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 2 /

Date: 8/9

JOBNAME: Duns PAGE: 3 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 445

In addition, between 1993 and 2007 various government ministries or agencies adopted several regulations that have implications for competition. In 1993 the State Administration of Industry and Commerce (SAIC) issued Certain Provisions on Prohibiting Public Utility Enterprises from Committing Restrictive Acts Against Competition to rein in monopolistic abuses by public utilities.13 In April 2001, in response to widespread protectionist policies of local governments, the State Council issued the Rules on Prohibiting Regional Blockades in Market Economic Activities.14 In 2003 the National Development and Reform Commission (NDRC) issued the Provisional Rules on the Prohibition of Monopoly Pricing, which implements the Price Law and prohibits a variety of conduct such as horizontal agreements on price or production volume, resale price maintenance, predatory pricing, and price discrimination.15 In 2003, four government agencies jointly issued the Provisional Rules on Mergers and Acquisitions of Domestic Enterprises by Foreign Investors, which created China’s first merger review regime.16 In 2006, six government agencies jointly issued the Rules on Mergers and Acquisitions of Domestic Enterprises by Foreign Investors (M&A Rules).17 The M&A Rules inherited most of the 2003 provisional rules and required that mergers exceeding notification thresholds be notified to the Ministry of Commerce (MOFCOM) and SAIC. The merger review regime under the 2003 and 2006 rules, however, only applied to mergers and acquisitions of domestic firms by foreign firms. 13

See ‘Certain Provisions on Prohibiting Public Utility Enterprises from Committing Restrictive Acts Against Competition’ (promulgated by the State Administration of Industry and Commerce on 24 December 1993). 14 ‘Rules on Prohibiting Regional Blockades in Market Economic Activities’ (promulgated by the State Council 29 April 2001 and effective 29 April 2001). 15 ‘Interim Provisions on Preventing Acts of Price Monopoly’, Order [2003] No 3 (adopted by the State Planning and Development Commission on 18 June 2003 and effective 1 November 2003). 16 ‘Provisional Regulations Regarding Mergers and Acquisitions of Domestic Enterprises by Foreign Investors’ (promulgated by Ministry of Foreign Trade and Economic Cooperation, State Administration of Taxation, State Administration of Industry and Commerce, and State Administration of Foreign Exchanges on 13 March 2003 and effective 12 April 2003). 17 ‘Rules on Acquisition of Domestic Enterprises by Foreign Investors’ (promulgated by Ministry of Commerce, State-Owned Assets Supervision and Administration Commission, State Administration of Taxation, State Administration of Industry and Commerce, China Securities Regulatory Commission, and State Administration of Foreign Exchanges on 8 August 2006 and effective 8 September 2006).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 3 /

Date: 8/9

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

446 Comparative competition law

These competition-related laws and regulations were adopted on an ad hoc basis and were enforced by multiple agencies, which reduced their effectiveness. Efforts to draft a comprehensive competition law, or the Antimonopoly Law (AML) as the law would later be called, started in the late 1980s, and the AML was officially placed on the government’s legislative agenda in 1994.18 The incentives for the government to use competition law to promote market competition and to rein in abusive state power, however, had not been strong.19 The drafting process for the AML languished for a decade until 2004, when the draft law suddenly gained momentum amid press reports of anti-competitive conduct by foreign firms in China.20 Even after a consensus appeared to have been forged on the necessity of the AML, its final passage was still not guaranteed because of debates on whether and how to subject the state itself and SOEs to the AML. Particularly, debates on whether to include the so-called ‘administrative monopolies,’ i.e., anti-competitive conduct by government agencies, as prohibited conduct under the AML prolonged the drafting of the law. The first several drafts of the AML released since 2002 all prohibited administrative monopolies. In 2005, however, the State Council deleted the entire chapter of the draft AML dealing with administrative monopolies and only kept a declaratory statement prohibiting administrative monopolies in principle. In 2006 the State Council approved a draft that did not contain a chapter on administrative monopolies. Several weeks later, however, the chapter on administrative monopolies was reinserted into a draft submitted by the State Council to the National People’s Congress Standing Committee.21 The final text of the AML eventually included a chapter on administrative monopolies, albeit a much more modest version.22 As for SOEs, a clause exempting SOE monopolies from the reach of the AML was once added during the law’s drafting process, but was soon deleted.23 As a compromise, the final version of the AML adopted a watered-down clause that protects the lawful 18

Zheng, ‘Transplanting Antitrust’ (n 1) 717. See David J Gerber, ‘Economics, Law & Institutions: The Shaping of Chinese Competition Law’ (2008) 26 Journal of Law and Policy 271. 20 Zheng, ‘Transplanting Antitrust’ (n 1) 718–19. 21 Owen and others, ‘China’s Competition’ (n 8) 254–55. 22 See Yong Huang, ‘Pursuing the Second Best: The History, Momentum, and Remaining Issues of China’s Anti-Monopoly Law’ (2008) 75 Antitrust Law Journal 117, 131. 23 Ibid 126–27. 19

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 4 /

Date: 13/10

JOBNAME: Duns PAGE: 5 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 447

operations of SOE monopolies but requires them not to abuse their monopoly power.24

2. THE ANTIMONOPOLY LAW: A LEGAL TRANSPLANT WITH CHINESE CHARACTERISTICS On 30 August 2007 the National People’s Congress Standing Committee approved the final draft of the AML, which went into effect approximately a year later on 1 August 2008. While the final text of the AML borrows heavily from competition laws of other jurisdictions, particularly the European Union, the US, and Japan, it also contains many provisions unique to China. Compared to the earlier laws and regulations on competition, the AML makes long strides in providing a comprehensive competition law regime. However, it also suffers from ambiguities and inconsistencies in several key aspects, in part because of the legal transplant nature of the law and in part because of the steep challenges in devising a competition regime for an economy that remains under significant state control. 2.1 General Provisions The general provisions in Chapter 1 of the AML set out the legislative purposes and general principles of the law. Article 1 states: ‘[t]his Law is enacted for purposes of preventing and curbing monopolistic conduct, protecting fair market competition, enhancing economic efficiency, protecting the consumer interests and public interests, and promoting the healthy development of the socialist market economy.’ While the other goals listed in this article are non-controversial, the inclusion of the phrase ‘public interests’ in this article appears to create the possibility of considering goals broader than protecting competition and consumer interests, goals that may include protecting national or business interests.25 The broad goals of the AML are further seen in Article 4, which 24

Article 7 of the AML provides that ‘in SOE-dominated industries concerning the health of national economy and national security, and in industries where state trading is authorized by law, the lawful operations of the undertakings are protected by the state’. Such undertakings, however, ‘shall not cause damages to the interests of consumers by virtue of their dominant or monopolistic positions’. AML Art 7. 25 See Xiaoye Wang, ‘Highlights of China’s New Anti-Monopoly Law’ (2008) 75 Antitrust Law Journal 133, 142–43.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 5 /

Date: 13/10

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

448 Comparative competition law

provides that ‘the State formulates and implements competition rules compatible with the socialist economy, strengthens and perfects macro regulation and control, and completes a unified, open, competitive and orderly market system.’ Article 2 sets the jurisdictional reach of the AML. It provides that the AML applies to ‘all monopolistic conduct in economic activities in the territory of the People’s Republic of China, as well as monopolistic conduct outside of the territory of the People’s Republic of China that eliminates or restricts market competition in China.’ This provision is apparently modeled after the effect doctrine under US antitrust law and expands the reach of the AML to monopolistic conduct outside of China.26 Chapter 1 then lists the categories of conduct prohibited under the AML. Article 3 states: ‘[t]he monopolistic conduct prohibited under this Law includes (1) monopolistic agreements between undertakings, (2) abuse of dominant market position by undertakings, and (3) concentrations among undertakings that have or possibly have the effect of impeding or restricting competition.’ Article 12 defines the term ‘undertakings’ as ‘natural persons, legal persons or other entities that are engaged in the production or distribution of products or services.’ Despite the straightforwardness of these provisions, two other provisions in Chapter 1, Article 5 and Article 6, appear to create ambiguities regarding the scope and nature of the prohibited conduct. Article 5 states: ‘[u]ndertakings may lawfully form concentrations through fair competition and voluntary consolidation to expand operational scale and to increase market competitiveness.’ While this provision allows firms to form concentrations only ‘lawfully,’ it emphasizes firms’ capability of forming concentrations under the AML and appears to create room for flexibility in the enforcement of the AML’s merger provisions. Article 6 states: ‘[u]ndertakings that have a dominant market position are prohibited from abusing that position to eliminate or restrict competition.’ As discussed in detail below, this provision appears to limit the abuse-of-dominance conduct prohibited under the AML to exclusionary conduct, i.e., conduct that eliminates or restricts competition. The AML reaches not only private individuals and entities, but also government agencies. Article 8 states: ‘Government agencies and other organizations that are authorized by law to have governmental functions shall not abuse administrative power to eliminate or restrict competition.’ This provision provides the legal basis for outlawing anti-competitive 26

Columns Design XML Ltd

Ibid 134.

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 449

conduct by government agencies, often referred to in China as ‘administrative monopolies.’ Article 7 concerns the legal status of SOE monopolies. It states: The state protects the lawful operations of undertakings in SOE dominated industries concerning the health of national economy and national security and in industries where state trading is authorized by law. The state supervises and regulates the operations of those undertakings and the prices of their products or services in accordance with law to protect consumer interest and to promote technology advancement.

Article 7 further states, ‘[u]ndertakings in industries referenced above shall conduct their operations lawfully, be honest and trustworthy, accept supervision by the public, and not hurt consumer interest using their dominant or monopoly power.’ The language of Article 7 is ambiguous enough to allow room for different interpretations. If the word ‘lawful’ in the sentence ‘the state protects the lawful operations of undertakings in …’ means lawful under laws other than the AML, then Article 7 effectively grants a blanket exemption for SOE monopolies. However, if the word ‘lawful’ means lawful under laws including the AML, then SOE monopolies are required to abide by the AML in addition to all other laws.27 Assuming that Article 7 does not create an exemption for SOE monopolies, it represents a compromise as to the legal status of SOE monopolies: on one hand, it respects the status quo of SOE monopolies; on the other hand, it provides that the business activities of SOE monopolies should be supervised and abuse of monopoly power should be prevented.28 Finally, Articles 9 and 10 concern the antimonopoly enforcement authorities. Article 9 provides that the State Council will establish an Antimonopoly Commission that will be responsible for conducting competition policy research, issuing antimonopoly guidelines, and coordinating antimonopoly enforcement activities. Article 10 provides that the State Council will designate an Antimonopoly Enforcement Authority to enforce the AML. What emerges under Articles 9 and 10 is a two-tier enforcement regime: the Antimonopoly Commission serves as a liaison and coordinating office, while the day-to-day enforcement of the AML is carried out by the Antimonopoly Enforcement Authority. At the time of the AML’s enactment, it was not clear which agency would be designated as the Antimonopoly Enforcement Agency. It was not until 27 28

Columns Design XML Ltd

Zheng, ‘Transplanting Antitrust’ (n 1) 697–98. Huang (n 22) 127.

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 7 /

Date: 8/9

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

450 Comparative competition law

after the AML went into effect that the Antimonopoly Enforcement Authority was finalized, with three agencies – MOFCOM, NDRC, and SAIC – splitting the responsibilities of the Antimonopoly Enforcement Authority. 2.2 Monopolistic Agreements Chapter 2 of the AML prohibits monopolistic agreements, defined as ‘agreements, decisions, or other concerted actions that eliminate or restrict competition.’29 Two categories of monopolistic agreements fall under this prohibition: horizontal agreements and vertical agreements. Article 13 prohibits horizontal agreements, or agreements between undertakings that compete with one another. Article 13 enumerates the prohibited horizontal agreements as follows: (1) agreements to fix or change prices; (2) agreements to restrict production or sales volumes; (3) agreements to divide market for product sales or raw materials procurement; (4) agreements to restrict the purchase or development of new technologies and new equipment; (5) joint boycott agreements; and (6) other agreements as determined by the antimonopoly enforcement authorities. Article 14 prohibits vertical agreements between undertakings that do not compete with one another. These vertical agreements include: (1) agreements to fix resale prices; (2) agreements to restrict minimum resale prices; and (3) other agreements as determined by the antimonopoly enforcement authorities. It is not entirely clear from the language of Articles 13 and 14 whether and to what extent the competitive effects of agreements will be taken into account in determining the legality of the agreements. On one hand, the prohibitions under Articles 13 and 14 appear to be absolute and are not qualified by considerations of competitive consequences. On the other hand, the term ‘monopolistic agreements’ is defined to only include agreements that eliminate or restrict competition. This ambiguity creates legal limbo for certain forms of conduct that may constitute agreements but may or may not have adverse competitive consequences. The Chinese courts have stepped in and clarified that the competitive consequences of horizontal agreements under Article 13 may be taken into account in determining the legality of the agreements. In January 2012 China’s Supreme People’s Court promulgated a Judicial Interpretation, effective 1 June 2012, that was aimed at clarifying issues arising 29

Columns Design XML Ltd

AML Art 13.

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 451

from civil litigation under the AML.30 Article 7 of the Judicial Interpretation provides that when the monopolistic conduct at issue involves a monopolistic agreement prohibited under Article 13 of the AML, ‘the defendant shoulders the burden to prove that the agreement does not eliminate or restrict competition.’ This indicates that the competitive effects of a horizontal agreement are indeed one factor in deciding its legality under Article 13 of the AML. In December 2012 a court in Guangdong province followed this view, ruling in favor of the defendants in a price-fixing case in part on the ground that the alleged price-fixing agreement did not have significant anti-competitive effects.31 This practice departs from the antitrust jurisprudence in the US and European Union where horizontal agreements of the kind under consideration are considered per se illegal.32 By contrast, the treatment of vertical agreements under Article 14 is more ambiguous. The Supreme People’s Court’s judicial interpretation conspicuously leaves out Article 14, raising questions as to the relevance of the competitive effects of a vertical agreement. A case in point of this ambiguity can be found in the treatment of resale price maintenance (RPM). In major antitrust jurisdictions, an issue regarding the legality of RPM has been whether it should be considered per se illegal. In the US, for example, RPM was initially subject to a per se rule under Dr. Miles Medical Co. v Park & Sons Co.,33 but has been analyzed under the rule of reason since the 2007 Supreme Court case of Leegin Creative Leather Products, Inc. v PSKS Inc.34 In China, prior to the enactment of the AML, minimum RPM (setting a minimum resale price) had been a standard practice in certain industries. Automobile dealerships in China, for example, had been following what was known in China as a ‘4S’ model (sales, service, spare parts, and surveys), a special franchising model in which the franchisee dealerships deal only in products of their franchisors (automobile manufacturers) and the franchisors set minimum 30

‘Guanyu shenli yin longduan xingwei yinfa de minshi jiufen anjian yingyong falv ruogan wenti de guiding’ (Provisions on the Application of the Antimonopoly Law in the Trial of Civil Disputes Arising from Monopolistic Conduct) (hereafter ‘Judicial Interpretation’). 31 See Adrian Emch and Jonathan Liang, ‘Private Antitrust Litigation in China – The Burden of Proof and Its Challenges’ (2013) 1 CPI Antitrust Chronicle 5. 32 Ibid. 33 220 US 373 (1911). 34 551 US 877 (2007).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 9 /

Date: 8/9

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

452 Comparative competition law

resale prices and allocate geographical areas for the franchisees. In July 2008, right before the AML went into effect, Toyota’s joint venture in the southern city of Guangzhou announced that it would abandon its minimum RPM and geographical limitation policies for its dealerships across China in order to comply with the AML prohibitions of the practices.35 In 2012, however, a court in Shanghai dismissed allegations that Johnson & Johnson Medical (China) Ltd and its Shanghai branch had set a minimum resale price in violation of the AML.36 The court found that the distribution agreement between the defendants and the plaintiff did contain an RPM clause, but held that the mere existence of an RPM arrangement was not sufficient for a finding of a monopolistic agreement under Article 14 of the AML.37 Citing the definition of monopolistic agreements under Article 13, the court held that the competitive effects of the RPM arrangement must be assessed to determine the legality of the practice.38 Article 15 provides an exemption mechanism whereby Articles 13 and 14 would be inapplicable to monopolistic agreements that serve the following purposes: (1) improving technology and researching and developing new products; (2) standardizing product specifications with a view to improving product quality, lowering costs, and enhancing efficiency; (3) improving the operating efficiency of medium- and small-sized undertakings with a view to enhancing their competitiveness; (4) promoting public interests such as conserving energy, protecting the environment, and disaster relief; (5) alleviating serious sales declines or excess capacities during economic recessions; (6) safeguarding legitimate interests in foreign trade and economic cooperation; and (7) other purposes stipulated by law and the State Council. Article 15 further provides that for monopolistic agreements falling under categories (1) to (5), undertakings seeking exemption from the AML have to prove that the agreements do not significantly eliminate or restrict competition and that consumers will share the benefits of the agreements. 35 ‘Guangzhou Fengtian fangkai jiage xianzhi; peihe fanlongduanfa’ (‘Toyota Guangzhou Eases Control on Retail Prices to Comply with the Antimonopoly Law’ 3 (24 July 2008) . 36 See Jessica Hua Su, ‘The Shanghai Court’s Position on Resale Price Maintenance in the J&J Vertical Price-Fixing Litigation’ (Kluwer Competition Law Blog, 4 June 2012) . 37 Ibid. 38 Ibid.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 453

The exemptions provided under Article 15 open the door to taking into account industrial, macroeconomic, and other policies in AML enforcement. Particularly, Article 15 appears to create a general exemption for crisis cartels and export cartels. The general exemption for crisis cartels departs from practices in the US and the European Union, where crisis cartels enjoy no or only ad hoc exemptions.39 The general exemption for crisis cartels gives AML regulators leeway to not enforce the AML or enforce the AML leniently when monopolistic agreements are used to address excess capacities, which, not coincidentally, are a widespread phenomenon in China.40 Excess capacities led to enormous downward pressures on prices in China’s export sector and prompted the imposition of antidumping duties and other trade remedy measures on Chinese products overseas. As a response, the Chinese government tried to stabilize export prices through export cartels implemented by industrial associations in the export sector.41 These export cartels appear to enjoy absolute exemption from the AML, as the requirements for exemptions on other grounds – that the agreements do not significantly eliminate or restrict competition and that consumers share benefits of the agreements – do not apply to export cartels. Speaking of industrial associations, Article 16 in Chapter 2 of the AML provides that ‘industrial associations shall not organize undertakings in their respective industries to engage in monopolistic conduct prohibited under this Chapter.’ The industrial associations referenced 39

In the United States, the US Supreme Court held in 1940 that an industry crisis was not relevant to antitrust analysis and that cartels could not be justified on the ground that they diminished competitive evils: United States v SoconyVacuum Oil Co., 310 US 150, 221 (1940). In Europe, the European Commission approved crisis cartels organized in the synthetic fiber and steel industries in the late 1970s and early 1980s as a response to excess capacities caused by recessions and distortive state aids: Zheng, ‘Transplanting Antitrust’ (n 1) 674 n 125. 40 Many of China’s industries suffer from chronic excess capacities, driven by distortive roles of the government, particularly local governments, in the investment process and the lack of credible exit mechanisms for failing stateowned enterprises. For more discussions: Zheng, ‘Transplanting Antitrust’ (n 1) 675–82. 41 Two Chinese export cartels, one in the vitamin C industry and one in the magnesite industry, were targeted in private antitrust lawsuits filed in US courts for price-fixing violations: ‘Courts Grapple With Novel Legal Issues Relating to Chinese Companies Sued in U.S. for Antitrust Violations’ (Paul Hastings StayCurrent, January 2009) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

454 Comparative competition law

here, also known as chambers of commerce in China, are quasigovernment organizations converted from government ministries that were disbanded during government reforms since the 1990s.42 Staffed by former government officials, industrial associations in China essentially have the same organizational structures and functions as their government predecessors.43 Article 16 of the AML prohibits industrial associations from facilitating conduct by individual undertakings that would violate Chapter 2. A provision in Chapter 1, however, creates ambiguity regarding the scope of permissible activities by industrial associations. Article 11 in Chapter 1 states: ‘[i]ndustrial associations shall strengthen industrial self-discipline, promote lawful competition by undertakings in their respective industries, and maintain the order of market competition.’ The phrase ‘industrial self-discipline’ here refers to a practice that was authorized by the government in 1998 and was once very common during the late 1990s and early 2000s.44 Under this practice, industrial associations organized industry self-discipline agreements as to price or production volume as a response to ‘excessive competition’ or ‘malignant competition’ resulting from massive excess capacities in China’s industries.45 These industrial self-discipline agreements are essentially government-sanctioned cartel agreements aimed at reinstating some sort of price control in industries with widespread structural distortions.46 The coexistence of Article 16, which prohibits industrial associations from facilitating monopolistic agreements, and Article 11, which requires industrial associations to engage in activities that essentially amount to monopolistic agreements, is another example of the ambiguities of the AML.

