This volume of essays contains contributions by a group of specialists in the area of competition law, including heads o
292 55 33MB
English Pages [632] Year 1998
INTRODUCTION
This volume is composed of the collected written works and edited transcripts of dialogue which occurred during the Second Workshop on European Competition Law, held at the Robert Schuman Centre of the European University Institute in Florence in June 1997. Following the successful example of the First Workshop on European Competition Law, held in April 1996, the workshop's participants were again a group of approximately thirty-five high level officials from the world's major competition enforcement authorities and international organisations, renowned scholars and private practitioners specialising in competition. Participants came mainly from the European Union and its Member States, but also from Canada, Japan, Mexico, Poland and the United States. During one and a half days, the workshop, organized in four panels, discussed different aspects of the fundamental question of what objectives are or should be pursued by competition policy. The choice of the overall theme of the Second European Competition Law Workshop was a direct consequence of its 1996 predecessor. Readers of the Robert Schuman Centre Annual on European Competition Law 1996 will remember that the First Workshop on European Competition Law was devoted to the problems of enforcing competition law in a 'federal' context. During the debate in April 1996 it became clear that some of the substantive and procedural/institutional problems with which the European Union has been struggling for a couple of years might be linked to a hidden agenda, i.e. implicit divergences about the objectives of the European Union's competition policy. Like the final objectives pursued by Member States of the European Union through European integration, the objectives of EU competition policy are rarely discussed, as discussion might reveal major differences of opinion and prove to be divisive. However, a better understanding of these objectives is useful—if not indispensable—both for substantive and procedural reforms, in particular a new approach to vertical restraints of competition, and for the institutional debate, i.e. the responsibilities of Member States' competition authorities and the controversy concerning the establishment of an independent European Competition Authority.
viii
Introduction
In addition to these problems internal to the European Union, a discussion about objectives of competition policy seemed appropriate in view of the efforts, undertaken in the framework of the World Trade Organisation, to determine the prospects for a multilateral competition code. A basic agreement on objectives could be as helpful as clear understanding of the differences of goals and perceptions of the major participants in the discussions under way in Geneva. These policy oriented considerations determined the sub-divisions of the overall theme which was examined in four successive panels. Thefirstdiscussed competition policy objectives in general. The second looked at competition policy objectives in the context of a multilateral competition code. The third tried to determine the objectives which are actually and in reality pursued in the European Union, Canada and the United States. The fourth was exclusively devoted to future reforms of the E.U.'s competition policy, in particular a new approach to Art. 85 of the E.C. Treaty. The reader of this volume who is interested in the international dimension (the prospects for a multilateral competition code) will therefore concentrate on the panel discussions and working papers of Panel One (Chapters 1 and 2) and Panel Two (Chapters 3 and 4). However, they will certainly find it useful to look also at the proceedings of Panel Three (Chapters 5 and 6) which explain more clearly than Part One the real situation in the European Union, Canada and the United States. Readers who are particularly interested in the future orientation of EU competition policy will focus on the discussions and working papers of Panels One, Three and Four (Chapters 1, 2, 5, 6, 7 and 8). In view of the extraordinary richness of the material and in order to get a quick overview, it might be useful to start reading the comprehensive introductions to the collection of working papers written for each of the panels by one of the (four) General Rapporteurs. This reading might be followed by a study of the working papers. The transcripts of the discussions will reveal their high added value against the background of the working papers which were known to participants before the workshop started. We cannot and it is not our ambition to summarise in a few pages the multitude of insights which the reader will collect from the study of this volume. In view of the division of the overall theme into four panels and the two main policy orientations of the workshop (the international debate and internal EU reform) it might however be useful to indicate briefly the main results of the discussions.
Introduction
ix
1. Objectives of Competition Policy in General With respect to the determination of objectives of competition policy, it was considered to be useful to distinguish between ultimate goals and intermediate (operational or direct) objectives of competition policy (a similar distinction would be the one between benefits and objectives). Ultimate goals reach beyond the intermediate objectives of competition policy. While the ultimate economic goal of competition policy is the wholesome development of the economy of a country or a group of countries, ultimate political goals are the underpinning and furthering of the democratic process and its essential elements, like pluralism, free enterprise, individual freedom etc. Compared with ultimate goals, intermediate (operational or direct) objectives of competition policy are much more confined. Participants agreed that these objectives are—or should be—limited to economic efficiency and consumer benefit. Other objectives (like industrial policy, the protection of small and medium sized enterprises, employment, the fight against inflation, the improvement of the environment etc.) should not be pursued by competition policy, but by other instruments which will normally produce better results. Ultimate goals will be furthered by competition policy, but positive economic and political effects will rather be by-products than directly intended results. Objectives are rarely defined expressly in competition statutes (Canada and Japan provide examples to the contrary). They have to be inferred from legislative provisions which are broadly worded. This allows adaptation of the law to changing circumstances, be it those prevailing in the markets, or new insights in economic and legal thinking. Competition authorities might in this respect act more freely than courts which feel obliged to interpret the statutes primarily in the light of their wording. Discussions (particularly in Panels One and Three) showed that during recent years most jurisdictions have seen a remarkable trend toward a significant narrowing of competition law objectives in the direction of a more economic based analysis, according ever greater weight to economic efficiency and consumer benefit. This trend is based on the growing influence of economists and economic thinking on competition authorities and courts. Most participants (notably those from Canada and the U.S.) stressed the benefits of economic tools for the quality and predictability of decisions in competition matters. Economic theory was considered to be a common language which furthers transparency, and which facilitates understanding and critical appraisal. Economic analysis was therefore considered to be a positive contribution to legal certainty. Though recognizing the usefulness of (and need for) a thorough examination of the facts in the light of the most reliable
Introduction
economic insights, sceptical voices observed that economic theory is not a natural science, that economics is a function of politics (leading to 'left-wing', 'centre' and 'right-wing' economic theories), that economists therefore often arrive at different results, if only in giving different weight to certain facts. However, it was recognized that economic theory is a strict intellectual discipline which contributes, if combined with transparency, to greater accountability. Discussions in Panel One, but even more so in Panel Three showed that the situation in the European Union is characterized by four particularities. The first concerns market integration. Until recently, market integration was the dominant operational objective of E.U. competition policy. Progress on the road towards the effective establishment of the Single Market will contribute to diminishing the relative importance of this objective in the future. However, it is unlikely that it will be abandoned and replaced by the generally accepted objective of economic efficiency and consumer benefit. It was noted that market integration is a purely economic objective in a long term perspective. However, in the short and medium term, it is also—and in situations of conflict between the objectives of market integration and of economic efficiency even primarily—a political objective. The second particularity concerns the long list of other objectives identified in the Report on the E.U.'s competition policy for Panel Three. The E.U. Report on objectives prepared for Panel One and the discussions in these two panels seem however to indicate that E.U. competition policy has evolved and continues to evolve in the same direction as competition policy in other jurisdictions. Not only does onefindthe basic (and during the entire workshop generally accepted) distinction between fundamental goals and intermediate objectives in the E.U. Report for Panel One. This report concludes also inter alia 'that competition policy and enforcement are not appropriate instruments to achieve ultimate objectives, such as prosperity and job creation directly. On the contrary, social objectives may conflict with the maintenance of competition in the short run. Therefore, the maintenance of effective competition must be the direct objective of enforcement'. Whether this (policy) statement is followed in reality can easily be checked if decisions are fully reasoned and published (as under the E.U. Merger Regulation). The situation is much less verifiable under Art. 85, because of the rarity of formal exemption decisions and the brevity of reasoning in of informal comfort letters. The third particularity is the relatively late arrival of economic analysis in the application of E.U. competition law, in particular Art. 85. The delay is closely linked to the historically dominant E.U. objective of market integration and the related objective to protect the freedom of action of market players. The
Introduction
latter is probably a consequence of the strong influence of the 'Freiburger Schule' on German competition theory and practice, exported by the first E.U. competition law enforcers to Brussels. In spite of earlier invitations by the E.U. Court of Justice, in-depth economic analysis started only with the implementation of the Merger Regulation since 1990. From there it has spread into the traditional areas of E.U. competition law. However, in these areas the process of alignment with practice in other jurisdictions is far from being finished. Attention was drawn in this respect to the remarkable achievements of the Italian Competition Authority which was only established in 1990. The fourth particularity is the well known uniqueness of the E.U. decisionmaking process in competition matters. All individual decisions for the implementation of E.U. competition law are taken by the E.U. Commission and not by an independent agency or by courts. In this respect, the E.U. continues to resist a world wide historical trend which clearly goes in the direction of conferring the responsibility for the day to day application of competition rules to bodies more or less detached from government. Discussions in all four panels clearly showed the relationship between the objectives of a given competition policy and the institutional arrangements for its implementation. The broader the objectives, the greater is the need for a centralized political body to pursue them. On the contrary, narrower objectives will facilitate the transfer of responsibilities from the central government to an independent agency and/or courts. In other words, narrower objectives will permit deconcentration and decentralisation. The special situation of developing countries (or countries in transition) was examined both in Panel One and Two. It becomes particularly clear through the proceedings of Panel One, for which the General Rapporteur summarised the pertinent working papers and his own experience (among others as Chairman of the ongoing W.T.O. Working Party on competition problems) as follows: 'In summary, in developed countries, the immediate goal of competition policy is to promote an efficient allocation of resources in the traditional economic sense, and therefore such policy focuses on market behaviour of firms and control of mergers. In contrast, in developing countries or countries in transition (i.e. countries in which the preconditions for a market economy are not fully established), the immediate goal of competition policy is much broader, because: there is less consensus among public policy officials or politicians about the desirability of competition policy; the economic, legal, social or political structures are less appropriate for the development of a free market economy; and the public places greater weight on the short term disruptions from a market economy than on its long term benefit. As a result, the immediate goal of competition policy in these countries appears to be the emergence
Introduction
of economic opportunities and entrepreneurship in a context in which more attention must be paid to establishing the political acceptance of a market economy (and therefore taking into account the populist goals of competition policy) .. . Competition policy in a developing country often takes a more regulatory approach, thereby allowing it to contribute actively to transformation of economic structures and behaviours ... Moreover, the advocacy function of the competition authorities vis-a-vis the rest of the government, and the propaganda function vis-a-vis the public, are seen as crucial.' 2. Competition Policy Objectives in the Context of a Multilateral Competition Code The task of Panel Two was to examine competition policy objectives in the context of a multilateral competition code. One of the astonishing results was the conclusion of the General Rapporteur for Panel Two that 'all speakers agreed that differences in competition law are not as such, an impediment to internationalisation.' This contrasted sharply with the statement of the General Rapporteur of Panel One that 'whatever the solution at the multilateral level, it requires agreement among the trading nations regarding the legitimacy of competition policy principles.' The explanation of this apparent paradox may be found in the experience of a seasoned international negotiator, that actors in international negotiations may be able to agree on rules, even if they each have different reasons for accepting common rules. On the other hand, consensus on competition policy objectives (as between the European Union and the United States) does not guarantee that this consensus leads to the same position with respect to the desirability of a multilateral competition code, as demonstrated by representatives of both sides during the proceedings of Panel Two. Because of the agreement that differences in competition law objectives are not an impediment to internationalisation, participants in Panel Two examined more the fundamental question of the necessity or desirability of a multilateral competition code and its possible content. Here, too, it is useful to repeat the conclusions drawn by the General Rapporteur for Panel Two according to whom 'while there was some divergence among panel members, there was again remarkable agreement, and near consensus that: —nations should adopt competition laws and be obliged to enforce them against anticompetitive restraints of international trade; —the GATTAVTO principles of transparency, national treatment and non-discrimination, and against beggar-thy-neighbour restraints, should be included in competition law obligations;
Introduction
xiii
—nations should continue to develop bilateral cooperation, including cooperation in discovery and other comity measures; —if work begins on substantive rules against certain private or hybrid restraints, one should also cautiously build the foundation, possibly focusing on effects resulting from conflicting national regimes, core principles on which there is already international consensus, and the private equivalents of G.A.T.T.-illegal government restraints (e.g. quantitative restrictions on imports and exports and strategic nationalistic behaviour).' However, not all participants agreed that a multilateral competition code is necessary or desirable. As it is well known, the United States is in favour of strengthened bilateral cooperation between competition authorities. It fears that multilateral negotiations might undermine bilateral efforts; agreement will be hard to reach and might result in the lowest common denominator; W.T.O. dispute settlement would be inappropriate or ineffective, as competition law enforcement is highly fact intensive and involves confidential information. Other participants challenged the basic assumptions of the 'internationalists,' They asked whether there are any significant inefficiencies resulting from negative spillovers (acts or conduct that decrease aggregate world wealth) that cannot be addressed in the existing W.T.O. framework. They did not find a sufficient economic rationale for a multilateral competition code, considering the large extent to which current international trade law could be (but has not yet been) used to solve existing problems. In their view, no international competition agreement should be pursued unless it can be shown to increase aggregate world wealth. For further observations, we would like to refer the reader to the proceedings of Panel Two which benefitted from an extraordinarily rich collection of detailed working papers. We now want to turn our attention to the second policy oriented subject of inquiry of the 1997 Workshop, i.e. the future reforms of the EU's competition policy, in particular a new approach to Art. 85.