42

Zheng, ‘Transplanting Antitrust’ (n 1) 669. Ibid. 44 In a directive issued in August 1998, the State Economic and Trade Commission required the implementation of ‘industrial self-discipline prices’ in certain industries: ‘Guanyu bufen gongye chanpin shixing hangye zilü jia de yijian’ (‘Opinion on the Implementation of Industrial Self-Discipline Prices for Certain Industrial Products’) 17 August 1998. The SETC directive required industrial associations to determine, in consultation with government authorities, ‘industrial self-discipline prices’ based on social average costs of production. 45 For detailed discussions on ‘excessive competition’ in China, see Zheng, ‘Transplanting Antitrust’ (n 1) 682–85. 46 See ibid 686–92. 43

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 12 /

Date: 8/9

JOBNAME: Duns PAGE: 13 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 455

2.3 Abuse of Dominance In Chapter 3, the AML prohibits undertakings with a dominant market position from abusing their market dominance. Article 17 in Chapter 3 provides a list of conduct that would constitute abuse of dominance: (1) selling goods at unfairly high prices or buying goods at unfairly low prices; (2) selling goods at below-cost prices without justifications; (3) refusal to deal with transaction counterparties without justifications; (4) restricting transaction counterparties to only deal with the undertaking or undertakings designated by the undertaking without justifications; (5) tying the sale of goods or imposing other unreasonable conditions on a transaction without justifications; (6) discriminatory treatment of similarly situated transaction counterparties with respect to price or other transaction terms without justifications; and (7) other conduct that constitutes abuse of dominant market position as determined by antimonopoly enforcement authorities. This list of conduct shows a striking resemblance to conduct that is generally considered abuse of dominance under the competition laws of the European Union and the US. The conduct specified in Article 17(1) – selling products at unfairly high prices or buying products at unfairly low prices – mirrors the conduct outlawed under Article 102(a) of the TFEU. The conduct specified in Article 17(2)–17(6) corresponds to what are known in the US and the European Union as predatory pricing, refusal to deal, exclusive dealing, tying, and price discrimination, respectively. Article 17, however, is not entirely clear as to the nature of the abuse-of-dominance conduct it prohibits. There are two types of abuseof-dominance conduct: exclusionary and exploitative. Exclusionary conduct excludes competitors from the market and creates or maintains the dominant firm’s market power, whereas exploitative conduct exploits other market participants without directly affecting the structure of the market.47 It is clear that one type of conduct prohibited under Article 17, excessive pricing, is exploitative. But the other types of conduct prohibited under Article 17 could be either exclusionary or exploitative, or both. For example, refusals to deal, which are prohibited under Article 17(3), could be exclusionary if directed at competitors, but could be exploitative if directed at consumers. Article 6 in Chapter 1 provides that ‘undertakings that have a dominant market position are prohibited from abusing that position to eliminate or restrict competition.’ While Article 6 appears 47

See Russell Pittman and Maria Tineo, ‘Abuse of Dominance Enforcement Under Latin American Competition Laws’ in Philip Marsden (ed), Handbook of Research in Trans-Atlantic Antitrust (Edward Elgar 2006) 325.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

456 Comparative competition law

to limit abuse-of-dominance conduct prohibited under Chapter 3 to exclusionary conduct, it is not clear to what extent Article 6 can be read to limit the reach of Chapter 3, particularly given that the prohibitions under Chapter 3 include at least one type of exploitative conduct, excess pricing. The question of whether Article 17 reaches only exclusionary conduct or both exclusionary and exploitative conduct is not purely academic. Since the AML has come into effect, plaintiffs have filed several lawsuits in Chinese courts against abuse-of-dominance conduct that was all said to be exploitative in nature. One lawsuit was filed by a business owner against a branch of China Construction Bank in Chongqing for refusing to honor the plaintiff’s requests to withdraw funds after the plaintiff refused to pay an account management fee charged by the defendant.48 Another lawsuit was filed against China Netcom, one of China’s major telecommunications carriers, by one of its customers who alleged that it violated the AML’s prohibition of discriminatory treatment under Article 17(6) by offering discount packages only to customers who have certificates of residence from Beijing.49 Yet another lawsuit was filed against China Mobile, China’s largest mobile carrier, by one of its customers for discriminatory treatment because it imposed a monthly fee on users of its global roaming services but not on other users.50 Presumably, the various forms of conduct involved in the above-listed lawsuits exploited consumers, but had no impact on market competition. Despite the fact that Article 6 of the AML refers to exclusionary conduct only, all three lawsuits were accepted by the courts for adjudication. The lawsuits against China Construction Bank and China Mobile were eventually settled, with the disposition of the lawsuit against China Netcom

48 See Junhua Hu, ‘Yinhangye fanlongduan diyi an: zhongguo jianshe yinhang qiangzhi shoufei cheng beigao’ (First Antimonopoly Lawsuit in Banking Industry: China Construction Bank Sued for Mandatory Fees) (Meiri Jingji Xinwen (Daily Economic News) 12 September 2008) . 49 See Xiaofang Ma, ‘Lüshi su Beijing Wangtong weifan fanlongduanfa zuo zhengshi li’an’ (Court Officially Accepts Antimonopoly Lawsuit Against Beijing China Netcom by Attorney) (Diyi Caijing Ribao (First Business Daily) 17 September 2008) . 50 See ‘Fayuan shouli zhongguo yidong beisu longduan an’ (Court Accepts Antimonopoly Lawsuit Against China Mobile) (Caijing Net (Business and Finance Net) 13 March 2009) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 457

unclear.51 In short, the uncertainty surrounding the question of whether Article 17 reaches exploitative conduct has not been resolved. The AML does not provide a definition of dominant market position. However, in Article 18, the AML provides a list of factors that are relevant for determining the market dominance of an undertaking: (1) the undertaking’s market share in the relevant market and the competitiveness of the relevant market; (2) the undertaking’s abilities to control the product markets and the raw materials procurement markets; (3) the undertaking’s financial and technological capabilities; (4) the degree of dependence of other undertakings on the undertaking; (5) the ease of entry into the relevant market by other undertakings; and (6) other factors that are relevant for determining the market dominance of the undertaking. Article 19 in Chapter 3 further provides for a rebuttable presumption of market dominance based on market shares of the undertakings. According to Article 19, market dominance can be presumed if one undertaking has a market share of at least one-half of the relevant market, if two undertakings collectively have a market share of at least two-thirds of the relevant market, or if three undertakings collectively have a market share of at least three-fourths of the relevant market. Market dominance will not be presumed, however, for any undertaking whose market share does not exceed one-tenth of the relevant market. Undertakings that are presumed to have market dominance can rebut that presumption with evidence to the contrary.52 2.4 Merger Review Chapter 4 of the AML lays out the basic framework of a merger review regime governing undertaking concentrations. Article 20 defines undertaking concentrations as (1) mergers between undertakings; (2) purchases of equities or assets that confer control rights; or (3) acquisition of control rights or influences through means such as contracts. Article 21 requires undertaking concentrations that meet notification thresholds to be pre-notified to the merger review authority. Article 23 sets forth the documents that need to be submitted to the merger review authority, including an application for concentration, an analysis of the competitive impact of the concentration, the underlying concentration agreement, the audited financial reports of the parties to the concentration for the preceding accounting year, and other documents that may be required by 51

For detailed discussions of the three lawsuits, see Zheng, ‘Transplanting Antitrust’ (n 1) 698–700. 52 See AML Art 19.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

458 Comparative competition law

the merger review authority. Article 24 requires applicants who submit incomplete documents to complete their file within the deadline set by the merger review authority; otherwise the concentration application would be treated as if it were never filed. Article 25 requires the merger review authority to make a preliminary decision as to whether to approve the merger within 30 days of the receipt of notification documents that comply with the requirements set forth in Article 23. Article 26 allows the merger review authority to initiate an extended review, but the extended review must be completed within 90 days, subject to a possible extension by another 60 days. Article 29 allows the merger review authority to veto or impose restrictions on concentrations, and requires the merger review authority to publish those decisions in a timely manner. Most importantly, Article 27 in Chapter 4 specifies a list of substantive factors that will be considered in merger reviews: (1) the market shares of the undertakings that are parties to the concentration and their ability to control the relevant market; (2) the degree of market concentration in the relevant market; (3) the impact of the concentration on market entry and technological progress; (4) the impact of the concentration on consumers and other undertakings; (5) the impact of the concentration on national economic development; and (6) other factors affecting market competition as determined by the merger review authority. Furthermore, Article 28 states that the merger review authority will prohibit a concentration that has or possibly has the effects of eliminating or restricting competition, but will not prohibit a concentration if parties to the concentration could demonstrate that the pro-competitive effects of the concentration outweigh its anti-competitive effects or that the concentration is in the public interest. While Chapter 4 sets forth the general principles and procedures for merger review under the AML, it leaves out many details, one of which is the notification thresholds. In August 2008 the State Council promulgated a regulation that supplied the notification thresholds.53 The regulation requires a concentration to be pre-notified to the merger review authority if: (1) the combined worldwide turnover of all undertakings concerned in the preceding financial year is more than RMB 10 billion yuan, and the nationwide turnover within China of each of at least two of the undertakings concerned in the preceding financial year is more than 53 See ‘Guowuyuan guanyu jingyingzhe jizhong shenbao biaozhun de guiding’ (Provisions on Thresholds for Prior Notification of Concentration of Undertakings) (Ministry of Commerce of the PRC Website, 13 August 1998) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 459

RMB 400 million yuan; or (2) the combined nationwide turnover within China of all the undertakings concerned in the preceding financial year is more than RMB 2 billion yuan, and the nationwide turnover within China of each of at least two of the undertakings concerned in the preceding financial year is more than RMB 400 million yuan. What emerges from Chapter 4 of the AML and other related regulations is a merger review regime that is, in general, similar to those found in advanced antitrust jurisdictions. The merger review regime under the AML subjects concentrations that meet the notification thresholds to prior review by a merger review authority, which has the power to block or impose restrictions on a proposed concentration based on its evaluation of the possible effects of the concentration. But the factors considered in merger review under the AML appear to be broader and less predictable than factors considered in merger review in advanced antitrust jurisdictions. The AML allows the merger review authority to consider the effects of a concentration on ‘other undertakings,’ ‘national economic development,’ and ‘public interest,’ factors that are not typically considered in advanced antitrust jurisdictions. These unusual factors would allow the merger review authority to take into account macroeconomic, industrial and other policies in its merger review decisions and, as a result, increase the unpredictability of merger review outcomes under the AML. 2.5 Administrative Monopolies Chapter 5 of the AML is devoted to administrative monopolies, or anti-competitive conduct by government agencies. Article 32 in Chapter 5 prohibits government agencies or other organizations charged with governmental functions from requiring firms or individuals to deal in, purchase, or use products supplied by designated undertakings. These ‘mandated dealings’ have their origin in the close personnel or financial ties that exist between many government agencies and firms in China. For example, in August 2008, China’s General Administration of Quality Supervision, Inspection and Quarantine (AQSIQ) was sued for requiring all companies in China to use the anti-counterfeiting products of a company that it partially owned.54 These mandated dealings are declared illegal under Article 32. 54 See ‘Fayun bu shouli 4 jia fangwei qiye zhuanggao zhijian zongju fanlongduan an’ (Court Not Accepting Four Companies’ Antimonopoly Lawsuit Against AQSIQ), Jinghua Shibao (Jinghua Times) (Beijing 5 September 2008) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 17 /

Date: 8/9

JOBNAME: Duns PAGE: 18 SESS: 5 OUTPUT: Wed Oct 21 12:16:51 2015

460 Comparative competition law

Articles 33 to 35 focus on regional protectionism, i.e., discriminatory measures by local governments against out-of-locality products or producers. Regional protectionism has long been a problem throughout Chinese history but has escalated since the early 1980s as economic and political reforms increasingly tied the welfare of local governments to the well-being of local firms. Particularly, since the late 1980s sustained capacity expansion in China’s industries and sluggish domestic demand began transforming China’s economy from one characterized by shortage to one characterized by oversupply. In order to protect the market shares of local firms and hence local fiscal revenues, local governments in China since the late 1980s have taken various measures to block or restrict the entry of products or firms from other localities, a practice known as ‘regional blockades’ in China. Regional blockades in China nowadays take many different forms, including discriminatory tariffs, taxes, or fees for out-of-locality products, express preferences for local products in local government procurement, and outright prohibition of out-of-locality products.55 These various forms of regional blockades are declared illegal under Articles 33 to 35 of the AML. Specifically, Article 33 prohibits government agencies and other organizations charged with governmental functions from: (1) imposing discriminatory fees or charges on out-of-locality products; (2) applying discriminatory technology standards, inspection standards, and quality certification standards to out-of-locality products; (3) using discriminatory licensing requirements to restrict the entry of out-of-locality products; (4) using border barriers or other means to block regional trades; and (5) using other means to impede the free flow of products. Article 34 prohibits government agencies or other organizations charged with governmental functions from imposing discriminatory requirements or standards on out-oflocality undertakings aimed at preventing them from participating in bid tendering or restricting their abilities to participate in bid tendering. Article 35 prohibits discriminatory treatment of out-of-locality undertakings by government agencies or other organizations charged with governmental functions in investment approvals and in approvals for establishing local branches. Although the AML strictly prohibits administrative monopolies, the remedies it provides against administrative monopolies are less than adequate. Article 51 in Chapter 6 provides that administrative monopolies 55 See Wentong Zheng, ‘Administrative Monopolies in China: An Overview and Recent Developments’ in Barry Hawk (ed), International Antitrust Law & Policy: Fordham Competition Law Institute (Huntington, NY: Juris Publishing 2011).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 18 /

Date: 21/10

JOBNAME: Duns PAGE: 19 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 461

should be ‘corrected by superior government agencies,’ and the antimonopoly enforcement authorities are only to ‘advise’ the superior government agencies on the correction measures. This provision appears to be a tacit acknowledgment that the problem of administrative monopolies is a problem stemming from an imperfect political system.56 But because superior government agencies do not have proper incentives or capabilities to address administrative monopolies,57 the AML’s prohibitions against administrative monopolies appear to be little more than exhortative. 2.6 Legal Liabilities Chapter 7 of the AML sets forth the legal liabilities for violating the AML. Articles 46 to 48 concern administrative enforcement by the Antimonopoly Enforcement Authority. Article 46 provides that undertakings that violate the AML’s prohibitions of monopolistic agreements could face a fine equal to 1 to 10 percent of the undertakings’ revenues in the preceding year. Article 46 further provides for a leniency program, under which undertakings that volunteer important information and evidence about monopolistic agreements to the Antimonopoly Enforcement Authority will be eligible for reduced fines and penalties. Article 47 provides that undertakings that violate the AML’s abuse-of-dominance provisions could face a fine equal to 1 to 10 percent of their revenues in the preceding year. Article 48 provides that undertakings that violate the merger review provisions of the AML could be ordered by the Antimonopoly Enforcement Authority to terminate the concentration in question and to divest equities or assets, and could also face a fine up to RMB 500,000 yuan. In addition to administrative enforcement, Chapter 7 also provides for a private right of action against undertakings that violate the AML. Article 50 provides that ‘undertakings that engage in monopolistic conduct and cause damages to others shall have civil liabilities under the law.’ Recall that Article 12 defines undertakings as ‘natural persons, legal persons or other entities that are engaged in the production or distribution of products or services.’ As a result, it appears that government agencies

56 This is the argument made by people opposing the inclusion of administrative monopolies in the AML during the AML’s drafting process: Huang (n 22) 131. 57 See Wang (n 25) 149.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 19 /

Date: 13/10

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

462 Comparative competition law

that are not directly engaged in the production or distribution of products or services may not be sued by private plaintiffs for violating the AML.58

3. THE ENFORCEMENT OF THE ANTIMONOPOLY LAW As discussed above, when it comes to enforcing the AML, the AML provides for both administrative enforcement and civil actions. Since the AML came into effect in 2008, China’s antimonopoly enforcement agencies and courts have come a long way in laying out the framework of an AML enforcement regime. Efforts to enforce the AML, however, have been hindered by deep-rooted structural problems in China’s economy as well as lack of expertise and experience in competition law matters on the part of regulators and judges. The challenges for enforcing the AML manifest themselves in the AML’s enforcement structure itself. Despite calls for a new, unified enforcement agency with sufficient authority,59 the responsibilities for enforcing the AML are divided among three existing agencies – MOFCOM, NDRC, and SAIC. MOFCOM is responsible for merger review, while NDRC and SAIC are responsible for enforcement against monopolistic agreements, abuse of dominance, and administrative monopolies. NDRC and SAIC further divide their enforcement responsibilities based on whether the conduct in question involves pricing. NDRC, China’s price regulator, is responsible for enforcing against non-merger conduct involving pricing, while SAIC is responsible for enforcing against non-merger conduct not involving pricing. Likely an outcome of the lack of a strong consensus on the necessity of the AML, this highly fragmented enforcement structure casts shadows on the prospect of an effective AML enforcement regime. 58 It is perhaps for this reason that the lawsuit against AQSIQ (discussed above in the section on administrative monopolies) was filed pursuant to China’s Administrative Litigation Law, not the AML. The court did not accept the lawsuit, reasoning that the statute of limitations for raising claims against AQSIQ under the Administrative Litigation Law had already run by the time the lawsuit was filed: ‘Chinese Court Rejects 1st Anti-Monopoly Case’ Xinhua (CRI English) (Beijing 5 September 2008) . 59 See Wang (n 25) 144–45; Owen and others, ‘China’s Competition’ (n 8) 261–62.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 20 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 463

3.1 Ministry of Commerce (MOFCOM) Among all three AML enforcement agencies, MOFCOM is the most active one in terms of workload. According to the latest statistics released by MOFCOM, between 1 August 2008 and 30 September 2012, MOFCOM reviewed 474 merger applications, 458 of which were approved without any conditions.60 Of the remaining cases, 15 were approved with conditions and one was blocked by MOFCOM.61 As required by the AML, MOFCOM publishes all its decisions that impose conditions on or block a proposed merger. Over the years, MOFCOM’s published merger decisions have become more sophisticated. MOFCOM’s first published decision in November 2008 conditionally approving the proposed merger between InBev and Anheuser-Busch was only a half page long and contained very little information on the rationales for its decision.62 By contrast, its decisions in later years were substantially longer and contained more sophisticated economic analyses.63 From a substantive point of view, MOFCOM’s published decisions are generally consistent with mainstream competition law analyses, with some notable exceptions. MOFCOM’s rejection in 2009 of the bid by Coca-Cola to acquire Huiyuan Juice, China’s leading juice maker, caused the biggest stir in the international antitrust communities. Although Coca-Cola accounted for over 60 percent of the market for carbonated soft drink in China, Huiyuan’s market share in that area was insignificant. However, despite the fact that the proposed acquisition would have little competitive impact, MOFCOM blocked the transaction, citing the ‘leverage’ effect of the transaction. MOFCOM reasoned that if the transaction went through, Coca-Cola would have the ability to leverage its dominant 60

See ‘Jingyingzhe jizhong fanlongduan shencha wu tiaojian pizhun anjian xinxi tongji qingkuang’ (Statistics on Unconditional Approvals of Applications for Undertaking Concentration) (Ministry of Commerce of the PRC Website, 16 November 2012) . 61 For MOFCOM decisions on these cases, see (Ministry of Commerce of the PRC Website, 2014) . 62 See ‘Zhonghua renmin gongheguo shangwubu gonggao No. 95’ (Public Notice of the Ministry of Commerce of the People’s Republic of China) (Ministry of Commerce PRC Website, 18 November 2008) . 63 See, for example, ‘Shangwubu gonggao No 22’ (Public Notice of the Ministry of Commerce) (Ministry of Commerce of the PRC Website, 23 April 2013) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 21 /

Date: 8/9

JOBNAME: Duns PAGE: 22 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

464 Comparative competition law

position in the carbonated soft drinks market to strengthen its position in the fruit drinks market, thus affecting other fruit drink producers and harming competition.64 The leverage effect of a conglomerate merger, however, has been subject to heated debates in international antitrust communities and has long ceased to be used as a basis for blocking conglomerate mergers in the US.65 A widely held opinion is that nationalism played a major role in MOFCOM’s decision in the case as Huiyuan is China’s best-known fruit juice brand.66 Similarly, noncompetition factors such as industrial policy may have accounted for the conditions MOFCOM imposed in a number of cases.67 Another area in which MOFCOM has been lacking is transparency. As discussed above, the AML requires MOFCOM to make a decision on a merger application within the timeline specified in the AML. But the clock for MOFCOM’s merger decisions starts clicking only after MOFCOM receives documents that comply with the requirements set forth in the AML. This gives MOFCOM the flexibility of not accepting a merger application and therefore not starting the clock for its decision by claiming that the documents submitted are incomplete. This way, MOFCOM can effectively reject an application without going on the record stating the reasons for the rejection. An example of this practice is the attempted acquisition by the Chinese internet portal Sina.com of Focus Media, a Chinese advertising and media company. The acquisition was notified to MOFCOM in December 2008 but was never accepted by MOFCOM,68 and the parties eventually abandoned the deal in September 2009 because of the uncertainty surrounding the approval.69 Another issue that has drawn much scrutiny is the treatment by MOFCOM of mergers involving SOEs. Unlike the previous merger 64 See ‘Zhonghua renmin gongheguo shangwubu gonggao No 22’ (Public Notice of the Ministry of Commerce of the People’s Republic of China) (Ministry of Commerce of the PRC Website, 8 March 2009) . 65 See Huyue Zhang, ‘Problems in Following EU Competition Law: A Case Study of Coca-Cola/Huiyuan’ (2011) 3 Peking University Journal of Legal Studies 96, 104–12. 66 See, for example, Yee Wah Chin, ‘The High-Wire Balancing Act of Merger Control Under China’s Anti-Monopoly Law’ (SSRN, 26 August 2012) 13 . 67 See ibid 14–29. 68 Ibid 9. 69 See David Barboza, ‘Sina.com Pulls Out of Deal After Delay’, New York Times (New York, 28 September 2009) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 22 /