3. Objectives of Competition Policy in the Context of future reforms of the E.U.'s Competition Policy A. Substantive Competition Law All participants in the 1997 Workshop agreed that the E.U.'s competition policy under Art. 85 has to be adapted. The discussion concentrated on this article, while Art. 86 was mentioned only rarely. However, those who referred to Art. 86 were rather critical of certain aspects of the Commission's decision
xiv
Introduction
making practice. Conversely the implementation of the Merger Regulation received very good reactions; it was generally praised as a positive development and major contribution to a modernized European competition policy. Participants agreed also on the reasons why practice under Art. 85 has to change. These reasons are a combination of several factors. Academics and practitioners expressed their—well known—dissatisfaction with the results of the actual policy, which spends too much of the Commission's rare resources on the control of vertical restraints, instead of concentrating on much more dangerous horizontal restrictions. In addition, this policy does not produce enough formal decisions and therefore lacks transparency, independently of the problems of legal security resulting from the use of non-binding comfort letters. Representatives of DG IV stressed the need to adapt traditional competition policy to changing circumstances, the most important being progressive effective integration of markets within the E.U. and globalisation of markets at the world level, together with the perspective of further enlargement of the E.U. which will put even more strains on the already stretched resources of DG IV. Enlargement will reinforce the now generally accepted need for decentralized implementation of E.U. competition law by both administrative authorities and courts of the E.U. Member States. Discussions showed further a basic consensus on the main reason for the existing inefficiency of E.U. competition policy under Art. 85, i.e. the overly broad interpretation of its paragraph 1, more specifically the too wide interpretation of the notion of restriction of competition. We have already mentioned the origins (or contributing factors) for this development: the historically dominant objective of market integration and the related objective of protecting the freedom of action of market players. Critical observers outside the Commission mentioned in addition 'lazy thinking' facilitated by the existence of paragraph 3 of Art. 85 and perhaps even self-interest of an administration which might be tempted to increase its exclusive powers to apply this provision. Consensus reached even further. Participants agreed on the remedy, i.e. on the necessity to reduce the scope of Art. 85 (1). This reduction should be operated through a narrowing of the notion of restriction of competition and not through a stricter interpretation of the requirement that trade between Member States has to be affected. The latter would have the undesirable effect of a retreat of E.U. competition law followed by a corresponding extension of the scope of application of non-harmonized and therefore diverging national competition laws. The narrowing of the notion of restriction of competition and the consequential reduction of the reach of Art. 85 (1) would automatically reduce the need for exemption decisions of the Commission under Art. 85 (3). It would therefore lead to greater legal security, reducing at the same time the
Introduction
xv
costs for industry which would no longer need to apply for formal exemption decisions or at least for comfort letters. This operation would also enlarge the possibilities of decentralized implementation of E.U. competition law by administrative authorities and courts of E.U. Member States. Until now, this decentralized implementation is limited by the broad scope of Art. 85 (3) which can only be applied by the Commission. Participants agreed that the narrowing of the notion of restriction of competition is not affected by disagreements on the (intermediate, direct, operational) objectives of E.U. competition policy. In other words, the road to re-interpretation of Art. 85 (1) is not obstructed by divergences of opinion on the objectives which are pursued under this provision. This seems to us to be the clear result of the discussions in Panels One, Three and Four and one of the major, if not the most important insights produced by the 1997 Workshop. The workshop did not have the ambition to examine in depth the precise lirriits of the notion of restriction of competition and the exact borderline between paragraph 1 and paragraph 3 of Art. 85. Discussions in Panel Four indicated, however, the fundamental options. Everybody agreed that the so-called 'bifurcation' in Art. 85 (i.e. the existence of these two paragraphs) does not hinder the re-interpretation of paragraph 1. But participants took different views as to how much of the appreciation which is traditionally carried out under paragraph 3 could be transferred to paragraph 1, i e. into the examination of the notion of restriction of competition. For some, the examination under the notion of restriction of competition would limit itself to some of the efficiency oriented evaluations of the agreement or concerted practice. The rest would have to be evaluated under paragraph 3. For others, all efficiency related considerations should take place under paragraph 1. The balancing under paragraph 3 would therefore concern only considerations other than efficiency oriented evaluations. It is obvious that the strength and force of conviction of the second group depend on the separate—but closely linked—question whether and to what extent paragraph 3 takes into account non-efficiency related considerations of an economic or even purely political nature. This fundamental question was even less discussed by the workshop than the borderline between the two paragraphs of Art. 85. It is however of fundamental importance for the correct understanding and application of both paragraphs of Art. 85. Without engaging in a detailed analysis of the nature of the elements, which Art. 85 (3) allows to be considered, participants referred to two important factors of interpretation. Partisans of a narrow construction of Art. 85 (3) drew the attention to the wording of paragraph 3 which seems to exclude purely political considerations. Others recalled that Treaty provisions have to be
xvi
Introduction
interpreted in the context of other Treaty rules. Some articles even demand explicitly that certain policy considerations have to be taken into account in the formulation of other Community policies (e.g. Art. 130 r which stipulates that 'environmental protection requirements must be integrated into the definition and implementation of other Community policies'). Interpreting Art. 85 (3) in a way which insulates competition policy from all other Community policies therefore seems to be ruled out. Participants discussed in depth the question whether the approach to Art. 85 should be achieved through re-interpretation or through Treaty amendment. Treaty amendment was advocated strongly by one participant. One of his main arguments was the desire to hierarchize Community policy objectives, to rank competition policy (together with the four fundament internal market freedoms) at the top, and to isolate it from other Community policies. All other participants expressed a preference for reform through re-interpretation. According to the majority view, the Court of Justice would not stand in the way of reform, provided the Commission establishes its new approach by determined and courageous action. After all, the Court had told the Commission already on several occasions that Art. 85 (1) is to be interpreted in a more realistic, complete and economically determined way than traditionally in the past.
B. Institutional Issues We have already mentioned the considerable benefits which decentralisation of the E.U.'s competition policy will obtain from the new approach to Art. 85 (1). It should be noted that these benefits were considered so substantial that no one argued in favour of sharing the Commission's monopoly to grant exemptions under Art. 85 (3) with national competition authorities or courts. Implicitly, participants considered (first) that the monopoly—substantially reduced in scope by the reform—would no longer be a serious obstacle on the road to efficient decentralisation, and (second) that the remaining tasks under Art. 85 (3) required centralized decision making. The pro's and con's of an independent European Cartel Office were however discussed, though only marginally. Readers of the preceding volume of this series will remember that the 1996 Workshop was profoundly divided on this issue. It is somewhat surprising that the 1997 Workshop showed greater support for the establishment of such an office than the 1996 Workshop. The debate about competition policy objectives and on the interpretation of Art. 85 (3) clarified however—at least implicitly—the limits of such an initiative.
Introduction
xvii
The arguments for an independent European Cartel Office are strong in so far as E.U. competition policy is determined by considerations of economic efficiency and consumer benefit, i.e. the purely competition policy oriented objectives. If the E.U.' s competition policy was strictly limited to these considerations, a case could even be made for a single step procedure, confined to the independent Office, without any sort of appeal to the Commission. This would be like the procedure before the independent Italian Competition Authority which—unlike the procedure in Germany—does not allow an appeal to the Government. The case for a single step procedure becomes weaker in so far as broader economic policy considerations are allowed under Art. 85 (3). As we have indicated above, it is likely that paragraph 3, interpreted in the context of the Treaty as a whole, is open to such considerations. An independent European Cartel Office could certainly integrate them into its investigations and decisions. However, it is difficult to deny that precisely these type of considerations plead for the possibility of intervention of a political body, like the Commission. The same would obviously be true if Art. 85 (3) permitted the taking into account of purely political arguments. However, we do not believe that this is the case. An interesting variant of the arguments for an independent European Cartel Office was the request for the establishment of an independent 'toothless watchdog' in competition matters, following examples in certain Member States. One of us has already earlier made a similar suggestion, referring to the example of the German 'Monopolkommission', which publishes at least once a year a report in which German competition policy is critically reviewed and reform proposals are made. It would indeed be highly useful to set up formally an independent group of eminent persons which would critically analyze every year the competition policy pursued by the Commission. The European Parliament and public opinion at large might find such expert analyses useful. They would probably contribute to a further strengthening of the competition culture in the E.U.
4. Conclusions The Annual Workshops on European Competition Law, held at the Robert Schuman Centre of the European University Institute, are academic in nature, but they also pursue policy objectives. After the second of these Workshops, we are tempted to ask ourselves whether one can detect any influence of the
xviii
Introduction
discussions in Florence on actual policy making. We believe it is no exaggeration to say that we detect such influences both with respect to the issue of decentralisation, discussed in 1996, and to a new approach to the interpretation (or application) of Art. 85 (1). The latter might become more visible once the Commission has concluded its reflections on the future treatment of vertical constraints. When this book is published, the results of these reflections will probably also be public. In the light of the lack of consensus among participants in both workshops, it is not astonishing to find that the debate about the independent European Cartel Office has not progressed. On the contrary, this debate seems today to be dead. Not only has Germany not found any allies for its proposal to establish such an office during the last Intergovernmental Conference. It seems also that Germany has discredited its own initiative through its practical behaviour during recent months. However, the impression of (premature) death might be wrong. The actual scene is dominated by the extraordinary achievements of Karel van Miert, responsible in the Commission for the E.U.' s competition policy since 1993. He has clearly demonstrated to what extent the Commission can and does pursue a vigorous, economic efficiency oriented competition policy, in spite of its political nature and the shameless lobbying on the part of some of the Member States. Karel van Miert's predecessors were similarly determined and courageous, though not always equally successful. The E.U. has therefore enjoyed some 15 years of progressively strong competition policy leadership from Brussels. However, there is no institutional guarantee that this situation will remain unchanged. We therefore believe that the idea of the independent European Cartel Office is not dead, but only profoundly asleep. Let us hope that this sleep will last through to the new millennium.
Winter 199711998
Claus Dieter Ehlermann Laraine L. Laudati Florence, Italy
BIOGRAPHICAL NOTES ON THE PARTICIPANTS
RODERICK ABOTT
was appointed Ambassador and Head of Delegation of the European Commission in Geneva in November 1996. In the 1960s, he began his career in the then Board of Trade in London, moving later to the Department of Economic Affairs. He joined the European Commission in 1973, and was appointed Head of Division responsible for G.A.T.T. Affairs. He moved to Geneva as Chef Adjoint in 1975, and on conclusion of the Tokyo Round returned to Brussels in 1979. From 1982-1996, he was a Director in DG I with responsibilities for O.E.C.D. and G.A.T.T., including preparation of the Ministerial Meetings in Montreal, Brussels and Marrakech, and in charge of third country tariff negotiations (e.g. after E.C. enlargements). GIULIANO AMATO
is Professor of Law at the European University Institute in Florence. From 1993-1997, he was Chairman of the Italian Antitrust Authority; from 19921993, he was Prime Minister of Italy; from 1987-1988, he was Deputy Prime Minister; from 1987-1989, he was Minister of the Treasury, and from 1983-1987, he was Under-secretary to the Prime Minister's Office. He became a Full Professor of Comparative Constitutional Law at the University of Rome, School of Political Science in 1975. Prof. Amato has also been full time professor at the Universities of Modena, Perugia, and Florence. He has written books and articles on economy and political institutions, personal liberties, federalism, and comparative government. His current area of research is competition law in Europe on a comparative basis. MiLOS BARUTCISKI
is a Partner at the firm of Davies, Ward & Beck in Toronto, Canada. He was Special Advisor to the Director of Investigation and Research of the Canadian Competition Bureau from 1991-1993. Mr. Barutciski was an Adjunct Professor at the Faculty of Law of the University of Ottawa from 1990-1994, where he taught international trade law.
Biographical Notes on the Participants
JACQUES H.J. BOURGEOIS
is a Member of the Brussels Bar and Professor at the College of Europe in Bruges. In 1962, he joined the European Commission, and its Legal Service in 1965. He was Head of the European Commission's Trade Policy Instruments Division from 1983-1987 and Principal Legal Adviser to the European Commission Legal Service from 1987-1991. He returned to private practice in 1991. He has chaired a G.A.T.T. and a W.T.O. dispute settlement panel. He lectured at the University of Brussels, the University of Michigan Law School and the Autonomous University of Barcelona. He has been Jean Monnet Professor at the University of Bonn Law School. He has published on E.C. competition and trade law and on G.A.T.T./W.T.O. issues. GABRIEL CASTANEDA
is a Partner at the firm of Castaneda y Asociados. He served as the Executive Secretary of the Federal Competition Commission from the agency's inception until 1995. Prior to that, he held the position of Legal Director General of the Economic Deregulation Unit at S.E.C.O.F.I. Mr. Castaneda is Secretary of the Antitrust and Trade Law Section of the International Bar Association and is currently the Chairman of the Antitrust Commission of the Mexican Chapter of the International Chamber of Commerce. GEOFF EDWARDS
is an Economic Analyst with the Australian Competition and Consumer Commission's Telecommunications branch in Melbourne. He previously worked with the Commission in its Mergers branch and Executive area. He has published articles on various topics of competition law. His main areas of interest include competition law and economics, and infrastructure access issues, particularly in the telecommunications industry. CLAUS DIETER EHLERMANN
is Professor of Economic Law at the European University Institute in Florence, and a Member of the Appellate Body of the W.T.O. in Geneva. In 1961, he joined the Legal Service of the European Commission, where he was Director General from 1977-1987. He served as Spokesman of the Commission from 1987-1990, and then as Director General for Competition from 1990-1995. He has lectured at the College of Europe in Bruges, at the Free University of Brussels and at the University of Hamburg, where he became Honorary Professor in 1983. His publications focus on European Community law and in particular on institutional and competition law issues.
Biographical Notes on the Participants
xxi
PIERO FATTORI
is Director of the Legal Office of the Italian Antitrust Authority. Before joining the Authority in 1991, he was a researcher in administrative law at Luiss University in Rome, and he was Assistant to the President of Consob, the agency in charge of stock market supervision andfinancialservices in Italy. His publications focus on antitrust issues. JONATHAN FAULL
is Director of Directorate A (Competition Policy, Coordination, International Affairs and Relations with the Other Institutions) of the Directorate General for Competition (DG IV) at the European Commission in Brussels. He is Professor of Law at the Vrije Universiteit Brussel and Visiting Fellow at the Centre for European Legal Studies, University of Cambridge. ALLAN FELS
has been Chairman of the Australian Competition and Consumer Commission since November 1995. He has a five year appointment. Professor Fels was Chairman of the former Trade Practices Commission from 1991-1995. He was also Chairman of the Prices Surveillance Authority from 1989-1992. He was Director of the Graduate School of Management, Monash University from 1985-1990, and is now Honorary Professor in the Faculty of Business and Economics at Monash University. Professor Fels is the Co-Chairman of the Joint Group on Trade and Competition at the OECD. ANNA FORNALCZYK.
is Professor of Economics at the Lodz Univerity Department of Economics and Sociology in Lodz (Poland), and a Member of the Committee of Management Sciences of the Polish Academy of Science. From 1990-1995, she was President of the Polish Anti-monopoly Office. In 1991, she took part in negotiations on the Europe Agreement between Poland and the European Union, with particular responsibility for competition and state aids. She has lectured at Central European University in Budapest. Her publications focus on competition concerns in contexts of restructuring, privatisation and deregulation. IAN S. FORRESTER
practices European law in Brussels with White & Case/Forrester Norall & Sutton. He also practices as an advocate in Scotland and, as a barrister, is a member of 2 Hare Court in London. He has participated in many cases before the European Courts, the European Commission, Scottish and English courts.