Date: 8/9

JOBNAME: Duns PAGE: 23 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 465

review regime, which only covered the acquisition of domestic firms by foreign firms, the merger review regime established under the AML applies to both foreign and domestic mergers. But it is not clear to what extent MOFCOM can conduct meaningful reviews for domestic mergers, particularly those involving SOEs. So far, the vast majority of mergers being notified to MOFCOM are between foreign parties. Some of the mergers involving China’s largest SOEs since the AML took effect were not notified to MOFCOM at all.70 For SOE mergers that are notified to MOFCOM, it remains to be seen whether MOFCOM can exercise independent review, given that the central government has a longstanding policy of promoting SOE mergers to consolidate the SOE sector and to cultivate national champions.71 3.2 National Development and Reform Commission (NDRC) In keeping with its role as China’s price regulator, NDRC is assigned the responsibility of enforcing against price-related violations of the AML in the areas of monopolistic agreements, abuse of dominance, and administrative monopolies. As discussed in detail below, NDRC has exercised its AML enforcement power only sparsely and selectively. One of NDRC’s primary responsibilities is enforcement against price cartels. However, since 2008 NDRC and its local offices have publicly announced only a handful of enforcement actions against price cartels. In March 2010, NDRC announced that it had punished several Chinese 70 The mergers among China’s ‘central SOEs’ i.e., SOEs directly controlled by the central government, were not among the list of mergers that received unconditional approvals from MOFCOM: See MOFCOM, ‘List of Unconditional Approvals (2008.8.1–2012.9.30)’ (Ministry of Commerce of the PRC Website) . 71 On 18 December 2006 the State Assets Supervision and Management Commission (hereafter SASAC) announced that it was its goal to maintain ‘absolute control’ by SOEs of seven ‘strategic’ industries, including national defense, electrical power generation and grids, petroleum and petrochemicals, telecommunications, coal, civil aviation, and waterway transportation: See ‘Guanyu tuijin guoyou ziben tiaozheng he guoyou qiye chongzu de zhidao yijian’ (Guidance on the Restructuring of State Capital and State Owned Enterprises) (Sina, 18 December 2006) . SASAC also stated that its goal was to foster 30 to 50 large ‘internationally competitive’ SOEs in those industries by 2010: See ‘SOEs to Maintain Absolute Control in Seven Sectors’ Shanghai Zhengquan Bao (Shanghai Securities Journal) (Shanghai, 19 December 2006) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 23 /

Date: 13/10

JOBNAME: Duns PAGE: 24 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

466 Comparative competition law

manufacturers of rice-noodles in Guangxi Province for price fixing.72 In July 2010 NDRC announced that it had fined a number of agricultural companies for conspiring to raise the price of green mung beans across China.73 NDRC took action against both the rice noodle cartel and the green mung bean cartel pursuant to the Price Law, not the AML. In January 2011 NDRC announced that it had imposed the maximum amount of fine allowed under the AML on a trade association in Fujian province for fixing the prices of certain paper products.74 This time, NDRC took action pursuant to both the Price Law and the AML. In October 2012 the provincial price bureau of Guangdong province for the first time applied the leniency provision under the AML, awarding a 50 percent reduction in fines to a member of a price cartel in the sea sand extraction industry.75 In January 2013, in the largest cartel case ever brought by NDRC, NDRC imposed hefty fines on Korea’s LG and Samsung and six Taiwanese producers of flat panel displays for price fixing.76 Other than these cases, NDRC has essentially turned a blind eye on cartels, particularly those involving large SOEs. For example, NDRC took no action against a well-publicized ‘joint price-pushing campaign’ in 2009 by China’s two largest petroleum firms, China National Petroleum Corporation (CNPC) and Sinopec.77 NDRC also took no action against Sino-Chem and Sino-Agri, China’s two largest manufacturers of 72

See Nathan Bush, ‘China’s NDRC Punishes Rice Noodle Cartel Members’ (O’Melveny & Myers LLP, 2 April 2010) . 73 See ‘China Fines Agricultural Companies for Agreeing to Raise the Price of Their Products’ (Eversheds, 16 July 2010) . 74 ‘Fagaiwei chufa fuyang zaozhi xiehui zhengzhi jiage longduan’ (NDRC Fines Fuyang Paper Trade Association; Tackles Price Monopolies) Meiri Jingji Xinwei (Daily Economic News) (11 May 2011) . 75 See David Livdahl and others, ‘NDRC’s Recent Enforcement of the PRC Anti-Monopoly Law: A More Aggressive and Transparent Direction’ (Paul Hastings China Matters, 2013) 2 . 76 See Chunmei He, ‘NDRC Punishes Six Makers of Flat Screens for Price Fixing’ (Caixin Online, 4 January 2013) . 77 See Kangpeng Wang, ‘Zhong shiyou zhong shihua zai tuijia’ (CNPC and Sinopec Start Price-Pushing Campaign Again) (Finance QQ, 21 March 2009) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 24 /

Date: 13/10

JOBNAME: Duns PAGE: 25 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 467

potassium fertilizers, which reportedly conspired to raise the price of potassium fertilizers in 2009.78 The lack of enforcement by NDRC against cartels is not entirely surprising, given that there are chronic excess capacities in many of China’s industries.79 Underlying China’s excess capacity problem are widespread structural distortions caused by the roles of the government in both the investment process and the firm exit process. Excess capacities exert enormous downward pressures on prices and lead to the much lamented problem of ‘malignant competition’ or ‘excessive competition.’80 As a result, NDRC might be tempted to tolerate cartels as a means of reinstating some sort of price control that had been relinquished in the reform era.81 Unless and until the structural distortions are removed from China’s economy, it will be nearly impossible for NDRC to enforce the anti-cartel provisions of the AML in a robust manner. NDRC has been even less active in enforcing the abuse-of-dominance provisions of the AML. In November 2011, NDRC announced that it had fined two pharmaceutical distributors in Shandong province for monopolizing the supply of a chemical compound that is an essential ingredient of a key high-blood-pressure drug.82 The two distributors signed an exclusive distribution agreement with the only two producers of the chemical compound and then sharply increased the price of the chemical compound. NDRC was not clear, however, about the legal basis of its action. One possible theory could be that the two distributors might have abused their dominant market position by charging excessively high prices for the chemical compound, thus implicating the abuse-ofdominance provisions of the AML. 78 See Qiyu Chen, ‘Jiafei diejia; zhonghua zhongnong yu lianshou taijia’ (Price of Potassium Fertilizers Falls; Sino-Chem and Sino-Agri Plan to Jointly Raise Price) Shanghai Zhengquan Bao (Shanghai Securities Journal) (Shanghai, 17 April 2009) . 79 See Zheng, ‘Transplanting Antitrust’ (n 1) 675–82, and n 40 and accompanying text. 80 For more detailed discussions of ‘excessive competition’ in China, see Zheng, ‘Transplanting Antitrust’ (n 1) 682–85. 81 Ibid 686–92. 82 See Henry (Litong) Chen, Alex An and Frank Schoneveld, ‘China NDRC Fines Two Pharmaceutical Distributors for Monopolistic Practice’ (McDermott Will & Emery, 21 November 2011) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

468 Comparative competition law

NDRC also became involved in an abuse-of-dominance dispute concerning China’s two major telecommunications carriers, China Telecom and China Unicom. It was reported in November 2011 that NDRC had launched an investigation into alleged monopolistic conduct by China Telecom and China Unicom in the market for broadband internet services.83 China Telecom and China Unicom, combined, controlled roughly two-thirds of the market for broadband internet services and allegedly charged rival cable-based internet service providers higher fees for accessing their broadband networks. The NDRC investigations were hailed as a breakthrough as they were the first investigations into major SOEs under the AML.84 Yet in retrospect, the competition law significance of the investigations appears to have been overstated. Shortly after the NDRC investigations were launched, China Telecom and China Unicom each issued a statement outlining steps to comply with the AML and each asked NDRC to terminate its investigations.85 The investigations then faded out of public sight, with no legal doctrines being clarified and no precedents being established. More likely than not, the NDRC investigations into China Telecom and China Unicom were an isolated event brought about by ongoing battles between China’s telecom carriers backed by the Ministry of Information and Industry and cable carriers backed by the State Administration of Radio, Film and Television over the central government’s initiative to merger China’s telecom, cable and internet networks into one ‘super’ network. 3.3 State Administration of Industry and Commerce (SAIC) SAIC is charged with enforcing against non-price related violations of the AML in the areas of monopolistic agreements, abuse of dominance, and administrative monopolies. Among all three AML enforcement agencies, SAIC is arguably the least active one.

83 See Loretta Chao, ‘China Telecom, China Unicom Face Monopoly Probe’ Wall Street Journal (New York, 9 November 2011) . 84 See, for example, Wang Yong, ‘Breakthrough for the SOE Monopoly Battle’ (Caixin Online, 1 December 2011) . 85 ‘China Telecom, China Unicom Pledge to Mend Errors After AntiMonopoly Probe’ (Xinhua Net, 2 December 2011) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 27 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 469

So far, SAIC has publicly announced only one enforcement action against cartels and one enforcement action against administrative monopolies. In a decision published in January 2011, the SAIC office in Jiangsu province imposed fines on an industrial association for the concrete industry in Lianyungang city of Jiangsu province and members of the industrial association for entering into a market allocation agreement under which association members were assigned specific market shares for the supply of concrete.86 In July 2011, it was reported that the SAIC office in Guangdong province helped overturn a regulation issued by an unnamed city in Guangdong province. The regulation in question required all GPS tracking service providers in the city to transmit their data to a provincial transportation authority through a designated GPS tracking service provider for a fee. Pursuant to Chapter 5 of the AML, the Guangdong SAIC recommended to the Guangdong provincial government that the regulation be revoked. The Guangdong provincial government adopted the recommendation and revoked the regulation.87 Although SAIC has been relatively quiet on the enforcement front, it has recently taken the lead in drafting an enforcement guidance document on issues related to intellectual property under the AML. In August 2012 SAIC released the fifth draft of its Guidelines on Antimonopoly Law Enforcement in the Field of Intellectual Property Rights for public comments.88 The draft guidelines address a wide range of issues such as refusals to license, restrictions on product sales, restrictions on purchases of new technology, and patent pools.89 Once finalized, the guidelines are expected to play an important role in shaping the enforcement of the AML in the area of intellectual property rights. 86

Susan Ning and others, ‘First Public Enforcement Decision by SAIC Against Concrete Manufacturers’ (China Insight, 28 February 2011) . 87 Susan Ning and others, ‘First Enforcement Action under Anti-Monopoly Law Against Administrative Monopoly’ (China Law Insight, 12 August 2011) . 88 An unofficial translation of the guidelines can be found at . 89 For a detailed analysis of the draft guidelines, see Mark Allen Cohen, ‘A Quick Read of the AML IPR Enforcement Guidelines (5th Draft)’ (China IPR, 26 August 2012) .

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 27 /

Date: 8/9

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

470 Comparative competition law

3.4 Courts Besides enforcement by MOFCOM, NDRC, and SAIC, civil actions in courts are an important avenue for seeking remedies under the AML. As discussed above, courts in China have already clarified certain important issues arising under the AML, such as whether competitive consequences can be taken into account in deciding the legality of monopolistic agreements and RPM.90 However, Chinese courts have had their share of confusion in handling lawsuits under the AML because of the complexity of the AML lawsuits and Chinese judges’ relative inexperience in competition law matters. Such confusion was on vivid display in China’s first private litigation under the AML. In October 2009 a court in Shanghai decided a case involving abuse of dominance allegations.91 The plaintiff, an internet portal in China specializing in publishing online fictions written by anonymous web users, sued the defendant, another internet portal, for forcing two novelists to stop publishing their novels on the plaintiff’s website. The two novelists had written a sequel to a popular online novel first published on the defendant’s website and started publishing it on the plaintiff’s website. The defendant asked the two novelists to discontinue writing the sequel, and asked other web portals in China not to publish the sequel. The two novelists complied with the defendant’s request. The plaintiff argued that in so doing, the defendant violated Article 17(4) of the AML, which prohibits an undertaking from ‘restricting transaction counterparts to only deal with an undertaking or undertakings designated by the undertaking without justifications.’ The court accepted the lawsuit, and spent a great deal of time hearing arguments about how to define the relevant market and whether the defendant had a dominant market position in the relevant market. The court failed to realize that the defendant’s conduct did not fall under the purview of Article 17(4) of the AML at all. The court eventually ruled in favor of the defendant on the ground that its conduct was ‘justified,’ but the court should not even go that far, as there was not a proper cause of action under the AML at all. An issue that has been a focus in private AML litigation so far that of burden of proof. In several high-profile private AML lawsuits, all of 90

See AML s 2 B. ‘Zhongguo shouli fanlongduan wangluo anjian yishen xuanpan’ (Court Decides China’s First Internet Antimonopoly Lawsuit) (Caijing Wang (Business & Finance Net) 23 October 2009) . 91

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 28 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Competition law in China 471

which concerned abuse-of-dominance allegations, the courts ruled that the plaintiffs failed to demonstrate that the defendants possessed a dominant position in the relevant market.92 This practice has made it very difficult for plaintiffs to win private AML lawsuits. Yet this practice found support in a key judicial interpretation promulgated by the Supreme People’s Court in May 2012 on private AML litigation.93 The judicial interpretation explicitly requires plaintiffs to shoulder the burden of proof as to market definition and abuse-of-dominance issues.94 Within the foreseeable future, therefore, plaintiffs seeking redress in courts against dominant firms will still face an uphill battle.

4. CONCLUSION Since the AML came into effect in 2008, efforts to establish a comprehensive competition law regime in China have definitely come a long way: the basic framework of a competition law regime covering all major areas of competition law has been established, the infrastructure of an enforcement regime has been put in place, and the competition agencies and courts have been making steady progress in enforcing the law. Yet one has to be mindful that these efforts are taking place in a setting where state control and structural distortions are still the norm. These efforts are also being hampered by the lack of expertise and experience in competition law matters. It might take quite a while, therefore, before China can develop a truly robust competition law regime.

92 93 94

Columns Design XML Ltd

See Emch and Liang (n 31) 8. See Judicial Interpretation (n 30). See ibid Art 8.

/

Job: Duns-Comparative_competition_law

/

Division: 20_Chapter16

/Pg. Position: 29 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

17. Latin American antitrust law and policy: An overview of three jurisdictions – Brazil, Chile and Colombia Javier Tapia and Alexandre Ditzel Faraco *

1. INTRODUCTION It has become almost a commonplace to say that, nowadays, more than 100 jurisdictions have adopted competition regimes. Within this context, Latin America is not the exception. So far, 22 countries have enacted national competition laws.1 Also, supranational competition statutes have been enacted within the framework of the Andean Community of Nations (CAN), the Caribbean Community (CARICOM) and the Mercado Común del Sur (Mercosur). Furthermore, several of these jurisdictions have recently amended their competition laws or are currently discussing the introduction of changes, with the aim of making the regimes stronger.2 At first glance, competition law and policy in Latin America seems booming. * The views expressed in this chapter are the authors’ solely and not necessarily the views of the Tribunal de Defensa de la Libre Competencia (TDLC) or any of its individual members. 1 The 22 countries are: Argentina, Barbados, Bolivia, Brazil, Chile, Colombia, Costa Rica, Ecuador, El Salvador, Guyana, Honduras, Jamaica, Mexico, Nicaragua, Panama, Paraguay, Peru, Dominican Republic, Saint Vincent & the Grenadines, Uruguay, Trinidad & Tobago and Venezuela. Guatemala is currently in the early stages of enacting a competition Act. 2 Among the countries that have recently amended their laws are Brazil (2000, 2007 and 2011), Argentina (2001), Andean Community of Nations (2005), Mexico (2006 and 2013), Panama (2006 and 2007), El Salvador (2007), Peru (2008), Chile (2002, 2003 and 2009) and Colombia (2009). At the time of writing, competition Bills are being discussed in Argentina, Costa Rica, Ecuador, Guatemala and Paraguay. Reforms cover a wide range of topics, such as institutional reforms (Brazil, Chile and Mexico); the introduction or enhancement of leniency programmes (Brazil, Mexico, Panama, El Salvador, Peru, Chile and Colombia); fines enlargements (Andean Community, Panama, Mexico, El Salvador, Peru, Chile and Colombia); the provision of more powers for competition authorities, including dawn raids and wire tapping (Brazil, El Salvador, Mexico

472

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 2 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 473

The microscopic reality is fairly different. There is a large disparity in how active and successful Latin American competition law regimes have been in shaping good policies and/or applying state-of-the-art legal and economic standards. In fact, most of them are ineffective. Although reasons vary, arguably a central one is the widespread reliance on the United States (US) and/or the European Union (EU) systems as main sources of reference for the creation or enhancement of competition regimes,3 without careful consideration of native realities. This goes against the general consensus on the idea that antitrust should be locally rooted, depending on the underlying economic conditions.4 The scarcity of comparative works focusing specifically on Latin American regimes may be negatively contributing to such trend. This chapter aims at contributing to fill this gap, providing a brief overview of competition law and policy in three Latin American jurisdictions: Brazil, Chile and Colombia.5 Since they have consistently followed the path of competition and Chile); the increase in the number of markets subject to competition law (Panama, Peru, Mexico and Colombia); the introduction of settlements and commitments (Andean Community, Panama and Mexico); the modification of merger thresholds (Argentina, Mexico and Colombia); the introduction of new exemptions to competition laws (Andean Community, Panama and Colombia); and the issuance of permission for third party (i.e., competitors’ and/or consumers’) intervention in proceedings (Panama and Colombia), among others. 3 This is no different than other new competition jurisdictions: see E Fox, ‘Economic Development, Poverty and Antitrust: The Other Path’ (2007) 13 Southwestern Journal of Law and Trade in the Americas 112, 113; R Whish and C Townley (eds), New Competition Jurisdictions: Shaping Policies and Building Institutions (Edward Elgar 2012). 4 Note that there are opinions expressing the opposite idea. See G Priest, ‘Competition Law in Developing Nations: The Absolutist View’ in D Sokol, T Cheng and I Lianos (eds), Competition Law and Development (Stanford University Press 2013): arguing that competition law of all nations should in principle be identical and only details of policy implementation may differ, because there is a single best-defined competition law to improve social welfare. 5 For general reviews of some countries not covered here (in Spanish): M Krakowski, Política de competencia en Latinoamérica: una primera apreciación: un análisis comparativo legal e institucional de las políticas de competencia en Latinoamérica (Managua 2001, Proyecto MIFIC/GTZ) (comparing several regimes); E Quintana, ‘Abuso de posición de dominio conjunta y colusión tácita: ¿infracciones sin contenido real?’ (2005) Thémis Revista de Derecho No 51: studying the Peruvian case law on tacit collusion and abuse of collective dominance; G Guevara Inciarte, ‘El tratamiento de las prácticas y comportamientos colusorios en Venezuela (1992–2006): problemas teóricos y prácticos’ (2008) 8 Revista de Derecho de la Competencia: studying the Venezuelan case law on collusion.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 2 /

Date: 13/10

JOBNAME: Duns PAGE: 3 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

474 Comparative competition law

at least during the last 10 years or so, arguably with success, they stand out as examples of how to apply competition law considering specific local settings. There are a number of social, political and economic variables that make the situation in Latin America fairly unique.6 First, most economies are of medium or small size.7 Second, despite important progress, social inequality still haunts most (if not all) countries.8 Third, the degree of concentration in most industries remains high. Oligopolies are (or have been until fairly recently) largely dominant in many markets, as a result of a long tradition of State-controlled economies combined with social conflicts and resentments9 and/or various concerns about ‘too much competition’.10 Fourth, there are powerful interests that have strong political power. It is therefore extremely difficult to go against these ruling elites. Indeed, in some sectors the threat of entry may come from imports – especially in some countries with very open economies.11 However, that threat is fairly limited in both tradable and non-tradable sectors. Particularly in the latter, there is no credible threat of entry, since

6

Some authors have also highlighted this point. See, for example, D Gerber, Global Competition: Law, Markets, and Globalization (Oxford University Press 2010) 237. 7 Gal defines a small economy as ‘an independent sovereign economy that can support only a small number of competitors in most of its industries’, but recognises that the definition is ‘arbitrary’: MS Gal, Competition Policy for Small Market Economies (Harvard University Press 2003) 1–2. According to the World Bank, in 2012 the GDP of Latin America and the Caribbean reached US$5.343 trillion, out of a world total GDP of US$71.92 trillion. 8 But see GA Cornia, Falling Inequality in Latin America: Policy Changes and Lessons (Oxford University Press 2014); GA Cornia, ‘Income Distribution Under Latin America’s New Left Regime’ (2010) 11 Journal of Human Development and Capabilities 85: showing improvements in the distribution of income during the years 2002–07 as a result of a decline in skill premium and fiscal prudence. 9 Gerber (n 6). 10 At least in principle, oligopolistic markets are more prone to cartelisation. This insight began with Stigler in 1964, but there is empirical literature which also offers support to this point: G Stigler, ‘A Theory of Oligopoly’ (1964) 72 Journal of Political Economy 44. 11 See for example Organisation for Economic Co-operation and Development (hereafter ‘OECD’), ‘Competition Law and Policy in Panama – A Peer Review’ (OECD 2011): highlighting openness with respect to the Panamanian economy.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 3 /