Biographical Notes on the Participants
He specialises in the fields of competition, trade, professional sport, broadcasting and pharmaceutical regulation. He has lectured on E.C. trade, competition, legal and political topics to academic and private bodies in many countries, and has published extensively on these themes. He was appointed Queen's Counsel in 1988 and Visiting Professor in European Law at Glasgow University in 1991. ELEANOR M. FOX
is the Walter J. Derenberg Professor of Trade Regulation at New York University School of Law. She is a member of the International Competition Policy Advisory Committee to the Attorney General and the Assistant Attorney General for Antitrust. She is a member of the Board of Directors and the Executive Committee of the Lawyers' Committee for Civil Rights Under Law. She was a partner and is counsel at the New York law firm Simpson Thacher & Bartlett. Her books include a casebook on the law of the European Union, an antitrust casebook, and a treatise on mergers (all co-authored). She has written numerous articles and essays in the area of competition. CALVIN S. GOLDMAN
is a Partner at the firm of Davies, Ward & Beck, Toronto, Canada, where he heads the competition law and trade practices group. After practising as defence counsel in competition law matters and as Assistant Special Counsel to the Attorney General of Canada in the Uranium cartel proceedings, in 1985 he was appointed Director of Investigation and Research and Head of the Competition Bureau in the Canadian Government. Prior to his return to private practice in 1990, he was appointed Queen's Counsel and was appointed to the Soloway Chair of Business and Trade Law at the University of Ottawa. Mr. Goldman is Chair of the Competition Policy Committee of the Canadian Council for International Business and is a past Chair (1995-1996) of the National Competition Law Section of the Canadian Bar Association. Mr. Goldman has published extensively and spoken widely in Canada, the U.S.A., Europe and elsewhere. BARRY HAWK
is a Partner at the firm of Skadden, Arps, Slate, Meagher & Flom LLP and Director of the Fordham Corporate Law Institute in New York. From 1968-1990, he was a Professor at Fordham Law School in New York. He has lectured at, among other institutions, Michigan Law School, New York University Law School and University of Paris. He has published numerous books and articles focused primarily on competition law and E.C. law and he
Biographical Notes on the Participants
xxiii
has edited the annual volumes of the proceedings of the Fordham Corporate Law Institute since 1974. ALBERTO HEIMLER
is Director of the Research Department of the Italian Antitrust Authority. He is Adjunct Professor of the History of Economic Regulation at Luiss University in Rome and Chairman of Working Party 2, 'Competition and Regulation,' of the Committee on Competition Law and Policy at the OECD. Before joining the Authority in 1991, he was a Senior Economist with the Confederation of Italian Industries. He has published extensively on applied economics and industrial economics. HERBERT HOVENKAMP
is a Professor of Law at the University of Iowa College of Law where his principal area of teaching and scholarship is federal antitrust law. He also teaches and occasionally writes in the fields of American legal history, law and economics, and property law. In antitrust, he has authored a standard one volume textbook (Federal Antitrust Policy: the Law of Competition and its Practice), a casebook (Antitrust Law, Policy and Procedure), and more than fifty other books and articles. He has also consulted extensively with the federal government, various state attorneys general, and many private firms. He began writing on the Antitrust Law treatise with Phillip Areeda in 1985. MARTIN HOWE
is a Special Professor in the School of Management and Finance at the University of Nottingham, and works in the competition policy and regulation fields in various advisory capacities in the U.K. and abroad. From 1973-1977, he was Senior Economic Advisor at the Monopolies and Mergers Commission. He moved to the Office of Fair Trading in 1977, where he became Director of the Competition Policy Division in 1984 and Deputy Director General in 1996. He retired from the Civil Service in 1996. ULRICH IMMENGA
is Professor of Law at the University of Gottingen, where he has been the Director of the Institute of International Economic Law since 1974. From 1986-1989, he was Member and Chairman of the German Monopolies Commission. He was also Professor of Law at Bielefeld (1970-1971) and Lausanne (1971-1974). He has been Visiting Professor at Georgetown University (U.S.A.), Kobe University (Japan), Nanking University (People's Republic of China), and several universities in France. His publications focus
xxiv
Biographical Notes on the Participants
on competition law, corporation law and European law, with reference to comparative or international aspects. FREDERIC JENNY
is Vice-chairman of the Conseil de la concurrence. Since 1997, he has chaired the W.T.O. Working Group on the Interaction between Trade and Competition Policy. In 1996, he was appointed Special Advisor to the French Minister of International Trade and Competition on international trade and competition issues. In 1994, he was elected Chairman of the O.E.C.D. Competition Law and Policy Committee. He was General Counsel (Rapporteur General) of the Conseil de la concurrence from 1985-1992. Prof. Jenny served as special assistant to the French Minister for Consumer Affairs in 1977, where he was in charge of the revision of the French antitrust law. He has been Professor of Economics at E.S.S.E.C. since 1972, and has published extensively in the areas of industrial organization and competition law. He has taught as Visiting Professor at Northwestern University (U.S.A.), Keio University (Japan), and the University of Capetown (South Africa). R. SHYAM KHEMANI
is Manager of the Business Environment Group in the Private Sector Development Department of the World Bank. Previously, he was on the Faculty of Commerce and Business Administration at the University of British Columbia in Vancouver, Canada. He has held several senior positions in the Canadian Competition Bureau, including Chief Economist and Director of Economics and International Affairs. He has participated in the work of two Royal Commissions in Canada and has published and edited several monographs, books, and articles in professional journals on issues relating to competition policy. CHRISTIAN KIRCHNER
is Professor of Civil Law, European and International Economics and Business Law at Humboldt University in Berlin. From 1977-1993, he held teaching and research positions at various universities, including Frankfurt University, Tokyo University and Hannover University. He served as government advisor with respect to China prior to 1989, and with respect to Central and Eastern Europe since 1991. His publications focus on German, European and international economic and business law and on law and economics. HlDEAKI KOBAYASHI
is Minister Plenipotentiary, Deputy Chief of Mission at the Embassy of Japan in Washington, D.C. In 1968, he joined the Ministry of Foreign Affairs of
Biographical Notes on the Participants
xxv
Japan, where he served as Deputy Director in the Treaties Bureau from 1983-1984, Director of the Asian Affairs Bureau from 1984-1988, Counselor at the Embassy of Japan in Australia from 1988-1990, Minister of the Embassy of Japan in Poland from 1990-1993, Deputy Director General for Consular and Immigration Affairs from 1993-1995, and Deputy Secretary General of the Japanese Fair Trade Commission from 1995-1997. His publications focus on Japanese competition law and trade and competition policy. VALENTINE KORAH
is Emeritus Professor of Competition Law at University College London and Visiting Professor at Fordham Law School, the College D'Europe, Bruges, and the Universities of Valencia in Spain and of Lund in Sweden. She has authored numerous books and articles on E.C. competition law. She is consultant to IBC on three E.C. competition law conferences annually, and is on the editorial boards of many publications concerned with competition law in Europe, the U.S. and New Zealand. LARAINE L. LAUDATI
is Research Fellow at the European University Institute in Florence, Italy, where she works with Professors Giuliano Amato and Claus Dieter Ehlermann on various research projects involving competition law. She has worked extensively as a legal consultant to the Directorate General for Competition of the European Commission. She has practised law in Washington, D.C. Dr. Laudati has published numerous studies and articles concentrating on E.C. competition law, and has been co-editor of the Robert Schuman Centre Annual on European Competition Law since 1996. ABBOTT B. LIPSKY, JR.
is Senior Competition Counsel to The Coca-Cola Company, Atlanta, Georgia and Secretary of the American Bar Association Section of Antitrust Law. He joined the U.S. Department of Justice Antitrust Division in 1976. He served as Deputy Assistant Attorney General in the U.S. Department of Justice from 1981-1983, and as Special Assistant to the Chairman of the U.S. Council on Wage and Price Stability in 1978. He has also been in private practice from 1979-1981 and 1983-1992. His publications cover a broad range of topics in the law, economics and policy of national and international antitrust and competition, economic regulation and deregulation. MITSUO MATSUSHITA
is Professor of Law at Seikei University in Tokyo and a Member of the Appellate Body of the W.T.O. in Geneva. In 1968, he joined the faculty of
Biographical Notes on the Participants
Sophia University, then moved to the University of Tokyo as a Professor of Law in 1984, which awarded the title of Professor Emeritus in 1994. He has served as a member of various councils attached to the government of Japan, including the Customs and Tariffs Council, the Industrial Structure Council and the Telecommunications Council. His publications focus on competition law issues in Japan. He is of counsel to Nagashima & Obno Law Office in Tokyo. FRANCINE MATTE
is Senior Deputy Director, Mergers Branch, of the Canadian Competition Bureau. From 1996-1997, she was Acting Director of Investigation and Research. In 1994, she became Senior Deputy Director (Mergers). Prior to her appointment, she was Head of Legal Services, Industry Canada in the market framework unit. In that capacity, she advised the Competition Bureau. She has also held various managerial positions and acted as legal counsel for many departments and organisations. She was special advisor to the Deputy Minister of Justice from 1983-1985, and Corporate Secretary to Investment Canada from 1987-1990. She was appointed to Queen Counsel in 1986. PETROS C. MAVROIDIS
is Professor for Public International Law and E.C. Law at the University of Neuchatel. He was previously with the Legal Affairs Division of the G.A.T.T. and worked as a consultant for the O.E.C.D. His most recent publications include The W.T.O. Law and Practice (co-author), State Trading in the 21st Century (co-editor) and Law and Policy in Public Purchasing, the W.T.O. Government Procurement Agreement (co-editor). ADRIAN OTTEN
is Director of the Intellectual Property and Investment Division of the Secretariat of the W.T.O., the responsibilities of which include intellectual property, investment measures, competition policy and government procurement. Previously, Mr. Otten has held posts with the Commonwealth Secretariat in London, where he worked on international trade questions, and with the Swaziland Government in Brussels, where he assisted in negotiations with the E.E.C. in the context of the first Lome Convention. He joined the G.A.T.T. Secretariat in 1975, where he has held a variety of posts. Between 1986 and 1993, he was Secretary of the Uruguay Round Negotiating Group on Trade Related Aspects of Intellectual Property Rights. DAMIEN NEVEN
is Professor of Economics at the University of Lausanne. He has previously taught at I.N.S.E.A.D., the European College in Bruge and the University of
Biographical Notes on the Participants
Brussels. His research focuses on trade and competition, with an emphasis on European matters and antitrust. His most recent publications include Trawling for Minnows, Agreements in European Competition Law (co-author). ALBERTO PERA
is Secretary General of the Italian Antitrust Authority. From 1987-1990, he was a consultant to various Ministers of Industry on antitrust and competition issues, and negotiated the E.C. Merger Regulation. Prior to 1987, he was a staff economist at the I.M.F., and Chief of Economic Research at I.R.I., Italy's state holding of public enterprises, and was involved in its privatisation program. He has taught economics at the University of Rome and public sector economics at Catholic University of Milan. He his publications focus on financial markets, publicfinance,privatisation and regulation. He has also been a consultant to the World Bank and O.E.C.D. on these matters. ROBERT PITOFSKY
is Chairman of the Federal Trade Commission. Formerly, he held positions as Director of the Bureau of Consumer Protection of the Federal Trade Commission; Commissioner of the Federal Trade Commission; Dean and Professor of Law at Georgetown University Law Center; Professor of Law at New York University School of Law and Visiting Professor of Law at Harvard Law School. He has also been counsel at thefirmof Arnold & Porter. His publications include various books and articles on antitrust law. JAMES F. RILL
is a Partner at the law firm of Collier, Shannon, Rill & Scott, where he specialises in domestic and international mergers, acquisitions, and strategic alliances and complex antitrust litigation. He is Co-chair of the International Competition Policy Advisory Committee to the Attorney General and the Assistant Attorney General. He is Vice Chairman of the Business and Industry Advisory Committee, O.E.C.D. Competition Law and Policy Committee. From 1989-1992, he was Assistant Attorney General for the Antitrust Division of the U.S. Department of Justice. During this period, he was responsible for the 1992 Horizontal Merger Guidelines jointly issued by the Federal Trade Commission and the Department of Justice. In 1991, he negotiated the U.S.-E.C. Antitrust Cooperation Agreement. CONSTANCE K. ROBINSON
is the Director of Operations and Merger Enforcement of the U.S. Department of Justice Antitrust Division. She has served as Department of Justice
Biographical Notes on the Participants
prosecutor for 21 years. Ms. Robinson has also served as Vice Chair and Chair of the Criminal Practice and Procedures Committee of the American Bar Association's Antitrust Section. ALEXANDER SCHAUB
has served as Director General for Competition of the European Commission in Brussels since May 1995. From 1990-1995, he was Deputy Director General for Industry. Mr. Schaub previously served in the cabinets of Ralf Dahrendorf, Guido Brunner, Viscount Etienne Davignon, President Gaston Thorn and Willy De Clercq. In 1988, he was appointed Director at the Directorate General for External Relations and Trade Policy. Mr. Schaub has written a book on the European Parliament and articles on economic and commercial topics. MARIO SIRAGUSA
is a Partner at the firm Cleary, Gottlieb, Steen & Hamilton. He is engaged in corporate and commercial practice and specialises in E.C. competition law. Mr. Siragusa began work at the firm in 1973. He is professor at the College of Europe, Bruges and at the Catholic University, Milan. He lectures at conferences in various European countries and in the U.S. on E.C. law. He has published numerous articles on E.C. law. FRANCOIS SOUTY
is Rapporteur in charge of multilateral affairs at the Conseil de la concurrence in Paris, and Associate Professor at the Universite de La Rochelle. From 19901996, he served as Charge des affaires internationales de concurrence at the General Directorate of Competition, Consumers Affairs and Fraud Repression (D.G.C.C.R.F.) in the French Ministry of Economics and Finance. From 1983-1990, he was an investigator with the D.G.C.C.R.F. He has lectured at the universities of Paris, Nantes, Toulouse (France) and Groningen (Netherlands). He has authored five books on competition law and policy and has written extensively in thefieldsof economic policy and economic history. ANTONIO TIZZANO
is Professor of E.C. law at the University of Rome La Sapienza. He has been Professor at the University of Naples and Catania, and at the National University of Mogadishu, Somalia, and has lectured at the Hague Academy of International Law. He has served for many years as Legal Adviser to the Italian Permanent Representation to the E.C. Prof. Tizzano was a member of the Italian delegation to the I.G.C. on the Single European Act and on the
Biographical Notes on the Participants
xxix
Treaty of Maastricht on European Union. His publications focus on E.C. law, mainly on institutional issues, freedom of establishment, financial services, telecommunications, and antitrust. His is founder and co-editor of the review, // Diritto dell'Unione Europea. JAMES S. VENIT
is a Partner at the Brussels office of Wilmer, Cutler & Pickering, where he specialises in E.C. competition law. Mr. Venit has frequently participated in E.C. competition law conferences, both in Europe and the U.S., and has written extensively on various subjects of E.C. competition law. MICHEL WAELBROECK
is Professor of Law at the University of Brussels and President of the Commission de la Concurrence, which was established in 1991 to advise the Belgian government on competition matters. He has been Visiting Professor at New York University Law School and University of Michigan Law School (1969-1970), the European University Institute ((1981), and the Parker School of International and Comparative Law, Columbia University (1980-1989). Professor Waelbroeck has written numerous books and articles on international law and European Community competition law. MARK WARNER
is Legal Counsel in the Division of Policy Inter-relations in the Trade Directorate of the O.E.C.D. Mr. Warner is also Chair of the International Antitrust Committee of the American Bar Association Section of Antitrust Law and a member of the A.B.A.'s Antitrust in the Global Economy Task Force. He is co-author of the Second Edition of the leading Canadian trade law treatise, The Canadian Law and Practice of International Trade (1997). He is also an Adjunct Professor of International Competition Law at the University of Leiden in the Netherlands. REIN WESSELING
is an Associate at the firm DeBrauw Blackstone Westbroek in Amsterdam. He is currently completing a Ph.D. thesis at the European University Institute in Florence, excerpts of which have been published. RICHARD WHISH
is a Partner and Head of the E.C. and Competition Law Department at Watson, Farley, & Williams, London, and Professor of Law at King's College London. He has authored several books including Competition Law (3rd
Biographical Notes on the Participants
edition 1993) and is General Editor, with Peter Freeman, of Butterworths Competition Law. DIETER WOLF
has been President of the German Bundeskartellamt, Berlin, since 1992. He previously was employed with the Federal Ministry of Economics where he headed the Industrial Policy Subdivision and the Competition Policy Section.
PANEL DISCUSSION
COMPETITION POLICY OBJECTIVES
GENERAL RAPPORTEUR:
Prof. Frederic Jenny, Vice President, Conseil de la Concurrence, Paris, France
PARTICIPANTS:
Gabriele Castenada, Partner, Castaneda Y Asociados, D.F., Mexico Anna Fornalczyk, Professor, Vice-president, Competition Development Centre Ltd., Lody, Poland Hideaki Kobayashi, Deputy Secretary-General, Japanese Fair Trade Commission, Tokyo, Japan Francine Matte, Deputy Director, Canadian Competition Bureau, Hull, Quebec, Canada Damien Neven, Professor, University of Lausanne, Lausanne, Switzerland Robert Pitofsky, Chairman, United States Federal Trade Commission, Washington, D.C., U.S.A. Alexander Schaub, Director General for Competition, European Commission, Brussels, Belgium Dieter Wolf, President, Bundeskartellamt, Berlin, Germany
1 - Competition Policy Objectives
• PROF. AMATO—The workshop this year is devoted to objectives of competition policy. This is not a new topic, but we are proposing to discuss it in the context of the European Union, Eastern Europe, and globalisation of the marketplace, where this topic has very new and extremely intriguing elements. It requires a frank discussion, because it is doubtful that we all agree on the goals of competition. Generally, however, we refrain from discussing it openly, and ambiguities remain. • PROF. JENNY—I believe there are three major reasons for discussing this topic. First, many jurisdictions are debating how competition law has been enforced, whether there were political motivations behind decisions by independent authorities, and if so, what those political dimensions of competition policy are. Second, in the context of O.E.C.D. countries, which have a long tradition of competition law and policy, we have considered the themes of convergence, harmonisation, and cooperation, and recognised that there are differences among national competition laws and policies. What justifies these differences? Do the implicit goals of competition policy differ from country to country, and if so, what does that imply with respect to harmonisation, whether soft or hard? Third, we must consider whether common minimum standards might be achieved for the international community through discussions within the W.T.O. or other multilateral fora, and if so, how. This also raises questions of whether there are common goals of competition policy. So whether one looks at the national, regional or international situation, we are always asking ourselves what the goals of competition policy are. I had expected the papers for this session to take a much tougher stance that competition policy should focus only on efficiencies and economic welfare, and everything else should be rejected. However, my impression from reading the papers was that the authors generally recognised implicit political goals as well as economic goals. I do not mean political goals which would be less noble than economic goals. Rather, that certain goals underlie competition policy which may not come straight out of a microeconomic theory textbook. A recurrent theme in the papers is the relationship between competition, economic democracy, and political democracy. If economic democracy will help promote political democracy, this constitutes a political goal which is beyond the scope of microeconomic theory. Referring to Europe, Mr. Schaub says that market integration is the overriding political goal, and is a different goal from achieving economic efficiency. The Commission's Green Paper on Vertical Restraints indicates that this goal is very important in the implementation of European competition policy. Mr. Castaneda says that irrespective of a desire to follow a strictly
Panel Discussion
economic approach to competition policy, the aspiration to achieve fairness and to provide opportunities are important goals of competition policy. Mr. Fels states that one of the political goals of competition policy is to establish limits on the powers of both government and private bodies. Thus, the panellists agree that several types of political goals might underlie competition laws. I would like to raise several questions for discussion. First, does competition policy share the same non-economic goals all around the world, or does it vary from country to country? Second, is it legitimate for competition policy to protect small and mediumsized firms? The consensus in the papers is that political goals are legitimate goals of competition policy, but that the protection of small-and medium sized firms is not a legitimate goal. This begs the question: if we admit that it is legitimate that there be political as well as economic goals to competition policy, can we distinguish between legitimate and non-legitimate political goals, and if so, what are the criteria for doing so? Third, to what extent do the legitimate political goals of competition policy interfere with the strict economic goal of economic welfare? Mr. Castaneda asserts that there is no agreement on what economic welfare is, in policy terms as opposed to theoretical terms. Thus, when competition policy is based on both economic and political goals, adjustments must be made; a strict economic approach must not be followed. If that is true, why are we still inclined to talk about competition policy in strict economic terms? Fourth, do differences in competition laws follow when underlying political goals vary? Prof. Fornalczyk suggests that competition policy in countries in transition regimes takes a more regulatory approach, and addresses problems such as controlling inflation or prices. Does this follow because these countries have a different political agenda? Are some infractions defined as per se unlawful because there is an implicit assumption that economic power has the potential to corrupt the political system and is bad per se, irrespective of the efficiencies that could result? Are the institutions themselves and their degree of independence the result of the political considerations that underlie competition policy? If the differences among countries in competition policies follow from different underlying political goals, what does that mean for the international debates, in terms of harmonisation, in term of convergence? Is it legitimate to pursue harmonisation, or should we recognise and accept that we have different political goals? What are the implications for the multilateral framework? • MR. CASTENADA—There are several interesting aspects of the Mexican case. First, the Mexican Constitution of 1857 contained the first constitutional provision on competition. The 1917 constitution, which is still in effect in Mexico,
1 - Competition Policy Objectives
5
also contains a competition provision that establishes a good foundation for a sound competition policy. Art. 28 of the constitution contains a sweeping prohibition against monopolies and monopolistic practices, as well as a clear mandate to eliminate all horizontal or collusive agreements, and a prohibition of all anti-competitive practices. A new Mexican competition law was enacted in 1993, which takes a fresh approach to competition policy. Four years of enforcement of the new law provides a good framework for competition in Mexico. Second, to advance competition policy, it is important that politicians be committed to its objectives. However, it is difficult to get politicians to focus on this area, because they have many other issues that they must address and competition may not be their most immediate concern. Moreover, in a country like Mexico, which has a highly concentrated economy, strict enforcement of competition rules may not appeal to politicians. Third, we must compare theory to practice. One need only consider decisions and rulings around the world, from judges and competition authorities, to wonder whether all adhere to any pure competition policy. Having a good competition statute is very different from having a good and actively enforced competition policy. For the latter, it is necessary to have transparency, clearly motivated decisions, and social backing. Since 1985, Mexico has been in the process of deregulating and opening the economy. These processes help build social backing for the new competition statute. Fourth, the government is also subject to the Mexican competition law. Finally, the Mexican system contains several safeguards designed to prevent the enforcement agencies from deviating. These include private enforcement and accountability measures, such as annual reports and advocacy from the agency itself. • PROF. FORNALCZYK—I would like to make four points. First, competition policy and competition law enforcement are very important for both industrialised countries and transforming countries, and the goals of competition policy in both are nearly identical: to prevent abuses of dominant position, anti-competitive agreements, and to control merger activity. However, the conditions and the scope of work, as well as the political and economic environment for competition law enforcement, are more severe in transforming countries than in industrialised countries. Second, my experience in Poland and other transforming countries leads me to conclude that the balance between economic goals and social goals is very fragile. Competition policy in transforming countries must focus on economic goals to counterbalance the social goals that are the sole focus of many politicians. All
Panel Discussion
over the world, competition authorities are viewed as troublemakers, but it is especially true in transforming countries. Thus, they must base competition policy on efficiency and consumer interests. Competition advocacy within the government is very important in restructuring programs, in programs of privatisation, and in many issues of government economic policy. Third, I spent one year as consultant to the Russian anti-monopoly committee, and I know how many problems exist in the former Soviet Union. The situation there is far more difficult than, for example, it is in Poland or the Czech Republic. Fourth, harmonisation and convergence of competition law is very important for transition countries. When I was president of the Polish Antimonopoly Office, I often argued to government ministers that we must harmonise our competition law with European Union standards. Such discussions are important in our countries to change the way of thinking and support market oriented reform. • MR. KOBAYASHI—Prof. Jenny's questions appear to be based on the assumption that a clear distinction exists between economic goals and political goals. The political goals suggested by Prof. Jenny include the democratic process, pluralism, promotion of free enterprise, market integration, fairness and social cohesion, among others. However, I question whether such a clear distinction can be made between political and economic goals. My impression is that most of the political goals suggested by Prof. Jenny can also be described as economic goals. For instance, regarding democratic process, "the democratic development of the national economy" is mentioned as an ultimate objective of the Japanese competition law. This was meant to create the basis for prevention of the re-emergence of "zaibatsu," which were gigantic family concerns that controlled the Japanese economy in the pre-war era. This clearly has a political element, but it was basically conceived as an economic objective. Regarding fairness, the provision of the Japanese anti-monopoly law specifying objectives refers to "fair competition". The concept of "fair competition" is generally understood to mean competition by means of intrinsic elements of business, such as price and quality, which again, is basically an economic concept. Moreover, the concepts of market integration and the promotion of free enterprise as goals of competition law are primarily considered from the economic point of view. In some countries, "pluralism" mainly signifies cultural or ethnic pluralism. In Japan and several other countries, however, it signifies the need to preserve a variety of sizes or types of enterprises, which is basically an economic concern.
1 - Competition Policy Objectives
Accordingly, political goals appear to be based on economic goals, primarily efficiency. Political goals may be viewed as intermediary steps that may ultimately advance the economic goal of efficiency. Thus, it appears that it is not necessary to view political goals and economic goals as mutually exclusive. Political goals have many economic elements, which are generally acceptable in most countries. • Ms. MATTE—The views expressed here are my own. The "purpose clause" of Canada's competition act summarises its objectives; I will not discuss that here, because my paper refers to it at great length. Each nation's competition law is adapted to its needs, and follows its historical, cultural and institutional framework. This probably explains why the level of political involvement varies. In Canada, competition is not viewed as an end in itself, but rather as a means to further various other objectives. Our law has evolved through the years. Historically, the major objectives of Canadian competition policy have been to maintain free competition, to prevent abuses of market power (and to protect consumer interests), and to promote economic efficiency. There has been considerable debate over the primacy of these objectives in Canadian government, business, and academic circles over the years. A consensus has emerged, however, that competition policy should focus on the goals of protecting and conserving competition, and promoting economic efficiency. In recent years, concern with economic efficiency as an objective of competition has further evolved to encompass the goal of international competitiveness. Increased efficiency is now widely acknowledged as the key to improving Canadian performance in world markets. Increasingly, the social and political dimension influences our thinking on policy and law. We consult more frequently with interest groups, and do not limit our thinking to just the economic dimension. For example, with regard to the deregulation of telecommunications, we have followed a more social approach. During the transition period, we did not base all of our decisions on pure economic reasoning. We even recommended use of targeted subsidies during a period of rate rebalancing when local rates were being increased. Given my belief that competition law must adapt to the specific needs of each country, I am not convinced that international harmonisation would be possible or would be desirable. Some degree of co-operation can be achieved, and in the U.S./Canada context, much has already been achieved to improve our ability to co-ordinate enforcement efforts. Moreover, through meetings like this, we are sensitised to each other's laws. In the future, this may lead to further convergence.
Panel Discussion
• PROF. NEVEN—I shall make four points. First, the objectives that we assign to the agency in charge of implementing the policy might be better achieved if drawn narrowly. This is a familiar principle from the theory of delegation. Second, we must recognise the constraints under which institutions operate. We must simultaneously consider the objectives of competition policy and the design of the institutions that implement the policy. This may also suggest that in some instances, the objectives should be narrowed. Third, I question whether the objective should be narrowed to competition itself, rather than considering all aspects of economic efficiency. This is the approach that was followed by the E.C. merger regulation, at least formally. I have reviewed the recent economic literature in that respect, and have drawn two conclusions. One conclusion is that productive efficiency, as opposed to allocative efficiency, has received more attention recently. We have learned in the last ten years of empirical work in industrial organisation that price distortions do not matter very much. Instead, what matters is that competition might promote efficiency of firms, and ensure that they run a tight ship, that costs are minimised. The second conclusion is that competition might also provide firms with incentives to enhance dynamic efficiency, but this is more debatable. I concluded from this literature that we might well presume that competition would enhance both productive and dynamic efficiency. This would provide a good case for narrowing the objectives of competition. However, this certainly requires a much more refined concept of competition than the one we have been using, and a more refined concept of what is a restriction of competition, certainly than the one that is currently used in the context of Art. 85(1). Fourth, I believe that the E.C. pursues the objective of promoting market integration very often at the expense of market efficiency. Price differentials among E.U. Member States are more a symptom of market power than a cause of anything. Given the existence of this market power, it is often not efficient to promote market integration, which runs against economic efficiency. • MR. PITOFSKY—First, on the issue of economic versus political goals as the basis for a competition policy, I think we all agree that the primary influence on competition policy should be economics. In pursuing the economic goals of competition policy, that is, by challenging market power, cartels, and abusive behaviour, we are also producing jobs, improving international competitiveness, promoting innovation. Therefore, the differences between economic and political goals are modest. If differences exist, and if there is a political component to competition policy, I believe it should be reflected in how rules are fashioned for future application. For example, a country may decide to challenge horizontal mergers at
1 - Competition Policy Objectives
9
the ten- percent level, or at the thirty- percent level, or at the fifty- percent level. It is not entirely an economic decision; in part, it is also a political decision. In the United States, we must respond to what the legislature thought when it enacted the antitrust laws, and when it revised them, and what the legislature would say today. Most of the support for an exclusively economic approach in the United States comes from academia. However, I do not believe it would be the approach selected by our legislature today if this question were put to a vote. If you introduce a political component, how do you do it? Are all "political" values included? For example, why in the United States do we not interpret our antitrust laws specially to preserve small business? I believe the answer is that we do not really believe that distorting antitrust principles would help small business. We cannot save small food stores from the chain stores by blocking chain store mergers. We cannot repeal the marketing revolution through antitrust enforcement. Moreover, a real commitment to helping small stores compete could be better pursued through tax policy or subsidies. • MR. SCHAUB— I agree with Mr. Jenny that the debate on the objectives of competition policy is very important. Moreover, there appears to be room for clarifications because the papers seem to show divergence of objectives. The differences, however, become smaller and less frightening when we try to structure the variety of goals listed in the papers. In my view it is necessary to distinguish between ultimate and direct or intermediate goals. Ultimate objectives such as prosperity and growth or even pluralism are those which lead political decision makers to adopt competition laws. These objectives are partly political, partly economic. Political objectives have economic consequences and vice versa. At this level the divergences are not that great. Direct or intermediate goals, on the other hand, concern the maintenance of effective competition. They tell us when to intervene and what criteria to use for the assessment. The two categories should not be mixed up. If we did, antitrust policy would be burdened with too long a list of objectives. Furthermore, the criteria for the assessment of cases would become less clear and transparent. Political, social or environmental aspects, in my view, have no place in the direct application of competition law. And—at least since I have been Director-General— criteria such as the protection of certain industries or the promotion of national champions have never been taken into consideration. Under Community law the protection of competition as the mechanism in a market economy is decisive. In addition, market integration is an explicit objective. Within the antitrust family this is rather unique and may be seen as
10
Panel Discussion
a break from the general principle of protecting competition as an indirect mechanism. The importance of this objective will diminish with ongoing integration. However, I personally believe that we still need market integration as a direct objective within our competition rules for its psychological and pedagogical effects.