Date: 13/10

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 475

financing is not readily available for new entrepreneurs outside these concentrated groups.12 In addition, with the exception of a few countries where competition policy has become an important part of the rules of the game, support for competition as a ‘value’ is generally weak. In large parts of society, competition is not seen as the best path to enhance productivity and achieve prosperity and economic progress. Conversely, in many countries ideological reasons and historical anti-market traditions still play a role, making industrial policy and control over the wider economy a favourite among many governments.13 This contrasts with the situation at the beginning of the 1990s, when many governments developed an agenda including trade liberalisation, deregulation and privatisation. The enactment of competition laws in many jurisdictions was part of those reforms. Recently, however, left-wing parties elected democratically in some countries have advocated an alternative for the so-called ‘neoliberal model’, with governments preferring the use of command-and-control regulation over sound competition policy. Overall, the real influence of competition remains scarce.14 Certainly, it is still possible to attribute the ineffectiveness of most regimes to other reasons. For example, there may be lack of maturity. In fact, only nine countries have more than ten years of practice enforcing the law and count on experienced competition authorities.15 Others have enacted their laws recently or have had no enforcement at all. The lack of economic and human resources or a faulty institutional design may also

12 S Khemani and A Carrasco-Martin, ‘The Investment Climate, Competition Policy, and Economic Development in Latin America’ (2008) 83 Chicago-Kent Law Review 766. 13 Historically, import-substitution policies were common in Latin America between the 1950s and the 1980s. See, among others, W Peres, ‘Industrial Policies in Latin America’ (2011) United Nations University-World Institute for Development Economics Research, Working Paper No. 2011/48. 14 As shown by international indicators. For example, in the Global Competitiveness Report 2013–14, only three Latin American countries stand out regarding the ‘effectiveness of anti-monopoly policy’, an indicator that enquires to what extent anti-monopoly policy promotes competition. Among 148 countries, only Panama (25), Chile (38) and Brazil (40) ranked above 50th place: World Economic Forum, ‘The Global Competitiveness Report 2013–2014’ (World Economic Forum 2013) 472. 15 We include in this group Argentina, Brazil, Chile, Colombia, Costa Rica, Jamaica, Mexico, Panama and Peru. Venezuela should be included in this group but is not because of the country’s political situation.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 4 /

Date: 13/10

JOBNAME: Duns PAGE: 5 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

476 Comparative competition law

play a role.16 None of these explanations, however, decreases the importance of considering local realities and taking examples from closer and more familiar environments. The rest of the chapter is organised as follows. Section 2 provides a brief historical account of the developments in the three countries. Unsurprisingly, several commonalities can be found. The next two sections focus on institutional regimes and substantive rules respectively. Due to space constraints, the focus is on antitrust – i.e., collusive agreements and unilateral conduct. Merger control regimes are not assessed. The analysis includes law and policy, providing also a general overview of the main cases. From an institutional perspective, the three countries present very different designs. Conversely, policy objectives, legal structures and substantive standards seem similar to a certain degree. Finally, Section 5 concludes.

2. A BRIEF HISTORICAL ACCOUNT The enactment of statutory provisions protecting competition in Brazil, Chile and Colombia started in the middle of the 20th century. Chile’s first competition legislation (title V of Law No. 13,305) was introduced in 1959, as part of several reforms recommended by a group of US consultants to tackle the then prevailing hyper-inflation.17 That same year the Colombian Competition Act (Law No. 155/59), which is still the current governing statute, was also enacted. Brazil would have to wait three more years to have its first competition statute in place: Law No. 4,137, which dates back to 1962.18 These early developments foresaw promising times for competition. However, at that time systems did not 16

D Sokol, ‘The Development of Human Capital in Latin American Competition Policy’ in E Fox and D Sokol (eds), Competition Law and Policy in Latin America (Hart Publishing 2009). 17 The consultants are known as the ‘misión Klein-Sacks’. The legislation consisted merely of few provisions within a broad statute containing several macroeconomic regulations. A history of competition law in Chile (financed by the competition agency) can be seen in Patricio Bernedo, Historia de la Libre Competencia en Chile 1959–2010 (Fiscalía Nacional Económica 2013). 18 Article 148 of Brazil’s 1946 Constitution was the basis for the antitrust system established by Law No. 4,137/1962. This law created the competition authority (the CADE) and established substantive and procedural rules. Before Law No. 4,137/1962 other laws were enacted to deal with economic crimes and offences, including cartels. However, Law No. 4,137/1962 was the first to create an administrative system to tackle this conduct.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 5 /

Date: 13/10

JOBNAME: Duns PAGE: 6 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 477

fall in fertile soil and remained a marginal part of the most general economic policy. Legislation was developed mostly within an environment of strong State intervention in the economy, so its enforcement was slow. In Brazil, Law No. 4,137/1962 was never fully implemented. In 1964 an autocratic military government began, which adopted a policy that deepened earlier industrialisation efforts and commanded the economy under a heavy State leadership.19 In Chile, the work of the competition authorities before 1973 was ‘exiguous and of little relevance’.20 Likewise, the Colombian Law No. 155/1959 was never systematically enforced for the purpose of preserving competition, but used instead as the legal basis for applying price controls.21 Overall, there was seldom political, economic or intellectual support for the newly created frameworks. Although there were some cases, regimes only started consolidating in earnest as the three countries shifted to a market-based economy. Chile was the first. It implemented policies such as trade liberalisation, privatisation and deregulation at the beginning of the 1970s,22 after the military dictatorial government began to heavily reduce the role of the State in the whole national economy. This policy would eventually lead to the enactment of the current Competition Act, Decree Law No. 211/1973 (DL 211/73), Law for the Defence of Free Competition.23 The regime, however, was rather weak, with institutions that did not respond to the demands envisioned by the new statute.24 In reality, Chile would

19 CADE, ‘Relatório Anual de Gestão 1997’ (CADE 1998) 12–13. Between the 1950s and 1980s, an economic model of import substitution and industrialisation promotion prevailed. It was characterised by strong State intervention and price controls. 20 See P Bermedo, ‘Historia de la Libre Competencia en Chile 1959–2010’ (2013) 215 Fiscalía Nacional Económica, 58. 21 Law No. 155/1959 was based on Article 32 of the 1886 Constitution, which assigned to the State the general conduct of the economy and empowered it to intervene in specific circumstances. The Law was amended in 1963 (Decree No. 3,307) and regulations were issued in 1964 (Decree No. 1,302). Both failed in boosting its effective implementation. 22 E Cruz and S Zárate, ‘Building Trust in Antitrust’ in Fox and Sokol (eds) (n 16) 157. 23 DL 211/73 has been amended on several occasions, being the main reforms introduced in 2003 (by Law No. 19,911) and 2009 (by Law No. 20,361). The Competition Agency was established during a reform in 1963. 24 For example, there was a criminal provision against cartels that was never applied, until it was repealed in 2003.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 6 /

Date: 8/9

JOBNAME: Duns PAGE: 7 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

478 Comparative competition law

only have a modern competition regime after its first serious restructuring 30 years later. In Brazil, it was only in the mid-1990s when the competition regime was modernised, after a period of transition to democracy and in which the role of the State in the national economy was reconsidered.25 This newly created environment, favourable to antitrust enforcement, would lead to the replacement of Law No. 4,137/1962 with a new Competition Act: Law No. 8,884/1994, which remained in force and was applied with success for almost two decades, until its replacement by the current Law.26 Currently, antitrust law in Brazil is governed by Law No. 12,529/2011, enacted on 30 November 2011 and entered into force on 29 May 2012.27 Finally, in Colombia antitrust enforcement was preceded by the enactment of a new Constitution (in 1991), which established ‘Freedom of Economic Competition’ as a constitutional right. This paved the way for the issuance of Special Decree No. 2,153/1992, which elaborated the types of conduct subject to competition laws and organized them in several categories. Along with Law No. 155/59, it remains the governing law. In parallel, several pro-competitive institutional reforms meant the creation of sector-specific regulatory agencies with special mandates to 25 Inflation was pervasive at the beginning of the 1980s, exceeding 1,000 per cent in the early 1990s. Hyperinflation and a financial crisis in the mid-1980s vindicated the need for economic reforms in the late 1980s, which implied a new economic model where trade was liberalised and deregulation and privatisation were implemented. The 1994 competition law was also enacted with the objective of controlling inflated prices. Brazil’s economic reforms were successful. For example, the country became one of the most important producers of several agricultural products in the world (alcohol, sugar, coffee, orange juice, soybeans, beef, tobacco, poultry, fruits and maize): see generally, M de Rezende Lopes and others, ‘Brazil’ in World Bank (ed), Distortions to Agricultural Incentives in Latin America (World Bank 2008); OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2010); CADE, ‘Brazil: Competition Law and Policy in 1997–1998’ (CADE 1999). 26 Law No. 8,884/94 was amended on several occasions. Among others: Law 9021 of 1995, Law 9069 of 1995, Law 9470 of 1997, Law 9781 of 1999, Law 10149 of 2000 and Law 11482 of 2007. Compare OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2010) 11. 27 The regime was updated mainly with the aim of making administrative proceedings more expeditious. The most significant changes were: the change in the administrative structure of the competition authority; the introduction of a new merger control regime; and some amendments to the investigation of anti-competitive conduct. See L Pagotto, ‘New Landscape in the Brazilian Antitrust Enforcement’ (2012) 8 Revista de Derecho de la Competencia 159–166.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 7 /

Date: 13/10

JOBNAME: Duns PAGE: 8 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 479

prevent competition within their respective sectors, creating a decentralised institutional model for enforcing competition law.28 In 2009 the competition regime was amended significantly, both institutionally (mainly by establishing a single authority to enforce competition rules in most sectors) and from a substantive point of view.29

3. THE CURRENT INSTITUTIONAL SETTING The current institutional setting in Brazil, Colombia and Chile is summarised in Table 17.1. This section analyses each of these aspects in more detail. Table 17.1 Institutional setting Brazil

Chile

Colombia

Current Law No. 12,529/ Competition 2011 Act

Law Decree 211/73

1. Law No. 155/59 2. Decree No. 2,153/1992

Competition CADE authority

FNE

TDLC

SIC

Functions of the agency

Investigative, prosecutorial and adjudicative

Investigative, prosecutorial

Adjudicative

Investigative, prosecutorial and adjudicative

Sanctions

Fines and/or behavioural or structural remedies

Judicial review

1. Federal Courts (JR) 2. Courts of Appeal (appeal) 3. Superior Court of Justice (appeal)

Supreme Court of Justice (reclamación [appeal])

1. Administrative Court (annulment) 2. State Council (annulment)

28 There was also a ‘Competition Promotion Office’ within the Superintendence of Industry and Commerce (SIC). See OECD, ‘Competition Law and Policy in Colombia: A Peer Review’ (OECD 2009) 12–13; Alfonso Miranda, ‘El Régimen General de la Libre Competencia’ in CEDEC III (Javegraf 2002) 24–26. 29 By Law No. 1,340/2009.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 8 /

Date: 8/9

JOBNAME: Duns PAGE: 9 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

480 Comparative competition law

3.1 Competition Law ‘in the Books’: Goals and Legal Mandates The three countries analysed protect ‘free competition’. This is a common statutory wording in Latin American jurisdictions. Both Brazil and Colombia give it explicit constitutional recognition. The Brazilian Federal Constitution of 1988 regulates the Economic and Financial Order in its Title VII, which explicitly mentions ‘free competition’ as a principle in its article 170, postulating the ‘repression of abuses of economic power intended for market domination, elimination of free competition and arbitrary increase of profits’ (article 173, section 4). Likewise, the Colombian Constitution, enacted in 1991, recognises ‘free economic competition’ as a constitutional right. Also, its article 333 mandates the State to ‘impede the obstruction and restriction of economic liberty and prevent or control any abuse by persons or firms of their dominant position in the national market’. Unlike these two countries, Chile lacks any explicit recognition of competition within its Constitution. The general framework is only provided by the constitutional guarantee of economic liberty and a strong constitutional protection of private property. However, at statutory level, the law sets forth the same purpose to ‘advocate and defend free competition in the markets’.30 Notwithstanding the wording, it is now generally accepted that competition is about promoting economic efficiency, so that ‘free competition’ must be interpreted in such a way. However, there remain possibilities of interpreting the goals differently. In Colombia, for example, although efficiency is put forward at the statutory level, it is done so in a rather contradictory form. Article 2 of Decree No. 2,153/92 directs the competition authority to process complaints or claims over potential violations that are ‘significant for achieving the following goals in particular: free participation of business in the market, consumer welfare, and economic efficiency’. Note that ‘free participation in the market’ may easily be used as protection of competitors instead of competition. The balance is left to the judiciary, which has not given clear guidance on the issue.31 30 Article 1, DL No. 211/1973. This is supplemented in article 2 by the mandate to competition authorities ‘to enforce the present law to safeguard free competition in the markets’. 31 For instance, the Colombian Constitutional Court has held that free competition is ‘a guiding principle for economic activity to the benefit of consumers and entrepreneurial freedom’: Judgment T-240 of 1993 (emphasis added).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 9 /

Date: 13/10

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 481

In the case of Brazil and Chile, there is no explicit statutory declaration of what free competition means. Hence there is the same reliance on judicial interpretation. However, both countries have introduced important institutional changes designed to ‘invest’ in more economic analysis. The creation of both the TDLC – a specialised tribunal comprising economists as judges – and the current CADE – with an internal Economic Department that puts economics at the forefront – is proof. As a result, economic efficiency has been reflected in many recent decisions and there is a growing use of economic analysis and evidence.32 Regarding the legal mandates, the three countries are characterised for proscribing most anti-competitive conduct in an enabling and general way. In Brazil, the basic legal framework is set up in one single provision – article 36 of Law No. 12,529/11, which deals with all types of anti-competitive conduct other than merger control. It prohibits acts that ‘have as [their] object or effect’ the limitation, restraint or, in any way, harm open competition or free enterprise; control over a relevant market for certain goods or services; an increase of profits on a discretionary basis; or engagement in market abuse. Its paragraph 3 sets forth a lengthy but non-exhaustive list of conduct that may be considered illegal, to the extent such conduct has the object or effect of distorting competition. The list includes both collusive practices as well as unilateral conduct. As in Brazilian law, in Chile one single provision establishes the prohibition. Article 3 of DL 211/73 contains a general clause allowing the sanctioning of ‘any person that enters into or executes, individually or collectively, any action, act or convention that impedes, restricts or hinders competition, or sets out to produce said effects’. A nonexhaustive list of conduct is also provided in the same article, colloquially known as ‘the letters’ of article 3. Despite the fact the letters use a language rather similar to the European Treaty, most cases are still enforced under the general provision. In Colombia, article 1 of Law No. 155/1959 also has a general section banning ‘… in general all types of practices, procedures or systems tending to limit open competition and to maintain or determine unfair prices’. Decree No. 2,153/92 comprises three specific categories of conducts that are deemed anti-competitive: anti-competitive agreements, anti-competitive acts, and abuse of dominant position. Theoretically, conduct that does not fall under the specific categories may be sanctioned under the general prohibition. However, the competition authority normally only prosecutes conduct that clearly falls under one or more of the 32

Columns Design XML Ltd

See infra, Section 4.

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 10 /

Date: 8/9

JOBNAME: Duns PAGE: 11 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

482 Comparative competition law

specific instances of prohibited conduct.33 In addition, article 5 of Law No. 155/59 prohibits interlocking directorates.34 3.2 The Competition Agency Brazil, Chile and Colombia have established three markedly different models of institutional design.35 Colombia has adopted the archetypical administrative model, whereby a single administrative agency (the Superintendencia de Industria y Comercio, or SIC) is in charge of enforcing, prosecuting and sanctioning.36 In this sense, the system is similar to the EU competition regime. Cases are initiated by private complaint or ex officio by the SIC. The head of the ‘Competition Promotion Office’ within the SIC leads the preliminary investigation and prosecution, a stage that ends with the issuance of a report addressed to the Superintendent recommending either that the case be closed or sanctioning of the investigated parties. The Superintendent may accept the recommendation or make a different decision.37 The SIC lacks independence from central government.38 It is part of the executive branch of the Colombian State, being supervised by the 33

Note the difference from the Chilean approach. The article is similar to section 8 of the US Clayton Act. In Chile, the FNE has issued guidelines on the issue: FNE, Participaciones minoritarias y Directores Comunes entre Empresas Competidoras, November 2013 . In Brazil, interlocking directorates may be analyzed within merger control or through behavioural controls, but there are no specific rules banning them. 35 On models of institutional design in competition law, see J Tapia and S Montt, ‘Judicial Scrutiny and Competition Authorities: The Institutional Limits of Antitrust’ in D Sokol and I Lianos (eds), The Limits of Competition Law (Stanford University Press 2012) 141–157; M Trebilcock and E Iacobucci, ‘Designing Competition Law Institutions: Values, Structure, and Mandate’ 41 Loyola University Chicago Law Journal 455 (2010). 36 Law No. 1,340/2009, which amended several aspects of the competition regime, established in its article 6 that the SIC would be the ‘national competition authority’. 37 Historically, however, the report has usually been endorsed. According to the OECD, ‘Competition Law and Policy in Colombia: A Peer Review’ (OECD 2009) 46, the likelihood of an anticipated termination of an investigation by settlement has significantly decreased. 38 In fact, an ‘Advisory Council’ must be convened and consulted before any fine is imposed (article 24 of Decree No. 2,153/92). This council comprises experts appointed and removable by the President. Their opinions are not binding. 34

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 11 /

Date: 8/9

JOBNAME: Duns PAGE: 12 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 483

Ministry of Commerce, Industry and Tourism. And despite having administrative, financial and budgetary autonomy,39 its head (the ‘Superintendent’) is appointed by the President and may be removed from post at any time.40 By contrast, since 2003 Chile has a model whereby the allocation of powers is ‘distributed’ between two different organisations. On the one hand, an administrative agency (the so-called Fiscalía Nacional Económica, or FNE) investigates and prosecutes cases. Cases may be initiated either by a third-party complaint or on the part of the FNE itself (ex officio).41 Investigations either result in an administrative decision closing the investigation or a referral to the Tribunal de Defensa de la Libre Competencia, or TDLC),42 which acts as decision-making authority, adjudicating the cases in essentially adversarial proceedings. An important feature is the composition of the TDLC: two of the five members must be economists, and the remaining three are lawyers with knowledge of the area.43 The head of the agency is also an expert in competition law. At least in theory, this mode of organisation should enable the employment of deep economic analysis and evidence.