Observations and Comments • MR. BELL—Within our increasingly globalized economy, effects on competition no longer solely originate from within a jurisdiction's domestic borders. Instead, instances of anticompetitive conduct occurring outside domestic borders that impact or affect competition are becoming more prevalent. In addition, with more than 60 jurisdictions having enacted some form of legislation governing anticompetitive conduct, the likelihood of disparate treatment of alleged acts affecting competition is growing. Since it seems infeasible at this time to expect or even desire substantive convergence of antitrust laws on a worldwide basis, cooperation among competition enforcement authorities appears to be the most viable alternative in addressing these concerns. Cooperation between enforcement authorities can occur on many different levels, depending on the various jurisdictions involved and the degree of confidence each agency has in each other's competition regimes. The flexibility derived from international cooperation can range from simple ongoing dialogue during a multijurisdictional merger review, to the establishment of positive comity principles between two jurisdictions. At either end of this spectrum, the benefits gained from cooperation outweigh the potential frictions caused by unilateral enforcement or extraterritoriality. Some view divergences in both substantive law and domestic social issues as impediments to the possibility of international cooperation; however, the extent to which these divergences create barriers to cooperation has been exaggerated. Cooperation does not entail harmonization of substantive law, but instead seeks to find mechanisms to bridge the variant antitrust laws in order to coalesce in a worldwide economy. While the benefits derived from the use of cooperation in the international competition arena to date have been noticeable, more initiatives need to be pursued to improve such cooperation. If cross-border cooperative efforts are not pursued as a means to avoid or resolve disputes, the remaining alternatives hold the potential to create significant friction between enforcement authorities. Alternatives to cooperation include: extraterritorial enforcement, use of trade remedies (including Section 301 actions), or formal dispute resolution mechanisms. While these mecha-
1 - Competition Policy Objectives
11
nisms may be successful at resolving specific and individual areas of conflict, the resulting friction imposed by their utilization could remain detrimental to future relations between the countries. At present, cooperation efforts appear to hold the greatest potential for minimizing conflicts that could arise between international competition enforcement authorities. • MR. GOLDMAN—I would like to raise several problems with defining the legitimacy of political goals of competition policy. It is impossible to plan submissions to competition enforcement authorities with the uncertainty created by a political override, or when norms are not widely understood. There are inherent risks when political goals are pursued, other than, of course, the initial goals that are set fourth expressly or implicitly in the legislation, which must be established at the outset. • PROF. BOURGEOIS—Is it not making policy when we take decisions on what sort of competition, or how much competition we should have, or where competition should be, in a concrete case? For instance, the European Commission's market integration objective is a decision to pursue competition at the Community level, and not to permit separation of markets. Are we not hiding behind our pencils when we say "no, no we are not pursuing political goals when we are concentrating only on competition"? • PROF. FOX—A distinction should be made between political goals that clearly override competition and efficiency, such as national industrial policy and protection of national champions and those that do not, such as entrepreneurial access to markets on the merits. Setting that aside, economics is a function of politics. This discussion has drawn a line between politics and economics. Some seem to have suggested that, if only we applied economics, the competition law of all nations would converge to the same rules. However, economics is a function of politics within the range of conflicts that arise under the antitrust laws. This is evident when one considers the various schools of economics that have arisen at various times in history, or even simultaneously. Today, certain economists are more sympathetic to the Community system, and others are more sympathetic to the, more laissez faire, U.S. system. For instance, Alexis Jacquemain in Europe and Frederick (Mike) Shearer in the U.S. are sympathetic to one view of "what is competition that should be protected". In contrast, Frank Easterbrook and James Miller in the U.S. are sympathetic to another very free-market idea of competition. Some leading analysts don't purport to separate economics from politics. When he was head
12
Panel Discussion
of the Federal Trade Commission, James Miller would often say that a libertarian value of freedom was guiding his view of economics. Finally, we ought to recognise differences where they exist so that we can deal with them. For instance, the U.S. and the European Commission held differing views with respect to how to evaluate IBM's conduct in the 1980s. In particular, IBM's changes of the interface of its mainframe computers, and the failure to disclose those changes of interface, disadvantaged the makers of peripheral equipment who were forced to rediscover the wheel before they could make a new or improved disk or memory that was plug compatible with IBM's mainframe. United States enforcers dismissed their own action, on the ground that it was pro-competitive and efficient for IBM to be allowed to make product changes whenever it chose. In contrast, the Commission issued a statement of objections, on the ground that these acts were anti-competitive and anti-efficient. Thus, policy values and perspectives are always important, even when we believe we are pursuing competition and economic goals. Economics is a function of politics. • MR. FAULL—It is interesting to hear that in 1857 the Mexican constitution contained competition laws, because people often suggest that the Treaty of Rome created the first competition laws of constitutional status some one hundred years later. At times, it seems the search for goals leads us down fairly dangerous paths. For example, the notion that we are protecting economic freedom, which sounds very noble, has impaired the interpretation of Art. 85 for the last 40 years. Two groups have emerged: those who believe that the objective of Art. 85 is to protect people's freedom to act, and those (amongst whom I count myself) who believe that the objective of Art. 85 is to protect the process of competition. Competition law can't save the world. Other laws and policies address other aspects of the economic system. I believe we should remain modest in our goals. Of course, within apparently neutral bureaucracies, policy choices are being made. Perhaps President Wolf is correct to assert that they should be articulated more openly so that there can be more accountability in the system. In day to day work, most competition authorities represented around this table believe that they are doing their best to ensure that the competitive process is functioning, to prevent at least the most egregious distortions of competition, and thereby to assist consumer welfare. • PROF. HOVENKAMP—In the U.S. today, there is no longer a significant debate about whether antitrust policy should encompass non-economic goals.
/ - Competition Policy Objectives
13
Literally thousands of law review articles have been written on this topic, and many hundreds of actions have been brought by private plaintiffs containing claims based on non-economic goals. However, non-economic goals have been very largely written off in the mainstream debate. No one of much consequence in the U.S. would any longer assert that consumer welfare should not be the central or even exclusive goal of antitrust, or that antitrust should be concerned about unemployment, inflation or other macroeconomic issues. However, a very significant political debate continues with regard to selection of appropriate economic theory. Antitrust continues to have a left, centre, and right in the U.S. All believe that economics ought to be the exclusive engine for determining antitrust outcomes, but opinions differ considerably with respect to selection of the appropriate economic theory. For instance, the 1992 U.S. Supreme Court's Kodak decision held that Kodak might have an obligation as a firm acting unilaterally to share its replacement parts with independent small businesses who service its machines. I regard that as a kind of left wing, scatter-brained, rather harmful decision that has had a lot of bad effects on antitrust in the United States. When Justice Blackman wrote that opinion, he didn't dispute the basic proposition that economics guides antitrust. Rather, he found market power where many would dispute that market power exists, and the conclusion that Kodak might have this obligation to share was an economic decision, not a political decision that recognised a duty under the antitrust laws to aid small business. • MR. HOWE—I shall comment regarding the relationship between objectives and the design of the institutions responsible for implementation of the law. I have recently provided advice to a small developing country regarding how to establish an antitrust law and enforcement system. It was easy to agree that the objective of the antitrust law should be economic efficiency in some broad sense of the word. It was also easy to agree that the main substantive provisions of the law should cover cartels, dominance, and mergers. However, problems began when we discussed institutional design. The World Bank advocates independent agencies to enforce competition law, but what is meant by independent agencies? In any event, this country was not prepared to delegate such authority to an independent enforcement agency, as members of government were concerned about what could happen when jobs are at stake, or when foreign companies want to enter the local market. I left them with an outline of a good modern textbook antitrust law, but with uncertainty regarding who would enforce it and how. If ministers are going to intervene in any event, perhaps it is preferable to have a system that gives them a political override, rather than to have a clear-cut antitrust law that
14
Panel Discussion
is then fudged in its enforcement. Politicians have a legitimate interest in protecting small firms, or indigenous industry or jobs. I believe it is preferable to make this explicit, rather than have apparently clear-cut rules which are then fudged in the enforcement process by the intervention of other criteria. • MR. ABBOTT—Regarding the conflict between economic and political goals, a distinction is being made implicitly between decisions or recommendations made by the competition authority versus those made by the government. In many other areas of government policy, politics may interfere with the achievement of the primary goal. For instance, in areas like shipping or civil aviation, in most countries government authorities make daily decisions, but issues such as effect on the environment are raised and the decisions become political. Similarly, regional development is a clear case where both a social objective and an economic objective are present. Anti-dumping decisions can also become highly political. No one here has suggested that the conflict in policy goals in competition differ from any other area of government policy. • MR. LIPSKY—Economics provides a tool to predict what the ultimate effect of a given form of business conduct, or a rule that regulates that conduct, will be on society's ability to generate wealth through its scarce resources. In the U.S., every antitrust decision must begin with a fundamental and economically rational analysis. Otherwise, the criteria of analysis, whether economic, political, or other, will be unclear. The economic rationality standard is what puts competition into practice in a particular case. Without it, decision-makers would have no way to distinguish behaviour which is per se legal from that which is per se illegal, or subject to any standard between those two. One further point: the proscriptions that will be delivered on an economic efficiency criterion are identical to those that will be delivered on a consumer welfare criterion, except in very narrow and unusual circumstances. • PROF. AMATO—It appears there is general agreement that if goals other than economic goals are pursued, this should be done in a transparent manner. This creates a major problem under the Community system. The Commission may grant a negative clearance under Art. 85(1) because an agreement is not restrictive, or it may exempt an agreement under Art. 85(3) because it is restrictive but beneficial in some other way. Use of the power of exemption violates the transparency principle. Therefore, we should employ an economic assessment and determine whether agreements are restrictive under Art. 85(1); we could then use the power of exemption for other purposes, if other purposes must remain, such as social, regional or industrial policy goals.
1 - Competition Policy Objectives
15
We must, of course, be aware that different views exist regarding economic criteria. In the Kodak case, both the majority and Justice Scalia employed economic reasoning, but according to the majority, Kodak had market power, and according to Justice Scalia, it did not. • PROF. FOX—I do not agree that all credible people in the United States now say that economics and efficiency are the sole accepted criteria for applying the antitrust laws, as has been suggested. There is a left, middle and right of economics, but how can that be the case unless policy or political considerations other than economics are assessed? In the U.S., everyone uses language of efficiency because that is, at the moment, the accepted discourse, but other policy goals are considered. Market access restraints provide an example. Even in the U.S., differences exist as to whether a restraint that blocks market access is illegal only if a plaintiff can prove that it also lessens output, and is in that sense inefficient, or whether such a restraint is illegal if and when it significantly impedes the flow of trade and artificially blocks market access. • MR. FAULL—On the few occasions that we pursue goals other than efficiency under Art. 85(3), our proceedings are transparent: decisions are published; before decisions are rendered, a notice is published calling for third party comment. That is more transparency than is employed by many other systems around the world. I believe it is too easy to say that the Kodak majority was wrong but still based in economics. Of course, the language employed in the decision is that of economics. However, it is not clear from reading the decision what policy that particular economic language was based upon. A cynical European sees in Kodak sympathy for the small dealer. That has not changed from earlier periods in American antitrust history; it is written differently, the language has changed. Many different views are possible on what is politically correct economic language. I don't think that answers the question at all. I do not believe that from reading the Kodak decision, one can conclude that economics is more paramount in the United States than it is anywhere else. • PROF. HAWK—This discussion on the U.S. and Kodak is very interesting. It is not unique to the U.S. As I see it, when a case is presented before the courts or enforcement authorities, it must be argued in the language of economics. To persuade the court or authority, economic analysis must be employed. Economics is supposed to be an intellectual discipline. It is not on a scientific level equal to that of physics, but there is a science to it. However,
16
Panel Discussion
politics also enters into it. We generally no longer argue explicitly in political terms, such as protecting small business; the arguments have become less transparent. However, there is a political motivation that is the basis for the decisionmaker's selection of the economic school that he or she will apply. Nonetheless, the necessity of making arguments in economic terms still constitutes a constraint, to the extent that economics is a science, an intellectual discipline. • MR. SCHAUB—Depending on which economic school forms the basis of the arguments, the capacity or judgement of the individual making the arguments, or the quality of the economist, a decision-maker can arrive at entirely diverging assessments. In specific cases, arguments are presented based on a given economic theory, but lead to diverging conclusions. For instance, regarding the Commission's famous decision in the DeHavilland merger, a divergence was evident in all the internal papers regarding market definition. Some argued that the market was larger, and others argued that it was smaller. This issue was determinative, however. The day after the decision was issued, in particular following the press reports it would have been impossible for the general public to conclude that the decision was based on economics rather than industrial policy and national champions. However, such arguments had been rejected. In many cases, the frustrating situation will exist that notwithstanding brilliant economic arguments, people will suspect that the real motive for a decision is not the stated economic reason, but some other hidden reason. However, as long as human beings are involved, this is inevitable. • MR. FORRESTER—When one goes to hearings at the European Commission or the Court, one can find Nobel scientists on each side of the table, one proving with beautiful charts and diagrams the precise diametric opposite of what the other has argued. It is very important to acknowledge that when law is applied, it will not be done in a scientific or universally predictable manner. It is unlike medical science or nuclear physics, because it is very highly arguable. So much depends on how the kaleidoscope is shaken. It would be an interesting Ph.D. thesis to correlate the asserted purity of the purely economic policies being pursued by some antitrust jurisdictions around the world, and how those policies are actually applied in practice. I suspect that such an analysis of the practice of a super right-wing, purely economics, nonindustrial, non-social, non-consumer, non-anything else administration, compared with such an analysis of the practice of a fuzzy, gentle, green, friendly one like the European model, would reveal that the actual implementation in individual cases could not be easily distinguished.
1 - Competition Policy Objectives
17
• PROF. WAELBROECK—As long as cases are being handled by human beings, we will never agree. However, at least now we all speak the same language, the language of economics. Previously, we didn't know exactly what language we were speaking. When people speak different languages, there appears to be less chance of agreement than if they speak the same language. At Commission hearings, I have frequently heard economists making arguments which are quite opposite from what the Commission says, which also is supposedly based on economic theory. For instance, I have heard very good economists argue that there was no reason to be concerned about a horizontal cartel, because it had no effect on price, and if there had been no cartel, price movements would have been exactly as they were with the cartel. People disagree about whether Kodak is good or bad. Personally, I favour the Kodak decision, perhaps because I would be classified on the left of the economic schools. Definition of the relevant market is important to the outcome of a case, and involves economic judgement. An early Supreme Court decision held that a market share increase from 7% to 10% is very dangerous, because there was a trend towards concentration in the industry involved. However, this analysis was incorrect, because a trend towards concentration may indicate that it is efficient forfirmsto be larger. • PROF. NEVEN—Being underrepresented here, as an economist, I am a bit disturbed by this discussion. I share the concern that a completely unaccountable clique of economists will eventually capture competition policy. However, the problem is not that there is left and right and centre economics, but that there is bad economics and good economics. It is important that the lawyers understand what are good economics and what are bad economics. Economists should be made accountable in the same way that lawyers are made accountable. The first step in making economists accountable is to develop standards for economic arguments, thereby ensuring that lawyers understand them better. The second is to publish what they argue with respect to particular cases. Economists would be much more careful about what they argue at Commission or administrative hearings if their arguments were published. • MR. KHEMANI—I belong to the underrepresented portion of this audience: I am an economist. Unlike the professional requirements of lawyers, there is no minimum entry standard for economists. This often leads to lawyers becoming economists, but not economists becoming lawyers. The political gloss that is often linked with competition in order to make
18
Panel Discussion
it saleable should not be confused with the objectives. Competition legislation is frequently unpopular; to get the legislation passed, it is frequently necessary to attach a political gloss. To find the objectives of U.S. antitrust law, it is necessary to go back to the Congressional debates because the objectives are not articulated in the Sherman Act itself. In contrast, a specific provision articulating the objectives was included in the Canadian competition law. The various objectives mentioned during the U.S. Congressional debates have been given different weights over the past century. The outcome of any particular case relates more to the different weighting of factors, and not necessarily bad economics or good economics, or left, centre, or right-wing economics. Regarding independence of a competition office, I believe it is an important aspect of institutional design, but so are accountability and transparency, and a system of checks and balances. Even with all those safeguards, there are still no guarantees. When I was the chief economist in the Canadian competition office, I had no difficulty in assigning weights or deciding on the primacy of the goals articulated in the law. The primary goals are to maintain and protect competition, in order to advance economic efficiency, with consumer welfare being an important factor. Secondary goals relate to total economic welfare, Until now, there has been no need carefully to weigh this goal in a specific case. The analysis usually ends with consumer welfare. I do not agree that economic efficiency is automatically related to consumer welfare. The latter is a subset that willflowfrom the former, but it will not necessarily lead to total economic welfare. • MR. HEIMLER—I am an economist, and I do not believe that in the Kodak case, there were two economic theories. Two different results were reached, but the theory behind each was the same. It was a matter of assigning different weights to different factors. People have different attitudes towards those factors, but the economic reasoning was the same. The two sides of the Kodak case could easily discuss their attitudes without challenging the hypothesis of each. The hypotheses were well understood. I agree that there are good economics and bad economics, and if results flow from good economics, they are accountable, and can be discussed. I believe that maximising consumer welfare is generally the objective of our laws. This is the case with respect to Art. 85 in its entirety. We should not distinguish results under Art. 85(1) and 85(3); rather, we should see the whole as one system. This allows us to see that Art. 85 as a whole must be interpreted under the constraint of maximisation of consumer welfare.
1 - Competition Policy Objectives
19
I do not believe a competition authority is the right institution to pursue objectives other than consumer welfare. Such other objectives should instead be pursued by a political entity, by government. Competition authorities should base their decisions on consumer welfare, because that makes them accountable and it makes their decision-making transparent. It also allows the theory of their decisions to be discussed. • MR. PERA—I believe that in a competitive system, consumer welfare and efficiency are essentially the same. However, I am concerned when we permit restrictive practices to continue, with or without undertakings imposed, because they improve consumer welfare. This risks allowing competition to become overly regulatory. A paper by Prof. Fox written several years ago discussed differences between the American and the European approach with respect to abuse of dominant position, and in particular the treatment of monopoly rents and excess pricing. The European law, probably for historical reasons, is interpreted to require regulation of the monopoly rather than to disallow the monopoly. I fear that the term 'consumer welfare' could easily be interpreted to mean consumer protection, imposition of maximum prices, and therefore regulation. • MR. MATSUSHITA—In referring to the international aspect of competition law, it is important to distinguish trade policy from competition policy. It appears that trade policy is not based on economic thinking. This leads to conflicts in these two areas of law, such as with respect to anti-dumping rules and voluntary export restraints. In technical areas of trade, the government sponsors many restrictive business practices. • MR. VENIT—One of the central issues we have been discussing is the role of economics in competition policy. Various people in this room have different models in mind when they approach competition policy. On the one hand, there is basic agreement that competition law is based on economics, and that the ultimate obligation that we have as practitioners, enforcement authorities or economists, is to identify the economic analysis that will form the basis of competition law decisions by a regulator. There are, however, some conflicting models. In the European model, the acceptance of economics has come later and more grudgingly. The European model has been essentially a social-political model. For historical reasons, its conception of restriction of competition has been linked to the notion of economic freedom. In the interest of transparency, it is important that we all recognise this difference.
20
Panel Discussion
• MR. CASTANEDA—I am amazed to see that we have some consensus around the table. Application of antitrust law has become more analytical as economic analysis has taken on a greater significance. However, there is also a need for legal certainty, and we lawyers must express our discomfort with respect to the often conflicting views of economists. If a group of economists were to have a round table discussion such as the present one, it would last a year, at the end of which there probably would be no consensus. Economics is far from an exact science, which is well illustrated by the history of economics in antitrust. Economics provides us with useful analytical tools, but we must be cautious about the precision and utility of those tools. Another good thesis topic would be to analyse court and administrative decisions regarding their consistency with economic theory. I doubt this would reveal a high degree of consistency. Second, within the government, there is a problem of conflicting goals. In Mexico, the competition law has accomplished a great deal by influencing policy development in important sectors of the economy, such as telecoms, energy, and electricity. It is important that government officials responsible for such regulatory matters take antitrust considerations into account. Thus, competition law in Mexico had become a consideration in policy development. I believe that competition authorities should be consistently advocating competition within the government as a whole. • PROF. FORNALCZYK—In Poland, it is critical that we have transparency in the decision-making process in antitrust cases. The public must know who is responsible for taking decisions, such as halting the restructuring programme in the coal-mining industry, which resulted in major losses to the public. Based on my experiences in Poland, I am deeply convinced that competition authorities should be independent in their role of enforcing the competition laws, and that their decisions should not be based on social or political concerns. They should also have an influence on other aspects of the government's economic policy. • MR. KOBAYASHI—It may be useful to compare competition policy with medical care. The direct objective of medical care is to cure the patient's illness, to remove impediments to health. Similarly, the direct objective of competition policy is to eliminate impediments to competition such as cartels, abuses of dominance, and so forth. There is a substantial parallelism between the two. The state of health of each individual is different from one person to another. The course of illnesses differs, the effect of injuries differs, and therefore the medical care given to each individual differs. However, the direct objective of med-
1 - Competition Policy Objectives
21
ical care remains the same: to remove the impediment to health. Similarly, with regard to competition policy, the economic conditions of each nation differ, and impediments to competition differ. Therefore, the measures taken under the competition law to remove those impediments must differ, although the objectives of the competition policy are the same: to protect the process of competition. Finally, regarding the ultimate objectives of medical care and competition policy, parallelisms exist as well. Some doctors are modest and curing alone might satisfy them, but many doctors hope that by curing a patient, that patient will be able to work effectively, creatively, and happily. But achieving the ultimate objective is not within the control of the doctor; rather, it is up to the patient to realise the ultimate objective. Similarly, the competition authority can remove the impediments to competition, but the ultimate objectives of development of the national economy, or improving international competitiveness of national industry cannot be achieved by the competition authority on its own. Rather, this depends upon the attitude of the private firms in question, and upon policy measures of other government agencies. • Ms. MATTE—Much of this debate has focused on consumer welfare and economic efficiency. The competition Bureau generally considers efficiencies within a total welfare test, which differs from a consumer welfare test. The total welfare test does not necessarily mean lower prices for the consumers. While the total welfare test is not necessarily a better test, it may be better suited to the Canadian economy. We are under growing pressure from outside groups, who may not have a good understanding of competition policy, to employ a public interest test, which does not exist under our present law. We were faced with such pressure in two recent cases. One was related to the newspaper industry and the other concerned the gasoline industry. Therefore, it appears that whenever we consider amendments to the Competition Act, we will be faced with such pressure. I do not believe that we have been successful in educating small- and medium-size businesses, or the average consumer, as to the significance of competition law. I surmise that this is why we are under such pressure. Therefore, we have begun to place greater emphasis on public education initiatives. • PROF. NEVEN—There is a somewhat surrealistic aspect to the debate about the role of economics, because there is no alternative to using economic analysis. Economists have made a strong case that they have a powerful approach. I believe we should strive to make better use of economic analysis, by making economists more accountable and by defining standards for economic analysis.
22
Panel Discussion
Regarding the link between laws and institutions, I agree that it is useless to design laws when there is no commitment to implement them. Institutions should be designed at the same time as laws, to ensure that the laws are going to be properly implemented. Consequently, it is likely that institutions will differ from country to country, depending on the circumstances within each country. Institutions in transition economies should be quite different from institutions in developed market economies, simply because they are facing different constraints. However, I believe there is a normative aspect with is common to all institutions: that transparency is always a positive element, irrespective of the institutional environment. More transparency will always result in more accountability, whatever the institutional framework. This must be helpful, not only with respect to the domestic matters of each nation, but also with respect to international co-ordination of competition policy. There is much confusion about co-ordination of competition policy at the international level, because there is a wide discrepancy between the declared objectives of national competition agencies and the perceptions of the various competition agencies regarding the practice of their counterparts. Thus, the international co-ordination of competition policy would be greatly supported by more transparency and accountability within each national agency. • MR. PITOFSKY—There is wide agreement here today that competition policy is designed to protect the free market, and that the free market in the long term will achieve a wide assortment of goals, some of which are economic, others social and political. Protecting the free market signifies protecting the process of competition and not individual competitors. Some of what has been said here is theflaringup of an argument that has perplexed people in the U.S. for some time, and that is whether the goals of antitrust are best achieved by a system that is exclusively economic or only predominantly economic. Prof. Hovenkamp states rather strongly that people who don't think that antitrust should be interpreted exclusively to serve economic goals are of no consequence and have been written off. I would like to emphasise that we disagree on this point, but the disagreement is quite narrow. A competition system is only sensible if it is predominantly guided and motivated by economic considerations. The problem with Prof. Hovenkamp's statement is the word 'exclusive.' I agree that in court, it is not constructive to argue about principles, or why the law is there at all. However, one must ask why the law is as it is. In the U.S., there is a per se rule against resale price maintenance and a statute addressing price discrimination—rules that might not exist in an exclusively economic universe. Perhaps
1 - Competition Policy Objectives
23
most importantly, U.S. merger policy does not accept Robert Bork's thesis that from an economic perspective, a merger is not harmful as long as two or three firms remain in the market after the merger. If economists controlled the law, perhaps our laws would be different in these areas. Thus, I believe these issues must be analysed as a matter of overall policy, and not on a case-by-case basis. Finally, I shall comment regarding the issue whether antitrust laws should be enforced in such a way as to support the enforcing country's ability to compete in international trade. Most people in the U.S. believe that the best way to help American firms compete in a global marketplace is to force them to compete vigorously at home. Thus, a firm could not defend anti-competitive acts by claiming that it is restricting competition at home in order to assist exports. However, I wouldn't expect that to be the rule in all countries—for example, in a newly privatised, decentralised economy that is concerned with protecting infant industries and facilitating exports. There are many different levels of development between the U.S. and other industrialised countries on the one hand, and newly decentralised economies on the other. I conclude that the political component of antitrust that is often present to a modest degree will vary according to place and time, and that is as it should be. • MR. SCHAUB—I agree with the comment that we should not hide behind our pencils and deny that enforcement of competition laws involves a political element. Of course, it is not the business of a competition enforcement agency, in its day-to-day work, to be open to pressures from political lobbies, powerful governments, or social groups. On the other hand, it is unhelpful and counterproductive if it behaves as if it were isolated from the broader political, economic system. Competition enforcement agencies are created by parliaments, by governments for clear reasons. They will be maintained as long as governments and the democratic process determine that they are delivering desirable results, not in the short term, but in the medium-long term. I believe it is a mistake to ignore this context or imply that these are nefarious links, because it threatens public support and acceptance of our work. Second, I believe that the basic logic of competition policy in developing countries should be the same as it is in our more experienced countries. However, in these countries, there are situations where a trade-off should be made between short-term pressures and long-term positive expectations. It is not desirable to do this, but it would be a mistake to ignore these constraints which exist in the real world. If we would like to assure the political acceptance of the competition system as it begins to develop in these countries, we may be forced to make certain compromises. However, what compromises can be made should be clearly set forth in the legislation, and thus be a political deci-
24
Panel Discussion
sion following the democratic process. It would be undesirable to allow such compromises to be made day to day, under unforeseeable nefarious influences or corruption. The real problem does not have to do with acceptance of a tradeoff, but with implementing a law that allows a trade-off in an acceptable and positive way, because there is an enormous temptation to succumb to pressures. Even a perfect government, and a perfect enforcement agency which resists these pressures and attempts to make such trade-offs in a positive way, would succeed only if it had a very rare combination. First, it would need an excellent economist who is able to foresee the economic consequences that will follow from competition-inspired decisions. Second, it would need good political judgement in order to anticipate the political consequences of its decisions. This combination is extremely rare in industrialised countries, not to mention developing countries. Finally, regarding the similarities and differences between the objectives of trade and competition policy, I believe there is a common element. We are all under enormous pressure in executing our day-to-day activities to consider trade policy, regional policy, industrial policy, and competition. The pressure is the same. The difference, however, is that while it is legitimate to balance these considerations in the trade area, in the social policy area, and in the industrial policy area, it is not legitimate that we give in to these pressures or negotiate in the competition area. • MR. WOLF—In Germany, we do not favour political trade-offs and compromises in the implementation of our competition policy. Competition has no lobby. Thus, if we allow negotiations with political forces in this area to take on the same importance as political decisions in other areas, competition will always be the loser. Institutional safeguards are needed to ensure that competition has at least one clear voice. If the institution can take into account other political goals, there is a problem. I do not believe it is a question of the role of economists versus the role of lawyers. In the Bundeskartellamt, we have 50% lawyers and 50% economists, and they have an equal impact on decisions. Of course, economic arguments must be taken into account in our decision-making. However, economists have learned to be advocates of the interests of their clients, just as lawyers have learned to do so. Thus, reliance on economic arguments does not protect the process. Independent institutions combined with judicial review by courts is the best way to guarantee objective results. Competition authorities must, of course, be staffed with lawyers and economists. However, as long as competition authorities are not independent, the risk is always present that specific
1 - Competition Policy Objectives
25
interests will influence their decisions. Therefore, I strongly advocate the creation of independent institutions that are able to guarantee the enforcement of competition policy. I believe that Europe is still deficient in this regard. • PROF. JENNY—To summarise this discussion, we have agreed that although the specific goals of competition policy differ from one country to another, the ultimate goals of competition policy are political goals that seem to be compatible and consistent with economic goals. If this is so, why do only sixty countries have a competition law, and why is there fierce resistance by many countries to adopt one? A second question: is it true that competition authorities deal only with the narrow agenda of economic welfare? I agree that competition authorities attempt to narrow their focus. Prof. Neven has provided some theoretical reasons to argue why this may be a good idea. However, the reason may not be that competition authorities are modest, as Mr. Faull suggested, but that they want to insulate themselves from political influence. Nonetheless, they take political objectives into account in the enforcement of competition laws, by making a choice of which economic theory—left, right, or centre—-to follow. Some participants, especially the lawyers, have argued that the ultimate goals of competition policy may be political, but competition authorities do not deal with ultimate goals. Rather, they deal with immediate goals, and in doing so, they rely on economic analysis. Others argued that in relying on economic analysis, political decisions are made. Economic consultants are advocates, able convincingly to argue either side of an issue. This implies that competition authorities must make a political choice that is not transparent to those outside the process. Therefore, they have a hidden political agenda, which follows from their discretion to base a decision on ambiguous economic evidence. I do not believe that the issue is whether economic theory is good or bad, or whether it is left, right, or centre, as some have suggested. Rather, I believe the issue is how the competition authority should interpret economic evidence that is very difficult to interpret. Economic theories may not be precise enough to provide guidance. Some have suggested that it is not a question of political agenda of competition authorities. Rather, the problem is that we're humans, we all make mistakes. The lawyers suggested that non-economic goals lead to more uncertainty than economic goals. I do not agree. A populist policy is very determined and constant over time. Some suggested that the political aspect of enforcement is hidden, and enters through unpredictable interpretation of the law. Thus, it is not the uncertainty of the ultimate goals or of the mandate of the competition
26
Panel Discussion
authority, but how the authority executes their responsibility that leads to uncertainty. Whether this uncertainty is sufficiently limited by requiring arguments to be based on economic theory is questionable. Prof. Neven recommended that decisions should be transparent, and that economists should be held accountable. However, considering the U.S. experience, many leading economists who participate in competition cases expose themselves to public scrutiny through publications. Many of them are recognised by their peers as being competent in their field, while still holding different views on how to define a market or whether a particular merger is going to lead to efficiency. Thus, I question whether being exposed to public scrutiny will solve the problem. The question was raised whether competition authorities should be independent. Some here have suggested that competition authorities should advocate competition with respect to various other government policies. However, I question whether it is legitimate for an independent agency to play the role of advocate, since, as an independent agency, it has no mandate to do so. I accept that a government may decide that all its departments should take competition considerations into account. However, it appears inconsistent with the idea of having an independent agency if it is to lobby government bodies, when it is not itself part of government. Finally, if competition policy is linked to the concept of democracy, what does that imply for countries which are not democracies? Should adoption of a competition policy not be advocated in those countries, or is competition policy a means which could be used to encourage these countries to change their political systems?