39

Article 1, Decree 2,153/92. The OECD acknowledged this weakness and suggested the SIC should transform into a collegiate body with more independence from the executive branch, more focused on competition and equipped with more investigative and decision-making powers: OECD, ‘Competition Law and Policy in Colombia: A Peer Review’ (OECD 2009) 62. A similar suggestion has been made by local commentators: see Luis A Zuleta and Lino Jaramillo, ‘Marco normativo para la aplicación de la política de competencia en Colombia’ in Centro de Pensamiento en Estrategias Competativas (ed), Competitividad e instituciones en Colombia: balance y desafíos en áreas estratégicas (Editorial Universidad del Rosario 2010) 104–109. In the past, the government has included some proposals in this regard (for example, in its National Development Plan). However, to date there has been no serious attempt to reform the authority to provide it with more independence. 41 Private initiative has generally been a bit more active than public initiative, although the FNE is involved in most of the main cases. See A González, A Micco and C Caicha, ‘El impacto de la persecución pública en los juicios de libre competencia en Chile’ (2013) 132 Estudios Públicos 39–69 . 42 Law No. 19,911/2003 replaced the ‘Competition Commission’, also a special court, with the Tribunal of Defence of Free Competition. 43 Article 6, section ‘b’ of DL 211/73. 40

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 12 /

Date: 8/9

JOBNAME: Duns PAGE: 13 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

484 Comparative competition law

Both the head of the agency and the judges of the tribunal are independent from government.44 The FNE is headed by the National Economic Prosecutor, who is nominated after a public contest mechanism set by law and may only be removed by cause, subject to a prior motion at the Supreme Court. Judges of the TDLC are elected after a public contest and remain in post for a fixed four-year term, which can be renewed only once. Unlike the Chilean design (and similar to the Colombian design), Law No. 12,529/11 in Brazil has consolidated investigative, prosecutorial and adjudicative functions into one agency: the Administrative Council for Economic Defence, or CADE.45 The accumulation of powers does not necessarily diminish the level of specialisation though. Tasks have been clearly separated internally. Currently, CADE has a rather complex structure that includes a tribunal, composed of six commissioners; a President; a General Directorate for Competition; an Economics Department; and two independent offices: Legal Services and the Federal Public Prosecutor’s Office.46 Each branch fulfils a different role. Thus, the independent tribunal adjudicates cases; the General Directorate for Competition is responsible for investigating them; and Legal Services can render legal opinions in all cases before CADE and represent the latter in court.47 CADE is fully independent of government.48 The three competition authorities – the SIC, the FNE/TDLC and CADE – are generally in charge of enforcing competition laws in all markets. In Colombia, the main exceptions are merger control of firms 44

Articles 33 and 39 of DL 211/73. Previously, Law No. 8,884/94 had reorganised CADE as a federal independent agency with the faculty of prosecuting and adjudicating cases. In addition, it had established the functions of two complementary agencies with specific enforcement roles: the Secretariat of Economic Law in the Ministry of Justice (SDE) and the Secretariat for Economic Monitoring (SEAE). The SDE was in charge of monitoring markets, identifying possible infringement and initiating investigations ex officio or upon request of interested parties. These three agencies constituted the so-called ‘Brazilian Competition Policy System’: OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2010) 11. 46 In addition, the Secretariat for Economic Monitoring (SEAE) has residual functions related to competition advocacy. 47 To some extent, such structure resembles the US Federal Trade Commission (FTC). 48 CADE’s Administrative Tribunal members may only be removed by decision of the Senate requested by the President, or by legal cause (article 7 of Law No. 12,529/11). 45

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 13 /

Date: 8/9

JOBNAME: Duns PAGE: 14 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 485

under surveillance of the Superintendence of Finance, and some mergers in the airline market.49 In Brazil, parallel to the administrative system, both the ‘Economic Crimes Law’50 and the ‘Public Procurement Law’51 give federal and State criminal prosecutors the sole enforcement responsibility on criminal matters. This means they may file indictments before criminal courts due to cartel conduct or bid rigging, acting independently of administrative authorities. Also, in regulated sectors, CADE enforces the law but interacts with the respective regulatory body.52 3.3 Administrative Adjudication Within the three countries, proceedings for infringements of competition are administrative in nature.53 So they are subject to ‘administrative adjudication’, i.e., primary adjudication by an agency within the executive branch of government, which can be reviewed by courts at the request of the affected party. As the institutional competition agency models differ greatly in the three countries, so do their respective models of administrative adjudication and the applicable standards of review.54 Both Brazil and Colombia follow the traditional system of external court review, but with differences. In the former, all decisions made by CADE are subject to judicial review before federal courts of first instance if requested by the private litigants. Then, the first instance decision may 49

Articles 8 and 9 of Law No. 1,340/2009. Law No. 8,137/1990, as amended by Law No. 12,529/11. 51 Law No. 8,666/1993. 52 There has been a debate over the competence over the banking sector, especially regarding mergers, since the Central Bank of Brazil claims to exercise the sole authority over competitive issues in this market. OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2005) 83–90. As a result of this unsettled institutional dispute, mergers in the banking sector are actually subject to review by both authorities in separate proceedings: see A Faraco, ‘Aplicação das normas gerais de concorrência nos setores regulados’ in AP Martinez (ed), Temas atuais de direito da concorrência (Singular 2012) 359. 53 However, in Colombia freedom of competition is a constitutional right that may be pleaded through a constitutional action (acción popular). This means judges are competent to decide on infringements of competition law by way of judicial proceedings. There have been several cases decided by courts. However, none of them has concluded that the right was violated. Furthermore, private litigation is still very uncommon and, generally speaking, discussions of competition law’s infringement have taken place within arbitration panels, not regular courts. 54 See generally, P Cane, ‘Judicial Review in the Age of Tribunals’ (2009) 3 Public Law 479. 50

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 14 /

Date: 8/9

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

486 Comparative competition law

be appealed before a regional Court of Appeals by private parties or the agency. Finally, the Court of Appeals’ decision may reach the Superior Court of Justice if the private parties or the agency apply for an appeal.55 Likewise, in Colombia parties may challenge the decision before regular administrative courts.56 Specifically, they are entitled to request the annulment of the primary decision before an Administrative Court, which judgment may be subsequently appealed to the State Council, the highest administrative tribunal in the country.57 Unlike both Brazil and Colombia, Chile has established a procedure whereby final decisions of the TDLC are subject to the direct judicial oversight of the Supreme Court of Justice (the highest court in the country).58 The Supreme Court reviews cases brought exclusively under a special recourse called ‘complaint recourse’ (recurso de reclamación).59 Since the scope of the review is not defined in the competition statute, it has been interpreted in the broadest possible terms. The Supreme Court has typically decided on questions of law, policy or fact, and, on occasions, substituting its judgment for that of the TDLC. That means the reclamación has functioned in practice as an appellate review. 3.4 Sanctions In the three countries, whoever is the decision maker (i.e., CADE, the TDLC or the SIC), it generally has latitude to address the causes of conduct and may impose fines and/or behavioural or structural remedies according to its findings. Sanctions apply to corporations, associations of firms or individuals (normally directors, managers or persons who participated in the infringement). 55 OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2010) 39–43, 53–56: for a detailed description of the procedures in antitrust conduct cases and the judicial review of government agency actions. In general, CADE prevails in the vast majority of the cases reviewed by courts: AP Martinez, ‘Brazil’ in Maurits Dolmans (ed), The Dominance and Monopolies Review (Law Business Research 2014) 41. 56 Since the Superintendent’s decision is an administrative act, they may also file for reconsideration (recurso de reposición) before the same authority. 57 The State Council has not annulled any decision so far. See infra, Section 4. 58 The Third Chamber on Constitutional Matters hears the competition cases. These cases, however, represent a very small share of this chamber’s workload; rather, the bulk of it comprises cases involving administrative law. 59 This means there is no reconsideration for final judgments, as in the Colombian case.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 15 /

Date: 8/9

JOBNAME: Duns PAGE: 16 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 487

Regarding fines, Brazil follows the model applied in many jurisdictions worldwide, based upon the turnover of the firm. For corporations, they range between 0.1 and 20 per cent of the company’s or group of companies’ pre-tax turnover in the economic sector affected by the conduct in the year prior to the beginning of the investigation. Nonetheless, CADE may resort to the total turnover whenever information on revenue derived from a relevant ‘sector of activity’ (as broadly defined by Resolution No. 3/2012)60 is unavailable. The fine must be no less than the amount of harm resulting from the conduct. However, due to challenges associated with quantifying damages, CADE has never calculated the harm caused by an instance of anti-competitive conduct to determine the applicable fine. Fines imposed for recurring violations must be doubled. In practice, CADE has been imposing fines of up to 5 per cent of the company’s turnover in connection with abuse of dominance violations; the level of fines being considerably higher in hard-core cartel cases (average of 15 per cent of the company’s turnover). Unlike in Brazil, in both Chile and Colombia fines are based on a fixed amount established in the Act.61 In Chile, the amount depends on the financial benefit received from the infringement, the severity of the breach and the offenders’ recidivism. The maximum fine is 30,000 ‘annual tax units’ (approximately US$26 million) for cartel offences and 20,000 annual tax units (approximately US$18 million) for other infringements.62 This maximum amount has been applied only in one cartel case.63 Currently, there are discussions to introduce a new fine system based upon the turnover of the firm. Finally, in Colombia, the maximum fine is 100,000 minimum monthly wages (approximately US$25,000,000) – a figure that has not been applied to any single infraction so far. However, in a recent case the SIC sanctioned a single

60

Resolution No. 3/2012 broadly defines 144 ‘sectors of activity’, which include, among others, beverages and agriculture. 61 For a criticism in the Chilean context, see J Tapia, ‘La aplicación de Multas a Agentes Económicos en el Derecho Chileno de la Libre Competencia: una Propuesta Metodológica’ (2013) 132 Estudios Públicos 71–105. 62 Tax units are a special monetary measure of value used by Chilean legislation to keep the value of sanctions, exemptions, tax purposes and others, in line with inflation. 63 Pollos (judgment 139/2014). Nonetheless, previously the Farmacias case (judgment 119/2012) ended up with the application of the highest fine, which at that time was only 20,000 annual tax units.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 16 /

Date: 8/9

JOBNAME: Duns PAGE: 17 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

488 Comparative competition law

firm for two different abuses, resulting in a fine that exceeded the statutory limit.64 Other sanctions are also incorporated in the legislation. Cease and desist orders can amend or eliminate anti-competitive acts, contracts, agreements, schemes or arrangements in violation of competition laws.65 In Chile, the TDLC can also order divestiture or dissolution of partnerships, corporations or business companies whose existence rests on anti-competitive arrangements. The Competition Act also gives the TDLC the faculty to propose executive amendments to the legislation.66 In Brazil, apart from fines, CADE may also adopt several other measures, such as ‘naming and shaming’ (e.g., ordering the publication of the decision in a major newspaper at the wrongdoer’s expense, and including the offender’s name in the Brazilian Consumer Protection List), prohibiting the exercise of activities or claiming benefits (e.g., prohibiting an individual from carrying out market activities on its behalf or representing companies for five years; prohibiting the offender from participating in public procurement procedures and obtaining funds from public financial institutions for up to five years; recommending the tax authorities to block the offender from obtaining tax benefits), among others. As in Chile, CADE can also adopt structural remedies, such as corporate spin-offs, sale of assets or transfers of controls.67 In addition, Law No. 64

Resolution No. 53,403/2013, of 3 September 2013, against Comcel (currently, Claro). 65 As with fines, in Colombia the Advisory Council must be convened and consulted before a cease and desist order is applied. In Chile, orders have been applied, for example, in GTD/EFE (judgment 76/2008) and Atrex/SCL (judgment 75/2008), where mandatory requests to modify internal procedures were made to private dominant firms. 66 Chilean Competition Act, article 18.3. The TDLC has used this faculty in several cases. For example, in both Transbank (sentence 29/2005) and CCS I (sentence 56/2007), the Chilean TDLC recommended the sectoral regulator (in both cases the financial authority) to apply the corresponding norms and regulations. Likewise, in Lan Airlines (sentence 55/2007), the TDLC proposed ‘the regulatory changes that were necessary and suitable to favour competition’ to be introduced by the customs agency; instructed the FNE ‘to keep watch the functioning of the airfreight transport market and the custom warehousing market’ and ordered the dominant firm ‘to restructure its tariffs for airfreight transport’, it also imposed several other regulatory measures on the dominant firm. 67 CADE has recently applied these remedies in a landmark decision in the cement sector. Administrative Proceeding No. 08012.011142/2006-79. See in general for a discussion regarding these other types of sanctions, A Faraco and

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 17 /

Date: 13/10

JOBNAME: Duns PAGE: 18 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 489

12,529/11 contains a general provision enabling CADE to impose any ‘sanctions necessary to terminate harmful anti-competitive effects’. Finally, both Brazil and Colombia incorporate criminal sanctions. In the former, the law imposes penalties of two to five years’ imprisonment. In Colombia, article 27 of the Anti-Corruption Statute (Law No. 1,474/ 2011) establishes bid rigging in public auctions as a crime that may be penalised with up to 12 years’ imprisonment. So far, no person has been sentenced for infringing this criminal rule. In Chile criminalisation was abolished in 2003.68 During the previous period there were no convictions.

4. SUBSTANTIVE LAW This section shows comparatively the use of substantive rules in the three countries analysed. As indicated before, due to space constraints the focus is on antitrust – i.e., collusive agreements and unilateral conduct. Merger control regimes are not assessed. 4.1 Collusive Agreements 4.1.1 Law and policy Collusive practices are explicitly recognised in competition statutes in the three countries. All of them ban price fixing, restrictive agreements on production, bid rigging and market partitioning, among other things. Also, legislation prohibits both explicit and tacit mechanisms for collusion.69 In Brazil, all collusive behaviour falls under two broad provisions, targeting respectively hard-core cartels and other concerted practices.70 In both Chile and Colombia, the structure of the prohibition is analogous. It comprises, first, a general, open prohibition against agreements or concerted practices against competition; and, second, an exemplary, non-exclusive list of anti-competitive collusive practices. In Chile, article others, ‘Sanções não pecuniárias por infrações contra a ordem econômica’ (2014) 46 Revista de Direito Público da Economia 9. 68 However, there have been some recent discussions to revitalise it. 69 There has been less prosecution of tacit collusion in Chile than in Colombia and Brazil. See JD Gutiérrez, ‘Tacit Collusion in Latin America: A Comparative Study of the Competition Laws and their Enforcement in Argentina, Brazil, Chile, Colombia and Panama’ in Fox and Sokol (eds) (n 16) 225–239. 70 Law No. 12,529/2011, article 36 paragraphs I and II.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 18 /

Date: 8/9

JOBNAME: Duns PAGE: 19 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

490 Comparative competition law

3 of DL No. 211/73 contains a general prohibition on agreements or practices that have the objective of limiting freedom of competition. In addition, letter ‘a’ in that same article refers specifically to actions, acts or conventions that impede, restrict or hinder competition, or which set out to produce said effects, express or tacit agreements among competitors, or concerted practices between them, that confer on them market power and consist of fixing sale or purchase prices or other marketing conditions, limiting production, allowing them to assign market zones or quotas, exclude competitors or affect the result of bidding processes. In Colombia, article 1 of Law No. 155/59 contains a general prohibition on agreements or practices that have the objective of limiting freedom of competition.71 This was expanded by Decree No. 2,153/92, which holds that agreements may be deemed illegal if their object or effect is contrary to competition. Article 45 defines agreement as ‘any contract, covenant, compromise, or practice that is concerted or consciously parallel among two or more firms’. Then, article 47 contains a non-exclusive list of 11 types of prohibited agreements if their effect or objective is anticompetitive.72 Fighting against collusion is at the core of competition agencies’ priorities in the three countries. However, both the pace at which such task has become central for their respective agencies and the level of success in prosecuting wrongdoers differs greatly. To a great extent (and this is certainly not different from other jurisdictions), the diverse experience with cartel cases is a reflection of the manpower and investigative tools available for the authorities. Brazilian authorities began focusing on cartel conduct by the year 2000. That year, what over time proved to be one of the most important legislative innovations was introduced: the establishment of a leniency programme.73 Structurally, the regime in itself is broadly aligned with other modern antitrust legislation worldwide and academic suggestions. Also, it incorporates the variant of a ‘leniency plus’. The programme was implemented for the first time in 2003. In 2006 the leniency programme was subject to further amendments, when a ‘marker system’ was formally introduced and companies were given the chance to present their initial 71

The SIC assimilates ‘objective’ with ‘potential harm to the market’. For an analysis of each of the anti-competitive agreements established in article 47, see M Velandia, Derecho de la Competencia y del Consumo (2nd edn, Universidad Externado 2011) 127–163. 73 Law No. 10,149/2000. The programme was managed by the former agency SDE: OECD, ‘Competition Law and Policy in Brazil: A Peer Review’ (OECD 2010) 15–17. 72

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 19 /

Date: 8/9

JOBNAME: Duns PAGE: 20 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 491

applications supplying only certain basic information. The 2011 amendment brought about one more significant change: the prohibition against ringleaders benefiting from the programme was removed.74 Since the introduction of the investigative tools, Brazil has been generally recognised as one of the leading jurisdictions in cartel prosecution.75 Arguably, in both Chile and Colombia the slow use of these tools – particularly leniency – has delayed similar success.76 To date, in both countries the leniency programme has not taken off. In Chile, for example, there have been only two cases where firms have applied for the benefits – and one of them was a multinational part of an international cartel already sanctioned in several other jurisdictions, including Brazil.77 The main reason for the failure is that, in practice, the programme does not assure full criminal immunity. Although there is no specific antitrust offence in Chile (i.e., there are no criminal sanctions, even for cartel 74 CADE has also a settlements programme for defendants that do not qualify for leniency. On 7 March 2013, CADE revised its regulation on settlement proceedings regarding pending investigations for alleged anticompetitive behaviour. The revised rules are primarily related to settlements in cartel cases and directly affect pending international cartel investigations. Under the new policy, to settle an ongoing cartel case, defendants (both individuals and companies) must admit their participation in the illegal behaviour, cooperate with the investigation and pay a settlement sum. Differently from the previous system, the new regulation establishes a predetermined level of discounts depending on the moment in time the defendant comes forward and the degree of expected cooperation. A settlement with CADE does not shield individuals from criminal liability and the fact that an admission of guilt is now mandatory requires interested parties to carefully consider the expected effects a settlement at the administrative level might have on the criminal front. 75 Of the 22 cases sanctioned by CADE in 2013, 18 referred to cartel investigations, nine of them regarding the fuel retail sector, while the remaining four cases concerned abuse of dominance, three of which involved exclusivity clauses in the health-care sector. 76 In Colombia, investigative tools were established by Law No. 1340/2009 and developed by Decree No. 2896/2010. In Chile, Law No. 19,911/2003 introduced them. Leniency exists only in its ‘classic’ version – for example, there is neither marker system nor leniency plus. However, in November 2014 the FNE launched for public consultation new Guidance on Leniency Procedures, which incorporates the possibility of leniency plus. 77 The case was Tecumseh (judgment 122/2012). A similar case had been ruled on in Brazil (Whirpool et al. 2009), as a result of a leniency agreement with the competition agency. Thereafter, there were simultaneous dawn raids in the US, the EU and Brazil. The case is considered a milestone in Brazilian anti-cartel policy. In Chile, there is a third, pending case initiated after a firm applied for leniency, in the market of asphalt.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 20 /

Date: 8/9

JOBNAME: Duns PAGE: 21 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

492 Comparative competition law

cases), in the past criminal prosecutors have tried to use a general, old ‘market fraud’ offence incorporated in the Criminal Code as the basis for condemnation in cartel cases. This has produced a negative incentive for parties to apply for leniency, undermining the protection against cartels. Nonetheless, cartel enforcement remains central for authorities. In Colombia, price fixing is the most common infringement investigated by the SIC, amounting to almost half of the proceedings. The majority of cases involve agreements between competitors, with only a handful of vertical price-fixing cases.78 Tacit collusion cases are as common as cases where explicit collusion is investigated.79 Also, during the last years the SIC has increased the number of bid-rigging investigations and cases.80 Unlike both Colombia and Brazil, historically Chile has devoted more attention to cases of abuse of dominance and vertical restraints.81 Cartel prosecution has only recently grown in importance. The situation slowly changed after the introduction of the new investigative tools. Since 2010 the FNE has followed a strategy of prioritisation, whereby goals and priorities are set internally on a yearly basis according to a methodology that weighs a number of criteria and indicators to determine the importance of sectors and the relevance of the conduct in relation to their impact on consumer welfare. Cartels are at the top of the list. In fact, in 2011 cartel enforcement work was moved to a dedicated Cartels Division and more cases have started to be prosecuted (albeit still in a limited number of sectors).

78

In the period 2003–2007, the SIC opened just one investigation on vertical collusion (Resolution 21821/2004, whereby the SIC sanctioned a producer of glass and three transport companies), while 23 cases of horizontal collusion were opened: OECD, ‘Competition Law and Policy in Colombia: A Peer Review’ (OECD 2009) 65. 79 Between 1999 and 2007, the SIC carried out 15 investigations where firms were accused of conscious parallel practices and on seven occasions the authority concluded competition law was infringed: Gutiérrez (n 69) 235. 80 In 2011, the authority published guidelines on detecting bid rigging, financed by the European Union. The same year the same practice was incorporated as crime: Gutiérrez (n 69). The increased attention on bid rigging was a reaction to several corruption scandals thoroughly covered by the press. 81 OECD, ‘Competition Law and Policy in Chile: A Peer Review’ (OECD 2004) 41. Reportedly, between 1973 and 2003 collusion cases represented only 6.3 per cent of the total anti-competitive conduct cases: E Cruz and S Zárate, ‘Building Trust in Antitrust’ in Fox and Sokol (eds) (n 16) 165.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 21 /

Date: 13/10

JOBNAME: Duns PAGE: 22 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 493

4.1.2 Case law analysis A common feature of the three countries’ approach to collusion is the debate regarding proof of market power as a necessary element to determine an infringement of competition law, regardless of the type of conduct that is prosecuted. This has been equated to a requirement to carry out a case-by-case inquiry into the net effects of an agreement. Therefore, the case law has not been clearly framed in the ‘per se’/‘rule of reason’ categories. This has produced an avid debate. While some commentators have tried to demonstrate the existence of a legal base for applying economic standards,82 others have preferred to explain the analysis of anti-competitive conduct in a way that closely resembles criminal law standards.83 In Colombia, the SIC has been reluctant to apply a per se rule and define explicit standards. Resolution 21,821 of 2004 did not accept that horizontal price fixing agreements should be considered illegal per se while vertical price fixing agreements should be weighed in terms of the efficiencies created. Its argument was that the law does not distinguish between both types of agreements.84 In Chile, there seems to be some basis to argue that the TDLC is taking important steps forward towards a position closer to a per se illegality of cartels. However, the TDLC has not fully embraced such rule so far. The case law establishes that in order to prove collusion, the plaintiff must show that the conduct has an ‘objective ability’ to affect competition.85 This was established for the first time in Asfaltos (judgment 79/2008), and has become one of the key arguments used in Chile to deny the application of a per se rule. In two further judgments though (Explora, judgment 113/2011, and Farmacias, judgment 119/2012), the Supreme Court softened the requirement. Later, in ACHAP (sentence 128/2013), the TDLC incorporated the Supreme Court’s reasoning and 82 In Colombia, see Alfonso Miranda, ‘El Régimen General de la Libre Competencia’ in CEDEC III (Javegraf 2002) 35–41. 83 See: in Chile, D Valdés, Libre Competencia y Monopolio (Editorial Jurídica de Chile 2006); in Colombia, Velandia (n 72) 57–68. The influence of criminal law is pervasive in all areas of Latin American competition law. 84 Although it did remind one of the party’s supposed dominance over the market and controverted the alleged efficiencies. Compare OECD, ‘Competition Law and Policy in Colombia: A Peer Review’ (OECD 2009) 19–20. 85 The TDLC assesses the evidence according to ‘due circumspection rules’ (in Spanish, reglas de la sana crítica), which gives ‘the judges the opportunity to assign value to every single proof in harmony with the merit of the process and his or her logic and experience’.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 22 /

Date: 13/10

JOBNAME: Duns PAGE: 23 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

494 Comparative competition law

made further advances towards such rule.86 Finally, in Collico (judgment 141/2014), the TDLC accepted that market power was evident when the agreement had produced effects in the price. CADE is also moving towards a position closer to a per se illegality of cartels. The agency has been adopting a presumption that anticompetitive effects exist whenever there is collusion. This means in practice that in cartel cases CADE is concerned only with evidence of the existence of collusive behaviour, without further evaluating market power or actual effects in the market.87 4.2 Abuse of Dominance88 4.2.1 Law and policy In the three countries analysed, the existence of market power is not considered to be a per se violation of competition laws. The plaintiff must demonstrate dominance and an actual or potential abuse in a relevant market. Regarding dominance, both Brazil and Colombia have statutory provisions referring specifically to the concept. Under the new Brazilian 86

In ACHAP, referring to one of the tendering processes, the TDLC stated that ‘the only effect of the agreement denounced on the tender process was, in practice, that [the conduct] produced an extension of the deadline to make the proposals. Nonetheless, if the FNE had not intervened, it is highly probable that the effect on the tendering process would have been greater’: ACHAP (sentence 128/2013) para 188. Note the TDLC based its conclusion on mere speculation, not on evidence. However, its statement implies that proof of market power is not required, so the standard of proof of the conduct has in practice been lowered vis-à-vis previous cases. Whether this becomes a trend remains to be seen in further cases. 87 For a review of the case law, see AP Martinez, Repressão a cartéis (Singular 2013) 162. 88 See generally: on Chile, S Roberts and J Tapia, ‘Abuses of Dominance in Developing Countries: A View from the South, with an Eye on Telecoms’ in M Gal and others (eds), The Economic Characteristics of Developing Jurisdictions and Their Implications for Competition Law (forthcoming 2015); S Roberts, J Tapia and M Ybar, ‘The Same and the Other: A Comparative Study of Abuses of Dominance in Chile and South Africa’ (2013) Centre for Regulation, Competition and Economic Development, University of Johannesburg, Working Paper 2/2013; J Tapia, ‘“Dime de qué presumes y te diré de qué careces”: el tratamiento jurisprudencial de los abusos de dominancia en Chile’ (forthcoming) 10 (10) Revista de Derecho de la Competencia, CDEC XIV. On Brazil, AP Martinez, ‘Abuse of Dominance: The Third Wave of Brazil’s Antitrust Enforcement?’ (2013) 9(2) Competition Law International 169, 169–181.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 23 /

Date: 13/10

JOBNAME: Duns PAGE: 24 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 495

legislation, dominance is presumed when a company or group of companies holds at least 20 per cent of a relevant market. The expression ‘group of companies’ encompasses firms belonging to different economic groups that could jointly abuse in a relevant market, even though no single member holds market power. The threshold – relatively low compared with the US or the EU – may be changed by CADE ‘for specific sectors of the economy’, although this has not been formally done to date. In Colombia, dominant position is defined in article 45 of decree 2153/1992 as the ‘possibility to determine, direct or indirectly, the conditions of a market’. The definition is broad and allows the authority to shape the criteria to determine if a firm has dominance. In a recent decision, confirming previous doctrine, the SIC considered that the elements that must be analysed include ‘the structure of the market, including market shares and market concentration, the characteristics of the demand in the market, the competitors, the barriers to entry and other factors that allow a firm to act independently in regards to the market’.89 Notwithstanding the broad definition, abuse-of-dominance cases have only been directed against a single firm with market power so far – i.e., there have been no cases on ‘collective dominance’. Unlike both Brazil and Colombia, Chile has no specific statutory requirements for dominance, and the TDLC has not been explicit on what amounts to a dominant position. It has said that the factors affecting the finding of dominance are broad, including market shares, costs, barriers to entry, the degree of technical innovation in the market, and product differentiation, among other things. Arguably, the very nature of the cases presented before the TDLC has decreased the need for adopting clearer rules. A large proportion of the cases have involved firms that were overwhelmingly dominant (i.e., over 75 per cent of market share). In addition, the TDLC has been rather flexible (if not unclear) when defining dominance.90 In addition to dominance, in all the three jurisdictions plaintiffs must prove an abuse of such position in a relevant market whose object or effect is distorting competition. This means that the rule of reason

89

Resolution 3694/2013, 24. For instance, while in some cases it has equated dominance to ‘commercial superiority’ (Voissnet I, judgment 46/2006), in others it has derived dominance from the ability to set prices above competitive levels (Emelat, judgment 93/2010). 90

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 24 /

Date: 8/9

JOBNAME: Duns PAGE: 25 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

496 Comparative competition law

applies,91 although its boundaries are not always clearly set in the case law. The measure of abuse is generally considered objective, so a query for the ‘intent’ is largely irrelevant.92 None of the three jurisdictions have ruled on a case where the harmful conduct was justified by procompetitive efficiencies. In fact, in Chile the negative effects have normally been quite clear and parties have not justified their conduct on efficiencies in any case. The abuse may be exploitative or exclusionary. In Brazil, article 36(3) of the Act contains a non-exhaustive list of acts that may be considered illegal, which includes various types of unilateral abuses. In the Chilean Competition Act, a similar (but not so lengthy) exemplary list is set forth in the letter ‘b)’ of article 3°. In both cases, listed practices include refusals to deal, predatory pricing, resale price maintenance and other restrictions. Unlike Chile or Brazil, in Colombia article 50 of Decree No. 2,153/ 1992 contains an exhaustive list (numerus clausus) of six types of abuses of a dominant position in the market:93 (1) predatory pricing; (2) vertical discrimination (exploitative); (3) tying and bundling; (4) horizontal discrimination (exclusionary); (5) regional predatory prices; and (6) obstruction of the access to markets or commercialisation channels to third parties. Colombian competition law also distinguishes from both Chilean and Brazilian laws in its explicit prohibition of certain unilateral conduct by agents that do not have market power. Thus, article 48 of Decree No. 2,153/1992 deems it anti-competitive: (1) to infringe advertisement rules of the Consumer Protection Statute:94 (2) to influence prices, either to raise them or to refrain from lowering them; and (3) to 91

In Brazil, this is set out in CADE’s Resolution No. 20/99, Annex II, issued under the previous law. 92 This was the position of the Chilean TDLC until recent times. However, CCS II (124/2012) expressly declares that ‘there are no records in the file to conclude that CCS has had the true intention of excluding [its rival] from the market …’: para 30. A similar statement is made at para 32. 93 For an analysis of each of the abuses of dominant position established in article 50, see Velandia (n 72) 165–193. 94 This prohibition overlaps with those contained in both the Consumer Protection Statute (particularly with article 61 of Law 1480/2011) and the Unfair Competition Law (particularly with the conduct prohibited in articles 10–15 of Law 256/1996). Therefore, in theory conduct may originate three different proceedings (in antitrust, unfair competition and consumer protection). In fact, some commentators have criticised the inclusion as an ‘anti-competitive act’, since the ability to restrict competition through an infringement of advertisement rules would be highly improbable, unless there were market dominance. See, for

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 25 /

Date: 8/9

JOBNAME: Duns PAGE: 26 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 497

discriminate or deny contracting or dealing, if such practice is a form of retaliation due to pricing policies. The enforcement of the abuse-of-dominance provision in the three countries is dissimilar. Colombia presents the weakest enforcement in terms of number of cases, with less than one-fifth of the investigations of the SIC being focused on abuse of dominance.95 Conversely, both Brazil and Chile present a large, and increasing, number of cases in this area – arguably indicating some preference for over-enforcement.96 The reasons underlying this large number of cases vary. In Brazil, some commentators have blamed either poor instruction from complainants or lack of resources from the authorities to further investigate the cases and reaching sound conclusions on their effects. In Chile, most cases may be considered ‘simple cases’, in the sense that both the dominance and the effects have been largely clear. Hence, the TDLC has been able to deal with them directly and rapidly.97 Despite the number, abuse of dominance cases have not been part of the enforcement priorities for Brazilian authorities, which have focused their efforts in prosecuting cartels and bid rigging instead. To date, CADE has shelved the majority of cases. By contrast, in Chile this type of conduct has always been important for authorities, with collusion being only recently prioritised. Abuse-of-dominance cases have arguably played an important part in structural controls, given the lack of a mandatory ex-ante system of merger control. 4.2.2 Case law analysis Notwithstanding the number of cases, what transpires from the analysis of the case law is a relatively low number of ‘good’ cases. Generally speaking, there are few examples of each of the kind of conduct traditionally associated with abuses of dominance. In Colombia, for example, Carlos A Uribe, Reforma al Régimen de Libre Competencia en Colombia: Perspectivas y retos (Universidad Javeriana 2009) 81. 95 OECD, ‘Competition Law and Policy in Colombia: A Peer Review’ (OECD 2009) 44. 96 In Brazil, there are around 70 pending investigations for alleged abuse of dominance, including allegations of sham litigation. Relevant unilateral conduct investigations have been settled with CADE, including investigations into the construction, telecommunications, tobacco, banking and financial sectors. The record fine imposed for an abusive practice was 352 million reais in connection with an exclusive dealing case in the beer market. See AP Martinez, ‘Brazil’ in Dolmans (ed) (n 55) 41. 97 Considering only this aspect, the system seems effective. See Roberts and Tapia (n 88).

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 26 /

Date: 8/9

JOBNAME: Duns PAGE: 27 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

498 Comparative competition law

example, all of the conduct listed as abuse of dominance requiring market power has been investigated by the competition authority, but only in some cases has the investigation concluded in penalties.98 Regarding abuses that do not require market power, the outcome is mixed. To date, investigations on the first infringement (against advertising rules) are very infrequent, and the SIC has never imposed a penalty based on it. By contrast, the second prohibition (influencing prices) has been investigated on several occasions with fines imposed on some firms.99 Furthermore, some cases have reached the State Council.100 Finally, the third type of conduct (discrimination or contract denial) has been investigated in few cases by the SIC.101 In both Brazil and Chile the situation is similar. There have been important cases on different areas,102 such as exclusive agreements,103 predatory pricing104 or resale price

98 For example, see Resolution 22,624/2006 (predatory pricing); Resolution 4,285/2002 (vertical discrimination); Resolution 8,328/2003 (tying); Resolution 53,992/2012 (horizontal discrimination); Resolution 15,653/2001 (regional predatory pricing); Resolution 3,694/2013 (obstruction of access); Resolution 4,907/2013 (obstruction of access); and Resolution 53,403/2013 (obstruction of access and misleading information). 99 For example, Resolution 27,263/1999; Resolution 8,231/2001; Resolution 25,420/2002; Resolution 8,310/2003; and Resolution 33,141/2011. 100 For instance, judgment of 19 November 2009. The State Council revised a decision whereby the SIC had fined a company that sent communications to a supermarket informing it that its discounts should not be transferred to consumers. 101 See for example Resolution 8310/2003. 102 Many of the cases across jurisdictions are very much alike. 103 In Brazil, see for example the cases AmBev (Administrative Proceeding No. 08012003805/2004-10) and Souza Cruz & Philip Morris (Administrative Proceeding No. 08012.003921/2005-10). AmBev has challenged the decision before the judicial courts and a final decision is still pending (Judicial Courts, 16th Circuit, 2009.34.00.028766-7). In January 2013 Souza Cruz agreed with CADE to end exclusivity arrangements with their dealers prohibiting the display of their competitors’ products and in-store advertisements. While Philip Morris, which was also the subject of the investigation, settled the case in July 2012 subject to a payment of 250,000 reais, Souza Cruz agreed to pay 2.9 million reais. In Chile, Fósforos (judgment 90/2009), Tabacos I & II (judgments 26/2005 and 115/2011, respectively), Beers (FNE resolution 62/2008) and Coca-Cola (FNE resolution 92/2011). 104 In Chile for example Arauco v D&S (judgment 103/2010). In Brazil, see Annex I of CADE’s Resolution No. 20/1999.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 27 /

Date: 13/10

JOBNAME: Duns PAGE: 28 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Latin American antitrust law and policy 499

maintenance,105 among others, but only a few of them can be considered relevant as precedent. Nonetheless, overall it is possible to observe an increase in the level of sophistication of the decisions made by authorities in the three countries. This has coincided with a transition from a formal-based approach to a more economics-based analysis. Indeed, this does not imply that competition authorities are actually applying that kind of analysis in full – at least in most cases. There is still some way to go. But serious efforts have been made to leave formality behind and embrace thorough economic analysis instead.

5. CONCLUDING REMARKS As in developed jurisdictions, competition law and policy is gradually gaining its central place in the legal and economic order of Brazil, Chile and Colombia. The protection of the competitive process is today essential in the three countries and may serve as a good example for other Latin American jurisdictions. However, competition law remains embedded in a context of changing politics, values and perceptions about the relationship between the State and the economy.

105 For example, in Brazil, see SKF do Brazil Ltda. (Administrative Proceeding No. 08012.001271/2001-44). Note that CADE has changed its approach regarding RPM from ‘rule of reason’ to a modified per se test, in which the conduct is presumed to be illegal, and parties would in theory have an efficiency defence.

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 28 /

Date: 8/9

JOBNAME: Duns PAGE: 29 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: 21_Chapter17

/Pg. Position: 29 /

Date: 8/9

JOBNAME: Duns PAGE: 1 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

Index Abbott Laboratories v Teva 157 A.B. Dick v Henry 143 absolute cost advantages 133 abuse of dominance 148, 158–159, 165, 240–242, 387 ACHAP case 493–494 advertising industry 51 advertising restrictions, in codes of ethics 104–105 airline industry 41–42 Albrecht v Herald Co 254 ALCOA 127–128 Alese, Femi 107 Allied Tube & Conduit Corp. v Indian Head, Inc. 162 allocative efficiencies 110, 111, 113 allocative inefficiencies 90, 111–112 AMD see Intel/AMD antitrust war American Airlines 407 American Banana Co. v United Fruit Co. 351 American Needle v National Football League 89 American Society of Mechanical Engineers v Hydrolevel Corp. 162–163 American Society of Sanitary Engineering, In re 163 amnesty programs 62 Andean Community of Nations (CAN) 472 anti-competitive agreements 4, 56–78 introduction 56–59 agreement distinguished from oligopolistic interdependence 69–73 antitrust treatment of ‘hard-core’ cartels 59–61 circumstantial evidence and 63–66 facilitating practices 73–77 oligopoly and analysis/proof 66–69 proof of cartel agreement 61–62 summary conclusion 77–78

anti-competitive agreements, content of 79–108 introduction 79–82 agreements as competitively neutral 91–92 agreements as pro-competitive 91–92 Article 101 (TFEU) 83–84, 87 assessment of anti-competitiveness in EU 101–103 assessment of anti-competitiveness in US 99–101 Australia law on anti-competitive agreements 85 characterization of 106–107 continuum of modified rules 103–107 definitions 85–89 EU categories of agreements 92, 96–98, 103 EU economic entity doctrine 86–88 market power and 90–91 other categorisation approaches 98–99 other flexible approaches 89 per se anti-competitive categorisation 92–93 prohibited per se categorisation 93–94 summary conclusion 107–108 UK Enterprise Act 84–85 US categories of agreements 92–93, 94–96, 99 US Sherman Act, Section 1 82–83, 85 US single enterprise doctrine 88–89 Antitrust Guidelines for the Licensing of Intellectual Property (US) 145 The Antitrust Paradox: A Policy at War with Itself (Bork) 124 APEC (Asia-Pacific Economic Cooperation) 15, 214, 348 Apple 165–167 Areeda, Philip 104 Argentina criminalization of cartel conduct 328

501

Columns Design XML Ltd

/

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 1 /

Date: 13/10

JOBNAME: Duns PAGE: 2 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

502 Comparative competition law private antitrust enforcement 387 Arizona v Maricopa County Medical Society 80 Arrow, Kenneth 160, 394 artificial entry barriers 124 ASEAN (Association of Southeast Asian Nations) Regional Guidelines on Competition Policy (2010) 17 Asfaltos case 493 assessment of agreements 80–81, 99–103 AstraZeneca 157–159 AstraZeneca v Commission 158–159 AU Optronics 56 Australia approval of Seagate/Samsung merger 205 categories of agreements 92, 103–104 Competition and Consumer Act/ Commission 85, 89, 98–99, 110–111, 113–114, 137, 223, 243, 275–276, 282, 288, 292, 380, 382 competition law, overview 16, 17 criminalization of cartel conduct 57, 312, 326–327 enforcement agency model 272, 275, 286–287 exiting firms 186 extraterritorial reach of 380–382 geographic dimension of markets 41–42 monopoly DWL in 111 predatory pricing case law 139–140 price signalling 296 raising rivals’ costs case law 140 refusals to deal case law 137–139 search and seizure powers 290 Australian Competition and Consumer Commission v Australian Safeway Stores Pty Limited 140 Australian Competition and Consumer Commission v Baxter Healthcare Pty Ltd 119, 139 Ayres, Ian 294

Columns Design XML Ltd

/

Bainian market power see strategic market power Baker, Donald I. 294 Baker, Jonathan B. 30–31, 123 Barnett, Thomas O. 286 BASF Aktiengesellschaft 321, 381 Beaton-Wells, Caron 286–287 Belgium criminalization of cartel conduct 312 Berkey Photo, Inc. v Eastman Kodak Co. 145 BHP Billiton 371, 439 bid rigging 97, 306, 313, 323, 324, 419, 435, 436–437 bilateral agreements 20–22, 209–211, 299 Blair, R.D. 258 block booking 144 block exemption regulation (BER) (EU) 238–240 Blomkest Fertilizer, Inc. v Potash Corp of Saskatchewan 71–73 BMI Music, Inc. v Columbia Broadcasting, Inc. 145 Board of Trade of City of Chicago v US 99–100 Boeing/McDonnell Douglas merger 210 Boral Besser Masonry v Australian Competition and Consumer Commission 139–140 Bork, Robert H. 111, 119–120, 124 Bosch 167 bottleneck method, of direct foreclosure 127 boycotts 96, 104 Braithwaite, John 294 Brandeis, Louis 99–100 Bray v F. Hoffman-La Roche Ltd 381–382 Brazil abuse of dominance 494–499 administrative adjudication 485–486 Administrative Council for Economic Defense (CADE) 193, 200–201, 203, 378, 481, 484, 485–486, 487, 488, 491, 494, 495, 497

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 2 /

Date: 14/10

JOBNAME: Duns PAGE: 3 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Index 503 administrative enforcement model 272 Brazilian Competition Policy System (BCPS) 313–314 collusive agreements 489–494 Competition Act (Law No. 8, 884/ 1994) 478 Competition Act (Law No. 12, 529/ 2011) 478, 481, 484, 488–489, 496 competition agency 482–485 criminalization of cartel conduct 305, 306–307, 312, 313–314, 327–328 current institutional setting 479 Economic Crimes Law 485 extraterritorial reach of 378 Federal Constitution (1988) 479 goals and legal mandates 480–482 history of competition law 476–479 Law No. 4,137/1962 476–477 leniency programs 490–491 multiagency enforcement 281, 290 National Anti-cartel Strategy (ENACC) 327–328 PMN (pre-merger notification) regime 192, 199 private antitrust enforcement 387–388 Public Procurement Law 485 regional overview 472–476 sanctions 486–489 summary conclusion 499 Bresnahan, Timothy F. 123 Breyer, Stephen 68, 257, 357, 361 Broadcast Music, Inc v Columbia Broadcasting System, Inc. 60–61, 106 Brunt, Maureen 120, 134–135 Brussels Regulation 370–371 bundling 139, 222, 233–234 Bush, Darren 296–297 Buxbaum, Hannah L. 349, 371 Cadbury Adams Canada Inc. 380 Cal. Computer Prods., Inc. v IBM Corp. 145

Columns Design XML Ltd

/

Californian Dentists Association case 104–105 Camesasca, Peter D. 90 Canada Budget Implementation Act (2009) 312 Competition Act (1985) 89, 189 Competition Act (2010) 98, 378–379 Competition Bureau (CCB) 275, 279, 282, 290, 379 criminalization of cartel conduct 306–307, 312, 325–326 extraterritorial reach of 378–380 prosecutorial enforcement model 274–275, 289 re-importing of pharmaceuticals 40–41 Canpotex 363 capital requirements, and strategic market power 133 Cardizem CD Antitrust Litigation, In re 152 Caribbean Community (CARICOM) 472 Carlton, Dennis W. 54, 118 cartels antitrust treatment of ‘hard-core’ 59–61 confessions by participants of 58, 61–62 historical background 10–11, 13, 15–16 orderly marketing of 90 proof of cartel agreement 61–62 stereotype of 56 see also criminalization of cartel conduct Caspary, T. 266 categorisation of agreements 80–81, 92–99 Cathode Ray Tubes case 440 cathode-ray tube cartel 57 Cement Manufacturers Association 17 Chicago School IP rights and 144 on NPVRs as procompetitive 224–225, 227, 251