Frederic Jenny Professor of Economics (E.S.S.E.C.) Vice-Chairman, Conseil de la Concurrence Paris, France Report of the General Rapporteur Of Session One
Introduction It is useful first to address a question that appears in Mr. Schaub's contribution: why should we be discussing the legitimate goals of competition laws? He observes that 'normally, these issues are debated when a political decision on the adoption of a competition law needs to be taken. One may ask why we are having this debate now, since we are not in the middle of adopting a completely new law, and the decision on the basic principles has already been made.' One reason this topic is important is that it will be a determinant of the future organisation of world markets in the face of the globalisation of business. The elimination of custom duties and other governmental impediments to international trade is not sufficient to create competitive international markets. Thus, market access and competition at the global level are still insufficient. The globalisation of markets compels trading nations to find a way to handle international competition issues, through the adoption and convergence of domestic competition laws combined with cooperation among national competition authorities, or through the adoption of minimum standards of competition law, or through the adoption of a supranational competition law in the context of the W.T.O. Whatever the solution to be adopted at the multilateral level, it will require agreement among the trading nations regarding the legitimacy of competition policy principles. At present, no more than 60 countries have adopted a competition law, and in some of those, competition policy does not play an important role. Thus, in more than 100 countries, the need for competition laws or competition policy is not strongly felt. The question, then, is whether the countries that have experience with competition law and policy can convince other countries that it would be in their mutual interest to promote competition through the adoption of domestic or regional competition laws, or through the multilateral adoption of competition standards.
30
Working Paper I
A. The Distinction Between the Ultimate Goals and the Direct Goals of Competition Policy The contribution of Mr. Kobayashi analyses Art. 1 of the Japanese Antimonopoly Act (A.M.A.), which states the goals of the Act. Observing that this article is rather lengthy, Mr. Kobayashi states 'when it was introduced 50 years ago, competition law was something totally unknown to political leaders, government officials and the general public in Japan'. Mr. Kobayashi then states that the A.M. A. has 'two types of objectives: direct and ultimate, where the former are instruments to obtain the latter'. In his contribution, Mr. Schaub adopts a similar view by stating that '[b]efore entering the issue of objectives, one should always distinguish two levels of the debate: [t]he level of political decision making and the level of law enforcement. This distinction is important, because the first level concerns more fundamental or ultimate goals, while the second concerns operational criteria for enforcement practice.' In examining the maze of goals assigned to competition law, it may be useful to follow Mr. Kobayashi's and Mr. Schaub's approach, and to distinguish between the ultimate objectives and the intermediate objectives.
B. The Ultimate Goals of Competition Policy Mr. Kobayashi states that Art. 1 of the Japanese A.M.A. establishes that the Act's ultimate goal is to 'assure the interests of consumers in general, and also to promote the democratic and wholesome development of the national economy'. This assumes that the promotion of competition is not a goal in itself, but rather a means to achieve two underlying goals: an economic goal (the wholesome development of the national economy), and a political goal (as competition is linked to the idea of economic 'democracy'). Following a similar line of reasoning, Mr. Wolf states that 'political and economic systems are interdependent. Political freedom goes hand-in-hand with economic freedom, since the freedom of economic activity is a vital constituent of the basic rights of personal and political freedom.' Mr. Schaub believes that the ultimate objective of competition policy is 'economic prosperity or the welfare of citizens, employment, and social cohesion'. But he adds that political, social, and philosophical arguments can also justify the adoption of a competition law by adding, 'one may debate whether competition rules could also be an instrument to serve pluralism and democ-
2 - Competition Policy Objectives
31
racy'. Finally, he adds that '[i]n the Community, competition has one additional fundamental objective: to promote market integration'. Although he does not categorise this goal, I suggest that it is political rather than economic. In his contribution on the Mexican competition law, Mr. Castafieda argues that competition policy is not designed only to achieve economic efficiency, which seems to be a result rather than an aim in itself. Rather, competition policy is also concerned with subtle political components. Further, he argues that elementary notions of economics feed the popular belief that market fairness is a value to be protected. He then offers the view that '[e]conomic competition also involves: providing opportunities for market entry; promoting of innovation; decentralising power; balancing political considerations; popular myths that generate votes; eliminating red tape needed to keep jobs and investments; promoting national champions; erecting of trade barriers; and so on'. Prof. Fels and Mr. Edwards state that 'there are political and social ends of competition, quite aside from efficiency effects, which make the fostering of competitive processes desirable. In the present timeframe, these aims are often neglected, as the focus is mainly on economic efficiency effects. However, it is these political and social arguments, rather than models demonstrating economic efficiency benefits, which have "tipped the balance of social consensus towards competition"'. More specifically, they add that competition leads to the fulfilment of political and social goals, such as limitation on the powers of both government bodies and private individuals, reduction of concentrations of economic power, sheltering the political process from the influence of economically powerful interest groups, protection of small businesses, freedom of opportunities, development of initiative, self reliance andflexibility,freedom of choice, more equal treatment of workers, and increased economic security, among other things. These developments suggest that there is wide agreement that although the most important ultimate goal of competition policy is economic development, competition policy can also help achieve political goals, such as preserving 'the democratic process', 'pluralism', 'free enterprise', 'fairness', 'market integration', etc. Moreover, these political goals may have more appeal to a country deciding to adopt a competition policy than the strict economic goal of seeking economic prosperity. Recognition that the ultimate goals of competition policy may be both political and economic raises several questions. First, are these political goals simply additional possible benefits of a competition policy basically designed to foster economic development (i.e. by-products that can be used by governments to 'sell' a competition policy), or are they intrinsically legitimate competition policy objectives? Second, if, in addition to the traditional economic
32
Working Paper I
goals, the ultimate objectives of competition necessarily include political goals, such as those listed above, does this imply that competition policy is desirable only to the extent that its implicit political goals are themselves desirable? If so, does this weaken the case for competition policy because some governments might not find the political goals of competition policy desirable? Prof. Fels and Mr. Edwards say that they would challenge the assumption that 'political arguments are necessarily against the fostering of competitive processes'. There is, however, political opposition in certain countries to the adoption of a competition law, and perhaps this opposition results not from rejection of the ultimate goal of economic development, but rather from opposition to the political implications of competition policy. Third, if Prof. Fels and Mr. Edwards are correct in stating that 'the adoption of a competition law is a political act, and as such, political considerations should be paramount', what are the acceptable tradeoffs between the attainment of the implicit political goals and the economic goal of competition policy? Is a competition policy based on the political goal of promoting 'fairness' or 'pluralism' (and in which presumably the under-achievers will be protected from the rigours of competition or from disappearing) legitimate? Will this not lead to a lower level of competition than a competition policy more squarely based on economic considerations? Stated differently, if we recognise that competition policies have implicit or explicit political goals, and that these political goals are not necessarily the same in all countries, is it not true that the concept of competition policy becomes rather fuzzy, and that the prospect for common standards of competition policies becomes quite remote? Can we draw up a list of'legitimate' political goals and a list of 'illegitimate' political goals for competition policy? The remainder of this report will hopefully shed some light on the answers to these questions.
C. The Direct Goal of Competition Policy In his contribution, Mr. Wolf hints that competition can be desirable irrespective of its contribution to economic efficiency. He recognises that 'competition should not take place in a vacuum. Considerations other than competition can be legitimate, and therefore should be included. In particular cases, overriding public interest arguments prevail over the competition principle, but the public should be well informed when that occurs.... If the decision making process in the context of competition law is not transparent, the cost of political intervention is no longer identifiable.'
2 - Competition Policy Objectives
33
In their contribution, Prof. Fels and Mr. Edwards suggest that competition is desirable for several reasons. First, 'the benefits to both static and dynamic economic efficiency conferred by competitive processes must be acknowledged. A perfectly competitive economy would yield an efficient use of resources in both the allocative and productive senses.' Second, competition leads to the fulfilment of desirable political and social goals. Thus, they conclude 'there are political and social ends of competition, quite aside from efficiency effects, which make the fostering of competitive processes desirable'. Mr. Kobayashi notes that the first article of the Japanese A.M. A. states that the direct objective of the Act is 'to promote free and fair competition' in order to 'stimulate the creative initiative of entrepreneurs, to encourage business activities of enterprises, to heighten the level of employment and people's real income'. Thus, competition is seen as means to attain certain economic results, which will contribute to the attainment of the ultimate political and social goals established in the Act. Although economic efficiency is not mentioned in the Act, Mr. Kobayashi asserts that stimulating the creative initiative of entrepreneurs, encouraging business activities, or heightening the level of employment and people's real income are not unrelated to economic efficiency. However, he observes that in cases where there could be a conflict between efficiency and competition, protecting competition may be considered more important than achieving efficiency. For instance, he states that 'an increase in efficiency would not, for instance, render legal an otherwise illegal merger'. Mr. Schaub considers that competition is 'an instrument or intermediate objective to achieve economic prosperity and employment. . . . Competition .. . forces companies to run a tight ship, to adjust to changes, and to innovate. It thereby benefits the consumer and promotes the welfare of society in general. This is why we protect competition, in Europe and abroad.' Thus, he believes competition is valuable because it leads to efficiency, which in turn contributes to economic development. Similarly, Ms. Matte states that, in Canada, 'three major objectives of Canadian competition policy have been identified. . . . They are: maintaining free competition; preventing abuses of market power, including the protection of consumers; and promoting economic efficiency. . . . In the 1970s, there was vigorous debate over the primacy of these objectives in government, business and academic circles. A consensus emerged that competition policy should focus on the goals of protecting and preserving competition and promoting economic efficiency. Subsequently, concern with economic efficiency as an objective of competition policy in Canada and elsewhere has further evolved to encompass the goal of international competitiveness.'
34
Working Paper I
Although Mr. Wolf might disagree, in these other contributions we find that the direct goal of competition policy is to promote static and dynamic efficiency. Because competition policy does this, it will necessarily contribute to the ultimate goal of economic welfare. The authors of these contributions tend to view the economic goal of promoting efficiency as the exclusive immediate goal of competition. Thus, they appear to conclude that the political benefits of competition policy, discussed above, are the only possible by-products of competition policy. If this is true, the authors may believe that the political goals of competition policy should not interfere with the attempt to promote economic efficiency. If there is a tradeoff between the political and economic goals in the design or implementation of competition policy, in most cases the economic goal of promoting efficiency should prevail. This seems to be in line with the statement of Prof. Fels and Mr. Edwards that: A tradeoff must often be made between the efficiency and equity effects of economic policies. Policies that offer only minor efficiency gains and have undesirable effects on equity should tend not to be implemented. However, competition policy has a sound efficiency justification, and other tools of social policy, such as taxation and government spending, may provide more appropriate and directed solutions to distribution problems. Many of the 'political' challenges made to competition law and policy are not, in reality, based on genuine public interest concerns. Rather, they are challenges advanced by sectional groups with political clout, advanced in their own narrow sectional interests, rather than in the true public interest.
D. The Direct Goal of Competition Policy in Developing Economies and Economies in Transition Mr. Castaneda offers a note of caution: '[tjhere is general agreement that competition policy should strive to achieve economic welfare, but there is no agreement on the definition of economic welfare, how it can be promoted by antitrust policy, and how it should be measured. This debate seems similar to the familiar and unresolved question of whether political considerations are more important than economic considerations in determining whether a country is able to function properly. It does not seem possible to arrive at a single definition of economic welfare.' Thus, Mr. Castaneda appears to suggest that for competition policy to contribute to economic welfare, certain conditions are necessary. These conditions
2 - Competition Policy Objectives
35
might already exist in most developed countries. However, they do not necessarily exist in less developed countries, where: economic concentration is likely to be higher; the legal system is likely to be less developed; and economic agents are generally less able to redirect their resources into promising fields due to cultural values, lack of entrepreneurship, or unequal access to education, etc. In such economies, the direct goal of competition policy cannot realistically be limited to improvement of the allocation of resources in a neoclassical sense. Providing economic and social opportunities through a gradual process of opening markets (i.e. increasing market contestability) might be a better characterisation of the immediate goal of competition policies in such countries. In her paper, Ms. Fornalczyk makes a similar observation: '[i]n emerging markets, it is important to build a commitment of society to competition in all areas of public activity. At the beginning of market reform, competition seems like a strange and undesirable phenomenon. Many individuals view it as the root of difficulties which they face. Promotion of competition should be pursued by consistently enforcing competition law, encouraging state authorities to consider competition when taking their decisions, and advocating competition to society. Each of these activities is an important element of bringing about progressive change in the mentality of the people and their behaviour as consumers and producers, as well as improvement in the efficiency of the economy.' In summary, in developed countries, the immediate goal of competition policy is to promote an efficient allocation of resources in the traditional economic sense, and therefore such policy focuses on market behaviour of firms and control of mergers. In contrast, in developing countries or countries in transition (i.e. countries in which the preconditions for a market economy are not fully established), the immediate goal of competition policy is much broader, because: there is less consensus among public policy officials or politicians about the desirability of competition policy; the economic, legal, social or political structures are less appropriate for the development of a free market economy; and the public places greater weight on the short term disruptions of a market economy than on its long term benefits. As a result, the immediate goal of competition policy in these countries appears to be to contribute to the emergence of economic opportunities and entrepreneurship in a context in which more attention must be paid to establishing the political acceptance of a market economy (and therefore taking into consideration the populist goals of competition policy). Ms Fornalczyk provides a concrete example of this when she states that because price liberalisation may lead to price increases, '[t]his placed enormous pressure on the competition enforcement agency to mitigate the price-making
36
Working Paper I
policy of monopolies. In the short run, many of even the most liberal reformers were interested in the agency's price control function.' Similarly she states that '[fjrom the perspective of consumers interest, monopoly profits are not justified, but they allow the inefficient enterprise to survive. The bankruptcy of such a large enterprise may cause significant unemployment and social unrest, which is often stimulated by political opposition and trade unions. Since the number of such enterprises is considerable in a transition country, this problem may substantially impede progressive changes in the economy. The problem concerns declining industries, which should be the objects of the government economic policy rather than the targets of competition law enforcement.' More generally, she correctly states that there is a tradeoff between competitionfriendly government policy and effective fulfilment of legitimate competition policy objectives. As a result, competition policy in a developing country often takes a more regulatory approach, thereby allowing it to contribute actively to the transformation of economic structures and behaviours. It may accomplish this through restructuring and privatisation of industries orfirms,or through price controls for dominantfirms.Moreover, the advocacy function of the competition authorities vis-a-vis the rest of the government, and the propaganda function vis-a-vis the public, are seen as crucial. This is not meant to suggest that the latter functions are not important in the competition policy of developed countries. For instance, Ms. Matte's paper states that in Canada '[t]he Bureau views competition advocacy and enforcing the Act in particular sectors to be complementary. Effective advocacy which results in liberalisation of regulations governing particular sectors must be followed by vigorous enforcement to prevent private anti-competitive practices that could nullify the gains from regulatory reform.' Similarly, Mr. Wolf states '[c]ompetition has no lobby, and one of the foremost tasks of competition policy is to make people aware of the competition principle.' It seems, however, that the scope of the advocacy function of competition authorities in developing countries must be much greater than it is for their counterparts in developed economies. If for no other reason, this is so because in developed economies the advocacy effort is principally aimed at eliminating sectoral regulations in order to allow competition to prevail. In contrast, in some developing countries, the advocacy effort must be aimed at establishing the conditions (regulatory, structural, legal or cultural) which will enable a market economy to function.