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 3 /

Date: 21/10

JOBNAME: Duns PAGE: 4 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

504 Comparative competition law private actions in Georgetown study and 405–406 on tying 228–229 see also classical market power Chile abuse of dominance 494–499 administrative adjudication 485–486 Cartels Division 492 collusive agreements 489–494 Competition Act (Law No. 211/1973) 477, 481, 489–490 competition agency 482–485 criminalization of cartel conduct 328 current institutional setting 479 FNE (Fiscalía Nacional Económica) 483–484, 492 goals and legal mandates 480–482 history of competition law 476–479 Law No. 13 305, 476 leniency programs 491–492 prosecutorial enforcement model 274–275 regional overview 472–476 sanctions 486–489 summary conclusion 499 China 8, 443–471 abuse of dominance 455–457, 467–468 administrative monopolies 459–461 Antimonopoly Enforcement Authority 449–450, 461 Anti-Monopoly Law (AML) (2008) 17, 314, 375, 446–462, 464, 466, 468, 470–471 Anti-Unfair Competition Law (AUCL) (1993) 444 approval of Seagate/Samsung merger 205 approval of Western Digital/Hitachi merger 205–206 civil court actions 470–471 competition law, overview 443–447 criminalization of cartel conduct 305 enforcement of AML 462–471 extraterritorial reach of 375–377 General Administration of Quality Supervision, Inspection and Quarantine (AQSIQ) 459

Columns Design XML Ltd

/

Guidelines on Antimonopoly Law Enforcement in the Field of Intellectual Property Rights 469 immunity from prosecution 285 legal liabilities 461–462 merger remedies 205 merger review 445, 457–459 Ministry of Commerce (MOFCOM) 193, 205–206, 284, 376–377, 445, 450, 462, 463–465 Ministry of Information and Industry 468 monopolistic agreements 450–454 multiagency enforcement 283–284, 295–296 National Development and Reform Commission (NDRC) 284, 375, 445, 462, 465–468 PMN (pre-merger notification) regime 192 Price Law (1998) 375, 444, 445, 466 Procurement and Bidding Law (1999) 444 prosecutorial enforcement model 274–275 Provisional Rules on Mergers and Acquisitions of Domestic Enterprises by Foreign Investors (2003) 445 resale price maintenance 451–452 Rules on Mergers and Acquisitions of Domestic Enterprises by Foreign Investors (M&A Rules) (2006) 445 Rules on Prohibiting Regional Blockades in Market Economic Activities 445 search and seizure powers 290 State Administration for Industry and Commerce (SAIC) 284, 445, 450, 462, 468–469 State Administration of Radio, Film and Television 468 state-owned enterprises (SOEs) 443–444, 446–447, 449, 464–465, 468 summary conclusion 471 China Construction Bank 456

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 4 /

Date: 21/10

JOBNAME: Duns PAGE: 5 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

Index 505 China Mobile 456 China National Petroleum Corporation (CNPC) 466 China Netcom 456–457 China Telecom 468 China Unicom 468 CIBA Vision Vertriebs GmbH, Großostheim 266 Ciprofloxacin Hydrochloride Antitrust Litigation, In re 152 class actions 402 classical market power 115–124 comparison to strategic approach 135–137 exercise of 118–120 identification of 120–121 measurement of 121–124 power over price 115–117 underlying assumptions 117–118 cluster market concept 33–34 Coate, Malcolm B. 47–48 Coca-Cola/Huiyuan Juice merger 463–464 codes of ethics, restrictions on advertising 104–105 Collico case 494 collusion 126–131 Colombia abuse of dominance 494–499 administrative adjudication 485–486 Anti-Corruption Statute (Law No. 1,474/ 2011) 489 collusive agreements 489–494 Competition Act (Law No. 155/1959) 476, 477, 478–479, 481–482, 490 competition agency 482–485 Constitution (1991) 478, 479 current institutional setting 479 goals and legal mandates 480–482 history of competition law 476–479 Ministry of Commerce, Industry and Tourism 482–483 regional overview 472–476 sanctions 486–489 SIC (Superintendencia de Industria y Comercio) 482–483, 484,

Columns Design XML Ltd

/

487–488, 492, 493, 495, 497, 498 Special Decree No. 2, 153/1992 478–479, 496–497 summary conclusion 499 comity 349, 350–351, 355–359 commerciality and markets 49–50 Competition Law and Policy Committee (CLPC) (OECD) 14 concentration ratio 123 Connor, John M. 305, 315, 316–317, 319–320 consumer surplus 118 consumer welfare standard 176 Container Corp. case see United States v Container Corp. Continental TV, Inc v GTE Sylvania 144, 231, 232, 251, 252–253 Conwood Co. v US Tobacco Co. 396–397 Copperweld Corp. v Independence Tube Corp. 88 Courage Ltd v Crehan 368, 371 Cournot model 68 court decisions commerciality and substitutionability case 50 on fines for cartels 56 on geographic boundaries of markets 41–42 Gillette case 54–55 on price fixing 60–61 Staples case 47–48 see also individual courts and cases CPTM 166–167 Crane, Daniel A. 275, 279, 280, 283, 284, 296 credit card interchange fees 50–51 Crest Theater 65–66 criminalization of cartel conduct 6, 18, 301–344 introduction 301–305 changes in the law 305–315 country comparison 341–344 development in other countries 312–315, 325–329 dual criminality 335–337

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 5 /

Date: 13/10

JOBNAME: Duns PAGE: 6 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

506 Comparative competition law enforcement and policy developments 315–329 EU/EU member state change in the law 305, 309–311 EU/EU member state enforcement developments 317–318, 322–325 extradition and Ian Norris case 335–337 impacts of Vitamins Cases 320–322 impediments to implementation 338–340 international cooperation 329–335 leniency programs 337–338 summary conclusion 340 US change in the law 307–308 US enforcement developments 318–320 critical loss analysis and SSNIP 48–49 customer restraints 223, 231 Data General Corp. v Grumman Systems Support Corp. 146 deadweight loss (DWL) 109, 110, 111 Dell, In re 163 demand-side substitutionability 44 Denmark prosecutorial enforcement model 274–275 deregulation 133 deterrence claims see private antitrust enforcement developing countries and competition law 18 direct foreclosure method 126–127, 128 Direx case 425 discriminatory pricing 118, 127 domestic jurisdiction see international governance of competition and extraterritorial jurisdiction double marginalisation 259–260 Dr Miles Medical Co v John D Park & Sons Co 249–250, 252, 254, 451 Dreyer’s Grand Ice Cream, Inc. 53 Dyestuffs case see Imperial Chemical Indus. v Commission dynamic bundling 222 dynamic efficiency 80, 91–92, 110, 394

Columns Design XML Ltd

/

dynamic injuries 393–395 Dynamic Random Access Memory (DRAM) chips 323, 379–380 E. Bement & Sons v National Harrow Co. 143 Eastman Kodak 146–147 e-commerce 40 economic entity doctrine (EU) 87–88 economies of scale 132–133 Edion case 424 effects doctrine, in US 351–355 see also private antitrust enforcement Egypt criminalization of cartel conduct 329 Department to Protect Competition and Prohibit Monopoly 329 Elzinga, Kenneth G. 48–49, 406 Emmerich, V. 263–264 Empagran case see F. Hoffman-La Roche Ltd. v Empagran S.A. enforcement see public enforcement enforcement pyramid approach see public enforcement Estonia criminalization of cartel conduct 324 multiagency enforcement 281 Etablissements Consten SARL and Grundig-Verkaufs-BmbH v Commission 235–236 Ethyl Corp. v Federal Trade Commission 76–77 EU Guidelines on competitor collaboration 86 on degree of control 87–88 infringement by object/effect 96–98, 101–103 European Commission (EC) 156, 158–159, 165–166, 169 Director-General for Competition (DG Competition) 273–274, 276–277, 289–290, 310–311 enforcement by 236–237, 238, 318, 364–365, 432 EU Article 82/102 Guidelines 242 on failing firms 186 fine against Intel 410 Green Paper (2005) 384, 386

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 6 /

Date: 13/10

JOBNAME: Duns PAGE: 7 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

Index 507 Recommendation on Collective Redress 369 on RPM 247–248, 261–262 vertical, use of term 220 Vertical Guidelines 237, 240 White Paper (2008) 384, 386, 400–403 European Competition Network (ECN) 310–311 European Convention on Human Rights (ECtHR) 336 European Court of Human Rights (ECHR) 336 European Court of Justice (ECJ) on cartel horizontal agreements 65 on infringement by object/effect 96 private antitrust cases 386 on single economic entity doctrine 367 on undertakings 87 see also individual cases European Court of Justice (EJC) on ‘pure’ oligopoly pricing 68–69 European Economic Community (EEC) 11, 365 see also European Union (EU) European Telecommunications Standards Institute (ETSI) 165 European Union (EU) administrative enforcement model 272–274 approval of Google/Motorola merger 206 approval of Seagate/Samsung merger 205 Article 101 (TFEU) 59, 74–75, 77, 83–84, 87, 96–98, 101–103, 156, 166, 169, 258, 260, 263, 364–368, 384 Article 101(1) (TFEU) 96–98, 101, 166, 235–237, 265 Article 101(3) (TFEU) 97, 103, 235, 237–240, 259–260, 262–264, 265 Article 102 (TFEU) 165–166, 240–242, 364–368, 384, 425 block exemption regulation (BER) 238–240

Columns Design XML Ltd

/

categories of agreements 92, 103 categories of anti-competitive agreements 96–98 competition law in member states 281 competition law, overview 16 cooperation agreements with US 21–22 criminalization of cartel conduct 309–311, 317–318, 322–325 economic entity doctrine 86–88 EU Guidelines 86, 87–88, 96–98, 101–103 extraterritorial reach of 364–371 fines on cartels 57 FRAND (fair, reasonable, and nondiscriminatory) agreements 162, 164–167 Guidelines on Vertical Restraints 225–227 immunity from prosecution 285–286 implementation test 364–366 imposition of structural undertakings on Western Digital/Hitachi merger 206 inter-enterprise agreements 86–87 Merger Guidelines 366 non-price vertical restraints (NPVRs) 234–242 on price fixing 60 private actions and jurisdiction 368–371 private antitrust enforcement 386, 393, 403, 437 resale price maintenance (RPM) 258–266 restrictions on acquisition 221–222 restrictions on resupply 223 search and seizure powers 290 single economic entity doctrine 367–368 Evans, David S. 52 Eversley, DeCourcey 271 exclusionary conduct 119–120 see also strategic market power exclusionary conduct, Salop’s four-step analysis 132 exclusive dealing see non-price vertical restraints (NPVRs)

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 7 /

Date: 13/10

JOBNAME: Duns PAGE: 8 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

508 Comparative competition law exclusive distribution 221–222 extradition and criminalization of cartel conduct 335–337 extraterritorial jurisdiction see international governance of competition and extraterritorial jurisdiction F. Hoffman-La Roche Ltd. v Empagran S.A. 356–357, 361–362, 379, 411–412 F. Hoffmann-LaRoche Ltd 321 facilitating practices 73–77 failing firm defence 185–187 Festival Records 50 First, Harry 340 first principles approach to prevailing price (Salop) 133–134 Fischer, Jeffrey H. 47–48 fixed RPM see resale price maintenance (RPM) flat panel display companies 466 Focus Media 464 Fox, Eleanor M. 298 France criminalization of cartel conduct 306–307 multiagency enforcement 281 FRAND (fair, reasonable, and nondiscriminatory) agreements 162, 164–167 FTC v Actavis 153–154 FTC v Brown Shoe Co 231–232 FTC v Superior Court Trial Lawyers Association 93 FTC v Watson Pharmaceuticals, Inc. 152–153 Fukui Prefectural Federation of Agricultural Cooperatives case 421–422 full-line forcing 222 Gans, Joshua S. 111 Gas Insulated Switchgear global cartel 366 GE/Honeywell merger 210 Gellhorn, E. 252 Gencor Ltd v Commission 366

Columns Design XML Ltd

/

generic drugs, delayed entry see pharmaceutical industry geographic dimension of markets 39–42 Georgetown Private Antitrust Litigation Project 404–406 Geradin, Damien 300 Germany criminalization of cartel conduct 306–307, 312, 324 Law against Restraints on Competition (GWB) 371 minimum RPM 266 Geroski, Paul 36 Gilbert, Richard J. 38 Gillette Company 54–55 Giovannetti, E. 261–262, 267–268 Glencore/Xstata merger 375–376 Global Financial Crisis 296–297 global LCD (liquid crystal display) cartel 374–375 globalisation and e-commerce 40 globalization of competition law 9–24 introduction 9–10 enhanced international competition 22–23 global competition law framework 15, 16–20 global nature of competition 14–16 historical background 10–14 multilateral vs. bilateral agreements 20–22 summary conclusion 23–24 Goldfarb v Va State Bar 60 Google 166–168 Google/Motorola merger 206 governance see international governance of competition and extraterritorial jurisdiction; individual organizations and countries graphite electrode cartel 327 Griffith, Rachel 36 Guidelines for the Licensing of Intellectual Property (US) 37–38 Gyselen, L. 259 Hammond, Scott 303, 320–321, 333, 337–338

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 8 /

Date: 13/10

JOBNAME: Duns PAGE: 9 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Index 509 hard core cartels, use of term 302 hard disk drive (HDD) market 205 Harding, Christopher 306–307, 315 Harris, Barry C. 48 Hartford Fire Insurance v California 352–353, 357–359, 360, 363–364, 379 Hausman, Jerry 122 Hay, George A. 120, 122 Haynes, J.S. 258 HBOS 297–298 Henry, David 300 Herfindahl-Hirschman index (HHI) 123 Hershey Canada Inc. 380 Hicks, Robert 138–139 High Fructose Corn Syrup case 70 Hitachi 205 Hoffmann-La Roche 381 Hoffmann-La Roche v Commission 241 Hogarty, T.F. 48–49 Holmes, Oliver Wendell 351 Horizontal Merger Guidelines (US) 52–53 Hydrotherm Gerätebau GmbH v Compact del Dott. Ing. Mario Andreoli & C. Sas. 367 Hyman, David A. 271–272 Iacobucci, Edward M. 276, 279 IBM Corporation 145 Iceland criminalization of cartel conduct 312 Image Technical Services, Inc. v Eastman Kodak Co. (Kodak II) 146–147 Imperial Chemical Industries Ltd v Commission 65, 367 IMS Health GmbH & Co. OHG v NDC Health GmbH & Co. KG 148 InBev/Anheuser-Busch merger 463 Independent Service Organizations Antitrust Litigation (Xerox), In re 147 India Competition Act (2002) 377–378 Competition Act (2007) 17, 314–315 Competition Commission of India (CCI) 17, 203–204, 284

Columns Design XML Ltd

/

competition law enforcement 296 criminalization of cartel conduct 305, 314–315 extraterritorial reach of 377–378 PMN (pre-merger notification) regime 192 private antitrust enforcement 389 Indonesia competition law, overview 17 criminalization of cartel conduct 312 industry structure actions against self-interest and 72–73 plus factor and 70–71 ‘The Informant’ (film) 320 innovation markets 37–38, 100–101, 160–161 see also research and development (R&D) Intel Corp. v Advanced Micro Devices, Inc. 410–411 Intel/AMD antitrust war 410–411 intellectual property rights, antitrust treatment of 141–170 introduction 141 history of 142–146 innovation markets 160–161 patent pools 168–169 pharmaceutical industry 149–151 pharmaceutical product hopping in EU 158–160 pharmaceutical product hopping in US 156–158 pharmaceutical settlements in EU 154–156 pharmaceutical settlements in US 151–154 refusals to license in EU 148 refusals to license in US 146–147 smartphone standard-setting 164–168 standard-setting, generally 161–164 summary conclusion 170 Intel/McAfee merger 210 Intergovernmental Group of Experts on Competition Law and Policy (UNCTAD) 13

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 9 /

Date: 14/10

JOBNAME: Duns PAGE: 10 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

510 Comparative competition law International Chamber of Commerce (ICC) 211 International Competition Network (ICN) 173, 174, 211, 212, 216, 298, 301, 304, 307, 315–316, 348 Cartel Working Group 332–333, 339 cooperation among enforcement agencies via 299–300 criminalization of cartel conduct 329–330, 332–335 ‘Defining Hard Core Cartel Conduct, Effective Institutions, Effective Penalties’ 333 on leniency programs 336 Manual on Anti-Cartel Enforcement Techniques 333 on merger efficiencies 189–190 on merger remedies 205, 207 on notification thresholds 197 Recommended Practices for Merger Analysis 182–184 scope of 14, 332–335 Statement of Achievements (2001-2013) 22 on timing of notification 199–200 on transparency in merger review 208 unjust enrichment principles 401–402 international governance of competition and extraterritorial jurisdiction 6–7, 345–383 introduction 345–346 concurrent jurisdiction 345–346 extraterritorial reach of Australia 380–382 extraterritorial reach of Brazil 378 extraterritorial reach of Canada 378–380 extraterritorial reach of China 375–377 extraterritorial reach of EU competition law 364–371 extraterritorial reach of India 377–378 extraterritorial reach of Japan 372–373 extraterritorial reach of South Korea 373–374

Columns Design XML Ltd

/

extraterritorial reach of US antitrust law 351–364 role of domestic jurisdiction in 347–351 summary conclusion 382–383 International League of Competition Law (LIDC) 211 International Salt Co. v United States 143–144 International Trade Organization (ITO) Draft Havana Charter 10–11 Interstate Circuit, Inc. v United States 63–65, 69–72 IP rights see intellectual property rights, antitrust treatment of Ireland criminalization of cartel conduct 307, 323–324 Israel criminalization of cartel conduct 312, 328–329 prosecutorial enforcement model 274–275 Japan 8, 415–442 abuse of superior bargaining position 423–426, 432 administrative enforcement model 272, 275, 278, 428–436 Antimonopoly Act (AMA) (1947) 371–373, 388–389 Anti-Monopoly Law (2009) 416–428, 431 Anti-Monopoly Law (2015) 434, 436–437, 440–441 appeal process 434–435 approval of Seagate/Samsung merger 205 cease-and-desist orders 431 competition law in cross-border cases 439–440 criminal enforcement 438 criminalization of cartel conduct 307, 312–313, 327 enforcement record of JFTC 435–436 ex ante merger control 439 extraterritorial reach of 372–373 fines/surcharges 431–433

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 10 /

Date: 14/10

JOBNAME: Duns PAGE: 11 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

Index 511 goals of competition law 416–417 history of competition law 415–416 horizontal restraints 417–420 investigatory powers 429–430 Japan Fair Trade Commission (JFTC) 282–283, 285–286, 290, 295, 371–372, 416, 418, 419, 420–422, 424–436, 438–442 leniency system for cartels 433–434 Liberal Democratic Party 440–441 merger control 427–428 Merger Guidelines 427 Ministry of International Trade and Industry 441 pre-order right to be heard 430 private antitrust enforcement 436–437 private monopolization 420–422 Subcontract Act 424–425, 426, 436 summary conclusion 440–442 unfair trade practice prohibition 422–426 JASRAC case 421 JCB Service v Commission 258–259 Jebsen, Per 122 Jefferson Parish Hospital District No 2 v Hyde 232–233 Johnson & Johnson Medical (China) Ltd 452 Joint Group on Trade and Competition (JGTC) (OECD) 14 joint ventures 199 definitions 85–86 EU economic entity doctrine and 87–88 hard-core cartels and 60–61 noncompete agreements 97 US single enterprise doctrine and 88–89 Jorde, Thomas M. 37 juries, in criminal cases 62 jurisdiction, use of term 346 Kansas Kansas Restraint of Trade Act (2013) 256–257 Kaplow, Lewis 69 Kauper, Tom 406

Columns Design XML Ltd

/

K-Dur Antitrust Litigation, In re 153 keiretsu (network) 426 Kiefer-Stewart Co v Joseph E Seagram & Sons 254 Klein, B. 267–268 Kloth v Microsoft 395 Koh, Harold 340 Kovacic, William E. 164, 271–272 Kraft Export Association (KEA) 365, 366 Krattenmaker, Thomas G. 113, 125, 126, 127–129, 131–132, 135 Kroes, Neelie 368 Lande, Robert H. 111, 112 LCD cartel 56 Leegin Creative Leather Products, Inc. v PSKS Inc. 254–257, 264, 267–268, 451 Lemley, Mark A. 92 LePage’s Inc. v 3M 396 Lerner Index 121 Les Laboratoires Servier case 155, 156 Leslie, Christopher R. 92 LG 466 Ling, Dong 299 Lloyds 297–298 Lombardo, J. Neil 126, 129–130 Lunbeck 156 lysine cartel 61, 320–322 Majoras, Deborah 164 Malaysia Competition Act (2010) 85, 89, 98 competition law, overview 17 non-price vertical restraints (NPVRs) 243 Manfredi v Lloyd Adriatico Assicurazioni SpA 368 marginal consumers 45 marginal cost 121–122 marine hose cartel 328, 334, 336–337, 374 Marine Hose case 324, 326, 431–432, 439–440 market dominance test (dominance test) 181–182 market identification, sources of 45–46

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 11 /

Date: 13/10

JOBNAME: Duns PAGE: 12 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

512 Comparative competition law market power, classical vs. structural approaches 5, 109–140 market power, regulation and establishment of 109–140 bundling 139 classical approach to 115–124 comparison of approaches 135–137 definitions 114–115 efficiencies 109, 111 predatory pricing 139–140 prioritization of concerns 113–114 problematic aspects 109–114 raising rivals’ costs 140 refusals to deal 137–139 rent-seeking behaviour 112–113 strategic approach to 125–135 transfers 109–112, 114 market sharing 97 markets and market power 4, 27–55 commerciality and 49–50 critical loss analysis and SSNIP 48–49 diminishing role for market definition 52–55 evidentiary basis for market definition and SSNIP 44–46 geographic dimension of 39–42 natural experiments and SSNIP 47–48 nature of markets 28–32 price correlations and 48 product markets 32–39 purpose of market definition 27–28 SSNIP test/hypothetical monopolist test 34–39 supply chain and 42–44 supply-side substitutionability 30–32 two-sided platforms/markets 50–52 Mars Canada, Inc. 380 Maryland Maryland Antitrust Act (2009) 255–256 Matra Hachette v Commission 261 maverick firms 187 maximum RPM see resale price maintenance (RPM) Melway Publishing Pty Limited v Robert Hicks Pty Limited 138–139

Columns Design XML Ltd

/

Mercosur 472 merger control rules 19 merger law 171–218 introduction 171–173 administrative structure and agency powers 192–193 administrative treatment 191–192 analytical approach 182–191 common filing form concept 215 cooperation, convergence, conflict 208–214 failing firm defence 185–187 international code for review 215 law and procedure differences 214–217 market definition and 187–188 maverick firms 187 national goals for international markets 179–181 rationale for regulation of 173–179 remedies 204–207 role of efficiencies in analysis 189–191 substantive law and analysis 181–182 summary conclusion 217–218 theories of harm 184–185 timeline for merger review 202–204, 217 transparency 207–208, 216 see also notification obligations (PMN), in merger law Merger Streamlining Group 211 mergers Baker on 125 diminishing role for market definition and 52–55 FTC challenges of 161 SSNIP test/hypothetical monopolist test and 36 US Guidelines on 85–86 see also individual companies/ corporations Mexico administrative enforcement model 272 criminalization of cartel conduct 313, 328 Microsoft 167, 394–395

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 12 /

Date: 13/10

JOBNAME: Duns PAGE: 13 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

Index 513 Microsoft Corp. v Commission of the European Communities 148, 242 minimum quantities condition 222 minimum RPM see resale price maintenance (RPM) Minn-Chem, Inc. v Agrium Inc. 363 Model Law on Competition (UNCTAD) 13, 213–214 monopoly thresholds 122 Monsanto v Spray Rite Serv Corp 252, 254 Monti, Mario 368 Morgan Crucible 336 Morganite 336 most-favored-nations clause (MFN) 76 Motion Picture Patents Co. v Universal Film Manufacturing Co. 143 Motorola Mobility (MMI) 165–168 movie theater industry 63–66 multilateral agreements 20–22, 211, 299 multilateral competition agreement proposals 347 music recording industry 50 Mutual Legal Assistance Treaties (MLATs) 334 ‘naked’ exclusionary right 127–128 National Collegiate Athletic Association v Board of Regents of the University of Oklahoma 104–105 national courts, jurisdiction over Article 101 (TFEU) 103 National Society of Professional Engineers v US 94–95, 99–100, 103, 105 natural experiments and SSNIP 47–48 NCAA (National Collegiate Athletic Association) 60–61 Negotiated Data Solutions (N-Data) 163–164 Nestlé Canada Inc. 380 Nestle Holdings, Inc. 53 ‘new sovereignty’ concept 349 New Zealand commerciality and substitutionability case 50

Columns Design XML Ltd

/

Competition Act 41–42 criminalization of cartel conduct 312 Nine West Group, Inc., In the Matter of 255 Nippon Steel and Sumitomo Metal Industries merger 427 Noel, Michael 52 noncompete agreements 97 non-essential patents (non-SEPs) 164–165 non-price vertical restraints (NPVRs) 219–244 introduction 219–220 definitions and taxonomy 220–224 economics of 224–229 EU approach 234–242 foreclosure barrier to entry 228 leveraging barrier to entry 228–229 other jurisdictions 243 restrictions by acquirers 223–224 restrictions on acquisition 221–222 restrictions on resupply 223 summary conclusion 243–244 US approach 229–234 Norris, Ian 335–337 North American Free Trade Agreement (NAFTA) 301 Northern Pacific Railway v US 93 Norway criminalization of cartel conduct 307, 312 notification obligations (PMN), in merger law 193–202, 216 fees 202 information requests 200–201, 216–217 thresholds for notification 197–199 timing of notification 199–200 voluntary vs. mandatory regimes 194–197 Novell 166–167 NT Power case see Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd NTT East case 421 O’Brien v Leegin Creative Leather Products Inc. 256–257

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 13 /

Date: 13/10

JOBNAME: Duns PAGE: 14 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

514 Comparative competition law oligopoly and analysis/proof 66–69 orderly marketing of cartels 90 Organisation for Economic Co-operation and Development (OECD) 11, 173, 211, 216, 298, 301, 304, 315–316, 348 Competition Committee 300 Competition Law and Policy Committee (CLPC) 14, 330 Council Recommendation Concerning Effective Action against Hard Core Cartels 330 criminalization of cartel conduct 312, 329–331 Global Forum on Competition 331 International Enforcement Cooperation (2013) 22–23 Joint Group on Trade and Competition (JGTC) 14 Model Law on Competition 13 recommendations on cartels 13 scope of 212–213, 334–335 output restriction method 126–127, 128–129 Padilla, Atilano Jorge 31 Pareto optimality 111 Parker Pen Holdings 54–55 patent pools 168–169 patent-antitrust intersection see intellectual property rights, antitrust treatment of Peeperkorn, L. 262, 267 People’s Republic of China (PRC) see China per se/rule of reason categorisation 4, 60–61, 92, 103–108 in EU 96–98 in other jurisdictions 98–99 in US 94–96, 99–101 see also intellectual property rights, antitrust treatment of; non-price vertical restraints (NPVRs) perfect competition model 118 Perloff, Jeffrey M. 118 Peru private antitrust enforcement 387 pharmaceutical industry

Columns Design XML Ltd

/

challenged merger attempts 161 in China 467 citizen petitions and 151 follow-on products 159–160 geographic dimension of markets 40–41 innovation markets and 161 IP rights 149–151 product hopping in EU 158–160 product hopping in US 156–158 reverse payments 151, 408 settlements in EU 154–156 settlements in US 151–154 vitamins cartel 326, 327, 374, 381–382, 391 Vitamins Cases 320–322 Polygram Holding Inc v FTC 105–106 Portugal multiagency enforcement 281 Posner, Richard 69, 70, 252–254 power over price see also classical market power predatory pricing 118–119, 139–140, 407 price correlations 48 price fixing 59–60, 97, 102, 106, 144, 335–337, 338–339, 379–380 see also individual cases private antitrust enforcement 7, 384–412 introduction 384–385 compensation 389–395 cross-border antitrust litigation 410–412 deterrence 389–390, 395–400 dynamic injuries 393–395 in EU 386, 393, 403 expansion of antitrust constituency 409 judicial backlash to 404–407 jurisdiction and extraterritoriality 411–412 prioritization of compensation/ deterrence 400–404 in South Africa, Asia, and South America 387–389 spillovers to public enforcement 407–409

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 14 /

Date: 13/10

JOBNAME: Duns PAGE: 15 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

Index 515 summary conclusion 412 in US 385–386, 390–393, 403–404, 410 product hopping see pharmaceutical industry product vs. economic markets 32–39 product-differentiation advantages 133 productive inefficiency 110 Professional Engineers case see National Society of Professional Engineers v US public education, on cartel offences 339 public enforcement 271–300 introduction 271 administrative model 272–274, 276–278 in China 283–284 compulsory acquisition of information/documents 289–290 detection and leniency policies 284–288 enforcement agency models 271–284 enforcement pyramid approach 291–294 during financial crises 296–297 in India 284 international cooperation 298–300 investigatory powers 289–290 in Japan 278, 282–283, 285–286, 295 multi-agency models 280–282 political influences and 295–298 promotion of compliance 291–295 prosecutorial model 274–276, 279 search and seizure powers 290 publishing industry 249, 470 ‘pure’ oligopoly pricing 68–69 pure vs. mixed bundling 222 Queensland Wire Industries Pty Ltd v Broken Hill Proprietary Co Ltd 137–138 Ralse case 424–425 Rambus, In re 163 RAND (reasonable and nondiscriminatory) agreements 162, 164

Columns Design XML Ltd

/

Rapp, Richard T. 38, 123–124, 133 real foreclosure method, of direct foreclosure 127–128 Reckitt Benckiser 159 recommended RPM see resale price maintenance (RPM) reduced intra-brand price competition 246–247 refusals to deal 137–139 refusals to license in EU 148 in US 145, 146–147 regional protectionism see China re-importation 40–41 remedies, in merger law 204–207 rent-seeking behaviour 112–113 resale price maintenance (RPM) 245–268 introduction 245–246 anti- and pro-competitive effects 246–249 in EU 258–266 summary conclusion 266–268 in US 249–258 resale prices 16–17 research and development (R&D) cooperative collaboration on 91–92, 101 innovation markets and 37–38, 159–160 responsive regulation theory 291 reverse payments see pharmaceutical industry Reysen, Marc 300 Rhone Poulenc 381 Rill, James F. 54 Rio Tinto 371, 439 Rockstar group 167 Rome Treaty, Article 85 365 Roosevelt, Franklin 11 RTE & ITP v Commission (Magill) 148 rule-of-reason see per se/rule of reason categorisation Russia competition law enforcement 296 criminalization of cartel conduct 305, 312, 314

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 15 /

Date: 14/10

JOBNAME: Duns PAGE: 16 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

516 Comparative competition law PMN (pre-merger notification) regime 192, 198 safe harbors, for competitor collaboration (EU) 101–102 safety zones, for competitor collaboration (US) 100–101 Safeway 140 Salop, Steven C. 125, 126, 127–129, 131–132, 133–134 Sammelrevers case 263–264 Samsung 165, 205, 466 Samsung Electronics Co. 379–380 Sanyo- Marunaka case 424 Schering-Plough Corp. v FTC 152 Schumpeter, Joseph A. 160, 394 Schwinn, US v General Motors Corp 231 Seagate 205 search and seizure powers see public enforcement Set of Multilaterally Agreed Principles and Rules for the Control of Restrictive Business Practices (UNCTAD) 11–12 sham litigation 147, 151 Shang Ming 193, 205 shareholders 396–399 Shell Oil Co. 88–89 Sherwin, R.A. 48 Simons, Joseph J. 48 Sina.com 464 Singapore competition law, overview 17 non-price vertical restraints (NPVRs) 243 single economic entity doctrine (EU) 367–368 single enterprise doctrine (US) 86–87, 88–89 single monopoly profit 119 Sino-Agri 466–467 Sino-Chem 466–467 Sinopec 466 Slaughter, Anne-Marie 349 smartphone IP standard-setting 164–168 Snyder, Edward 406

Columns Design XML Ltd

/

software platforms 51 Sotomayor, Sonia 73 South Africa Competition Act 387 criminalization of cartel conduct 305, 315 enforcement agency model 272, 274–275 private antitrust enforcement 387–389 South African Airways 387 South America, competition law overview 472–476 see also individual countries South Korea criminalization of cartel conduct 312, 313, 318, 327 extraterritorial reach of 373–374 Korea Fair Trade Commission (KFTC) 373–374 Monopoly Regulation and Fair Trade Act (MRFTA) 374, 388 private antitrust enforcement 388 Spahr v Leegin Creative Leather Products 256 Spain multiagency enforcement 281 Spratling, Gary 336 SSNIP (small, sustained, not insignificant increase in price from the competitive level) 187 critical loss analysis and 48–49 defined 34 evidentiary basis for market definition and 44–46 natural experiments and 47–48 test/hypothetical monopolist test 34–39 two-sided platforms/markets and 50–52 Stallibrass, D. 261–262, 267–268 standard essential patents (SEPs) 164–167 Standard Oil Co of California v US 231, 232 Standard Sanitary Manufacturing Co. v United States 143 standard-setting in IP 161–164

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 16 /

Date: 13/10

JOBNAME: Duns PAGE: 17 SESS: 4 OUTPUT: Wed Oct 21 12:16:51 2015

Index 517 standard-setting organizations (SSOs) 162–163 Staples 54 Staples case 47–48 State Oil v Khan 254 Stevens, Robert 122 Stigler, G.J. 48 Stigler school see also classical market power Stigler-Sherwin test 48 Stiroh, Lauren Johnston 123–124, 133 strategic market power 113, 125–135 comparison to classical approach 135–137 exercise of 126–131 identification of 131–135 incentives and effects 125–126 Structural Impediments Initiative (SII) 416 substantial lessening of competition test (SLC test) 181–182 Sunshine, Steven C. 38 supply chain product markets and 42–44 supply/demand elasticity 120 supply-side substitutionability debate commerciality and 50 defined 30–32 supra-competitive prices 72–77 Swaine, Edward T. 299 Switzerland criminalization of cartel conduct 313 Sylvania case see Continental TV, Inc v GTE Sylvania T Mobile Netherlands BV v Raad van Bestuur van de Nederlandse Mededingingsautoritleil 107 Tamoxifen Citrate Antitrust Litigation, In re 152 Tampa Electric Co v Nashville Coal Co 231–232 Tatham, T. 252 TDLC (Tribunal de Defensa de la Libre Competencia) 481, 483–484, 486, 488, 493–495, 497 Teece, David J. 37 Telser, L.G. 250, 252–253

Columns Design XML Ltd

/

Tennessee Tennessee Trade Practice Act 256 Terminal Rail Road case see United States v Terminal Rail Road Association territorial restraints 223, 231, 236 Texaco, Inc. 88–89 Texaco, Inc. v Dagher 88–89 Thailand competition law, overview 17 criminalization of cartel conduct 312 Theatre Enterprises v Paramount Film Distributing Corp. 65–66, 70–71 theories of harm, in merger law 184–185 third-line forcing 222 Timberlane Lumber Co. v Bank of America 356–357, 359 Tirole, Jean 132–133 Todd v Exxon Corp. 73 Tokyo/Osaka stock exchange merger 427 Townshend v Rockwell International Corporation 147 Toyota 452 Toys’R’Us case 424 transaction cost economics 91–92 transfers of market power 109–112, 114 transport services 41–42 Treaty on the Functioning of the European Union (TFEU) 27 see also European Union (EU) Trebilcock, Michael J. 271, 276, 279 TriCor 157 Tsuchiya kigo- v Japan Fair Trade Commission 419–420 two-sided platforms/markets 50–52 tying arrangements 119, 143–144, 145, 146, 147, 222, 228–229, 232–233, 242 UK Supreme Court 336 undertakings 87, 206 Union Oil Company of California, In re 163 United Kingdom (UK)

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 17 /

Date: 21/10

JOBNAME: Duns PAGE: 18 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

518 Comparative competition law Competition and Markets Authority 281 Competition Commission/Office of Fair Trading 29–30, 53 criminalization of cartel conduct 307, 312, 324–325 Enterprise Act (2002) 324 National Health Service 159 Office of Fair Trading (OFT) 159, 291, 297–298 private enforcement 369–371 prosecutorial enforcement model 274–275, 292 removal of RPM on books 268 search and seizure powers 290 UK Competition Act (1998) 84 UK Enterprise Act (2002) 84 United Nations Conference on Trade and Development (UNCTAD) 213–214, 301, 335, 348 Intergovernmental Group of Experts on Competition Law and Policy 13 Model Law on Competition 13 Set of Principles 11–13, 20 United Nations Economic and Social Council (ECOSOC) 11 United States of America (US) administrative enforcement model 272 Antitrust Criminal Penalty Enhancement and Reform Act (2004) 307–308 Antitrust Enforcement Guidelines for International Operations 355–357 Antitrust Guidelines for the Licensing of Intellectual Property (US) 145–146 Antitrust Modernization Commission 47 approval of Google/Motorola merger 206 approval of Seagate/Samsung merger 205 Bankruptcy Act (1898) 186 categories of agreements 92–93, 99

Columns Design XML Ltd

/

categories of anti-competitive agreements 94–96 Clayton Act (1914) 229–234, 280, 422 Consumer Goods Pricing Act (1975) 250 cooperation agreements with EU 21–22 criminalization of cartel conduct 307–308, 318–320 Discount Pricing Consumer Protection Act, proposed 257 effects doctrine 351–355 exclusive dealing 231–232 extraterritorial reach of 351–364 Federal Courts Improvement Act (1982) 145 Federal Trade Commission Act (1914) 164, 231, 422–423 Federal Trade Commission (FTC) 76–77, 104–106, 149–150, 163–165, 167, 169, 255, 273, 274, 277, 280–282, 398–399, 407, 408–409, 422–423 Foreign Trade Antitrust Improvements Act (FTAIA) 353, 360–364 Guidelines for the Licensing of Intellectual Property 37–38 Hatch-Waxman Act 150 Horizontal Merger Guidelines 186, 188 imposition of structural undertakings on Western Digital/Hitachi merger 206 McGuire Act 267 Miller-Tydings Fair Trade Act 267 MLATs of 334 as model for laws in other jurisdictions 110–111 monopoly thresholds 122 multiagency enforcement 280–281 non-price vertical restraints (NPVRs) 229–234 Patent Misuse Reform Act (1988) 145 private actions and jurisdiction 359–363

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 18 /

Date: 13/10

JOBNAME: Duns PAGE: 19 SESS: 2 OUTPUT: Wed Oct 21 12:16:51 2015

Index 519 private antitrust enforcement 385–386, 390–393, 403–404, 407, 410 prohibited per se categorization 93–96 RAND (reasonable and nondiscriminatory) agreements 162, 164 re-importing of pharmaceuticals 40–41 resale price maintenance 16–17 resale price maintenance (RPM) 249–258 restrictions on acquisition 221 role of comity in jurisdictional test 355–359 search and seizure powers 290 Sentencing Guidelines 308 Sherman Act (1890) 16, 27, 40, 59–60, 68, 69, 82–83, 85, 88, 94, 112–113, 142, 143, 229–230, 231–234, 251, 307, 352, 358, 384, 420, 422–423 single enterprise doctrine 86–87, 88–89 tying and bundling 232–234 US Guidelines 85–86, 90, 100–101, 107 Webb-Pomerene Act (1918) 365 see also US Justice Department (DOJ); US Supreme Court and individual cases United States v Aluminum Co of America 127–128 United States v Arnold, Schwinn & Co 250–252 United States v Container Corp. 94 United States v Line Material Co. 144 United States v Loew’s 144 United States v Terminal Rail Road Association 127 unjust enrichment principles 401–402 Uranium Antitrust Litigation; Westinghouse Electric Corporation v Rio Algom Ltd 354 US Guidelines on characterization of agreements 107

Columns Design XML Ltd

/

on competitor collaboration 85–86, 90, 100–101 US Justice Department (DOJ) 144, 166–167 antitrust actions and share price 398–399 Antitrust Division 308, 319–320 consent judgments 293 criminalization of cartel conduct 308, 319–320 enforcement agency model 275, 280–281, 289, 293 extradition of Ian Norris 335–337 fines on LCD cartel 56 Guidelines for the Licensing of Intellectual Property 37–38 Horizontal Merger Guidelines 52–53, 55 predatory pricing suits 407 US National Football League 89 US Supreme Court on cartel horizontal agreements 63–66 on price fixing 60–61 on private enforcement 385, 389, 410–411 see also individual cases US v Aluminium Co. of America 352 US v Colgate & Co 254 US v Microsoft 233–234 US v Nippon Paper Industries Co Ltd 358 US. v The Gillette Company 54–55 van den Bergh, Roger 90 VBVB and VBBB v Commission 262–263 Verizon Communications v Trinko 147 vertical conduct see non-price vertical restraints (NPVRs); resale price maintenance (RPM) vertical integration and supply chain 43 vertical price fixing see resale price maintenance (RPM) videotaping of cartel meetings 61–62 Vietnam competition law, overview 17

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 19 /

Date: 13/10

JOBNAME: Duns PAGE: 20 SESS: 3 OUTPUT: Wed Oct 21 12:16:51 2015

520 Comparative competition law vitamins cartel 326, 327, 374, 381–382, 391 Vitamins Cases 320–322 Walgreen Co. v AstraZeneca Pharmaceuticals LP 157–158 Werden, Gregory 28 Western Digital 205–206 Whish, R. 21–22 Whitaker, Mark 320 Whole Foods 54 Wild Oats 54 Willig, Robert 47 Wood, William 406 Wood Pulp case 364–366, 368

Columns Design XML Ltd

/

World Bank 295, 348 World Trade Organization (WTO) 295 lack of competition law forum in 213 multilateral competition agreement proposals 12–13 scope of 14, 15, 20 Xerox case see Independent Service Organizations Antitrust Litigation (Xerox), In re zaibatsu (large family-owned conglomerates) 417, 428 Zambia criminalization of cartel conduct 312

Job: Duns-Comparative_competition_law

/

Division: Index

/Pg. Position: 20 /

Date: 14/10