2 - Competition Policy Objectives
37
E. Goals of Competition Policy in the Context of Multilateral Discussions The panellists hold sharply varying views on the extent to which it is legitimate to use competition law to achieve economic results that are not strictly related to the narrowly defined concept of economic efficiency (such as preventing price increases, maintaining the level of employement, etc.). For instance, Mr. Fels and Mr. Edwards stated that '[generally, competition law should not be used directly to achieve economic objectives which are not strictly economic efficiency objectives, such as preventing price increases, maintaining employment levels, etc. That is, competition law should not be distorted to achieve these objectives. Indirectly, these other economic objectives should be aided by an effective competition law.' Similarly, Mr. Wolf states that '[s]ocial considerations ought to be taken into account, but only after competition has performed its allocative function.' In contrast, Ms. Fornalczyk stated that '[i]n the short run, many of even the most liberal reformers were interested in the agency's price control function' to prevent inflation. Accordingly, the fundamentalist notion that competition policy's immediate objective should be strictly based on efficiency considerations may be appropriate in some countries but not in others. Economic analysis demonstrates that the social benefits from competition are to be reaped in the long run, but that competition entails private costs in the short run. These short run costs may be quite high in countries in which the economic structures, the legal environment, the political system, and the culture are not conducive to a stable and smooth-functioning market economy. Moreover, when considering the adoption of a policy, the discount rate used to compare the long run benefits and the short run costs is, in any event, dependent on individual preferences. We cannot claim that governments or citizens can never legitimately consider that the short run costs of adopting a policy outweigh the long run benefits. Assigning to competition policy political goals unrelated to the search for economic efficiency (or allowing it to include politically-based exemptions) may, in some cases, mitigate the short run private costs associated with the adoption of such a policy. This could increase its political acceptability, as well as the possibility that, at a later stage of economic development, a consensus will emerge in favour of the adoption of a competition policy more squarely based on efficiency considerations. Mr. Castafleda's discussion of the 1917 Mexican Constitution shows how competition policy provisions inspired by a combination of economic and
38
Working Paper I
political goals, and allowing wide possibilities of regulatory exemptions reflecting populist concerns, can eventually serve as the foundation on which to build an efficiency-based competition law. In order to attain the ultimate economic objectives of competition policy in developing countries, it may be advisable to start with a less pure, more global approach, in which the immediate objectives of competition policy reflect a combination of economic and socio-political or populist concerns. Once the structural and cultural preconditions for a properly functioning market economy have been attained, it would then be possible to shift to a more economically orthodox, efficiency-based competition law. Three questions should be considered: 1. Can, or should, safeguards be found to ensure that this gradual process will actually occur, and that the populist goals originally included in the competition law will not defeat the ultimate purpose of achieving an efficiencybased competition policy? 2. With respect to countries that do not see the need for competition policy, either at the domestic or international level, is it counterproductive to present the argument that such a policy is desirable because of its long run efficiency-increasing global benefits? Would it not be more productive to acknowledge that competition policy that is strictly efficiency-based may entail short run costs that are too high for some countries, and instead insist on the desirability of a competition policy based on a broader concept? For instance, such a broader concept may be the need to ensure that economic agents, whether consumers, firms or countries, should not use their economic power unfairly to deprive other agents of their economic freedom or opportunities. 3. What are the core principles, or minimum standards, of competition policy that one could propose in the context of the international globalisation of markets? This depends in part on the answer to the second question. If the goal of competition policy should be to enhance economic efficiency, minimum standards are likely to be derived from the experience of countries that have adopted and enforced such a law. If a more broadly based concept of competition policy is accepted, the definition of common standards or core principles may first require a re-examination of the relationship between the political economy of development and competition. These three questions, among others, will be discussed in the context of the WTO Working Group on the interaction between trade and competition policy, which was set up following the 1996 W.T.O. Singapore conference. It would be
2 - Competition Policy Objectives
39
entirely inappropriate for the author of this report to attempt to second guess the outcome of this Working Group, and that is not his intention. In any case, the Working Group will be devoted to the interaction between trade and competition policy, rather than to competition policy alone. Nonetheless, the above discussion of the papers presented in Session One of this conference suggests that when thinking about competition policy, several points should be kept in mind: 1. Competition policies and competition laws are not the same in all countries even though, broadly speaking, they share the same ultimate goal of promoting economic efficiency and economic development. The differences among competition laws or policies in different countries may reflect legitimate differences in the immediate goals assigned to such laws, which themselves are the product of differences in the economic, legal and political environments in which such laws or policies are implemented. However, as economic development sets in and as markets become more globalised, there is a natural tendency toward convergence of competition laws and policies across countries. 2. For a competition law or policy to be effective in the long run in achieving the goal of economic efficiency, it is essential that the enforcement of competition policy and law be, at all stages, transparent and non-discriminatory. As time goes by, and as economic development increases, this transparency is probably the best guarantee that the initial competition policy and law will be modified, that it will become less regulatory, and that its enforcement will be progressively fine tuned in order better to serve its long run goal. 3. In the initial stages of economic development, or in countries in transition, competition policy or competition law must be complemented by the development of a broadly-based competition and entrepreneurship culture. The existence of such a culture cannot be assumed to exist in all cases. Accordingly, in addition to enforcing the law, competition authorities have an important advocacy role to play. 4. As the interpenetration among national markets progresses, and as national economies become more interdependent, cross-border anticompetitive practices become more important. National competition authorities, having limited territorial jurisdiction and enforcing different competition laws, are not usually in a position to meet the new challenges raised by the growing relative importance of such practices. Thus, the international community must devise a mechanism which will provide a way to discourage such practices, while at the same time allowing for national differences in competition laws.
II Gabriel Castaneda Partner, Castaneda y Asociados D.F., Mexico
Introduction Most antitrust scholars and practitioners agree on what are the elements of competition policy, but few agree as to its objectives. Is it aimed exclusively at attaining allocative efficiency?1 If so, what exactly is allocative efficiency and how do we balance it against anticompetitive effects? Is competition policy an instrument to prevent concentration of excessive power? Does it seek to preserve opportunities for the many by preventing abuse by the few? Is it a political tool to keep essential balances and decentralise power? Is it a device to feed a populist creed? Is it a component of democracy translated to markets? Is it a noxious regulation that only increases transaction costs?2 Is it an exotic regulation unenforceable in developing countries, an elegant but futile task, as in W. Herzog's movie 'Fitzcarraldo', which depicts the formidable task of bringing Caruso's opera to the middle of the Amazon jungle? The debate is far from completed. Even the 'purest' economic thinking about antitrust is plagued with inconsistencies and lack of clarity; empirical work to assess the impact of antitrust enforcement is highly controversial; and government antitrust enforcement is often criticised for hindering competition and for not being economically accountable.3 Such debate is not trivial, since principles guide legal instruments that cause specific investment decisions and, often, costly disputes. In order to reap practical results, however, the debate should perhaps be moved from general terms to specific issues. It would be helpful to avoid common confusion of 'legitimate goals of antitrust' with how those goals are implemented. There may be consensus regarding what the general objectives of antitrust are, but there are many ways to implement them. Objectives are ideals that are seldom obtained in real world antitrust 1
As distinguished from 'productive' efficiency, allocative efficiency has to do with society's welfare: what is the best allocation or assignment of economic resources across the whole system. 2 Thomas E. Sullivan (ed.), The Political Economy of the Sherman Act: The First One Hundred Years (1991); William Letwin, Law and Economic Policy in America: The Evolution of the Sherman Act (1965). 3 William J. Baumol & James A. Ordover, Use of Antitrust to Subvert Competition, 28 J.L. & Econ. (1987).
42
Working Paper II
enforcement. As in most areas of economic regulation, there are no requirements as to language or principles. However, as a minimum, a common framework should be designed to measure the level of consistency between antitrust goals and results in those areas where it may be empirically possible. Thus, two questions should be addressed: (i) whether enforcement is consistent with what the statutes require; and (ii) whether results are economically consistent with the legal rationale. This paper explores the rationale and language of the Mexican constitutional provisions on antitrust in an effort to trace the guiding values of competition statutes and enforcement in Mexico. It begins with a brief analysis of those constitutional provisions and how they were translated into the first comprehensive and effectively enforced Mexican competition statute. The main argument of this paper is then set forth: that no single goal is identified as the paramount value of competition policy in Mexico. Rather, a complex mixture of interrelated values is pursued in a process designed to establish and maintain open markets. Allocative efficiency is the least volatile and the most accountable of the principles underlying the Mexican antitrust law. The discussion will focus on Art. 28 of the Mexican Constitution and the related competition statute, the recently enacted Federal Law on Economic Competition (F.L.E.C.), which empowered the Federal Competition Commission (F.C.C.), formally regarded as an independent government agency, with exclusive enforcement powers. The legislative history regarding the Mexican antitrust rules is not conclusive as to the goals or objectives of competition law and policy, as is the case with its us, Canadian and European counterparts. Legal precedent in Mexico provides little help, as it is patchy and vague. However, the Mexican case is interesting because Mexico has the oldest written constitutional provision on antitrust, adopted in 1857, more than thirty years before the U.S. Sherman Act and the Canadian Combines Act; and it has one of the most recent competition statutes, in force since 1993. It thereby provides substance for further discussion as to the role of competition policy in a developing economy. Art. 28 of the 1857 Mexican Constitution states the following: There shall be no monopolies or tax corners of any kind, nor prohibitions intended to protect (any) industry. The only exceptions shall be those related to currency issuance, mail and those privileges that, for a limited time, are granted to inventors or perfectionists of any improvement. This can be read to convey a clear efficiency goal: monopolies or privileges are harmful to economic welfare, and therefore should not be allowed, except (i)
2 - Competition Policy Objectives
43
those designed to grant limited incentives for creativity and technology; (ii) mail, which appeared to be a natural monopoly and to raise confidentiality concerns which justified public control; and (iii) currency issuance, which required the backing of a unique central governmental creditor and exclusive regulator.
A. General Prohibitions Aimed at Creating Level Playing Fields Art. 28 of the 1917 Constitution, which is currently in force, is perhaps the most complete constitutional antitrust text in the world. Thefirstparagraph reads as follows: In the United Mexican States, monopolies, monopolistic practices, tax corners and tax exemptions are prohibited, as provided by Statute. The same treatment shall be granted to restrictions allegedly directed to protect industry.
This paragraph contains a sweeping prohibition against anticompetitive practices. Although it does not define the term monopoly, it nonetheless includes a broad prohibition of monopolistic practices. It also broadly prohibits Congress and the executive branch from granting regulatory exceptions, tax advantages, or protective industry shields. Accordingly, this paragraph covers two important competition issues: the need for level playing fields and the potential damage from regulatory intervention which grants privileges. The language of this provision is straightforward; its lack of adjectives confirms a neutral, pro-efficiency stance towards the creation of equal economic opportunity, and an impediment to regulatory capture by industry. Most importantly, this paragraph sets forth the main objectives of competition policy: to combat monopolies and anticompetitive practices, and to create level playing fields for all economic agents, including those in which federal and local governments intervene, directly or through the issuance of regulations.
B. Protection of the Right to Market Entry and the Competition Process The second paragraph of Art. 28 contains the core provisions on competition: In consequence, the law shall severely punish and the authorities will effectively prosecute every concentration of power in one or few hands, especially those
44
Working Paper II
involving consumer goods, when aimed at obtaining price increases; every agreement, procedure or combination of producers, manufacturers, merchants or service entrepreneurs that by any means eliminate free competition among them and force consumers to pay exaggerated prices; and, in general, everything that shall constitute an exclusive and undue advantage in favour of one or several persons which results in damage of the public in general or some social group.
Thus, this paragraph contains a populist goal of combating the greedy merchant or manufacturer that 'hides' products or restricts output to raise prices artificially. Moreover, it prohibits horizontal agreements or combinations that eliminate competition or free market entry, but includes an unnecessary qualification: that consumers are forced to pay exaggerated prices. This qualification is unfortunate because not every collusive arrangement raises prices, and it is impossible to demonstrate what 'exaggerated prices' are. The F.L.E.C. solved the latter problem by including a per se economic presumption that all collusive arrangements, whether involving price fixing or horizontal output restrictions, are anticompetitive and cause an artificial and undesirable effect on prices. It also contains a sweeping catch-all provision: everything that grants an undue advantage shall be unlawful; and undue advantages are against the competitive process, and therefore against the public interest. The latter part of Art. 28's second paragraph provided a constitutional base for the F.L.E.C. to prohibit conduct that creates abusive advantages in the market-place, applying rule of reason analysis. Accordingly, the second paragraph contains the core mandate for an active, pro-efficiency competition framework: a per se prohibition against collusive behaviour and a flexible device, based on the rule of reason, for elimination of practices that are proven to be more anticompetitive than procompetitive. The F.L.E.C. has followed its constitutional mandate. It contains a per se prohibition against horizontal price fixing, market allocation, information exchanges and bid-rigging. Further, it contains a prohibition based on the rule of reason for abusive conduct (vertical price fixing and market allocation, resale price or conditions maintenance, tying, exclusive dealing, boycotts, etc.) by dominant firms in a relevant market, technically determined through standard economic analytical tools. Accordingly, the second paragraph of Art. 28 has allowed the F.L.E.C. to strike a balance between the inevitable stiffness of collusive per se prohibitions and the inevitable vagueness of rule of reason vertical offences. It thereby sought to achieve two goals: legal certainty, by including a category oiper se offences; and economic efficiency, by prohibiting only those practices (and mergers) likely to eliminate competition.
2 - Competition Policy Objectives
45
Thus, the second paragraph of Art. 28 is designed to protect the public interest by disallowing privileges, excessive prices and collusive power. It is complemented by an economic rationale. This provision was written in 1917, and although it was influenced by then existing us antitrust laws (Sherman Act, F.T.C. Act), the crafting of these provision demonstrates original thinking, based on the political, legal and economic objectives of that time. Most of those objectives are still valid today.
C. The Anti-market Provisions ('Prime Necessity Items') The third paragraph of Art. 28 introduces an unfortunate provision allowing Congress to enact laws to plan centrally the national economy through the use of instruments reminiscent of those used during wartime. This provision is at odds not only with basic market principles, but also with the paramount right of every citizen to choose his or her own profession or activity. It thus allows the great Leviathan the possibility of absolute intervention: The statutes will set the bases for fixing the maximum prices of items, raw materials or products that shall be considered necessary for the national economy or popular consumption, and shall impose modalities for the organisation of distribution of such items, materials or products, in order to prevent unnecessary intermediation from causing insufficient supply and price increases. The law shall protect consumers and induce their organisation for the best protection of their interests.
Thus, under this paragraph, Congress will enact statutes establishing maximum prices and 'modalities' for production of items, materials or products of prime necessity for 'the economy' or public consumption, as well as for their distribution, when an 'excessive' middle man intervenes, causing price increases or insufficient supply. It also provides for enactment of a complementary statute for consumer protection purposes. Based on this paragraph, in 1950, Congress enacted the 'Law on Attributions of the Executive Branch on Economic Matters'. This was an overexpansive war-time statute that not only empowered the president to impose maximum prices on almost anything, but also to require what should be produced, how to produce it and how to distribute it. This statute created the basis for a centrally planned economy that formally lasted more than thirty years, creating a culture of arbitrary decisions divorced from market principles and protection of inefficient industries.
46
Working Paper II
This Act was repealed in 1993 by the F.L.E.C, because it imposed a conflict in economic philosophy and because it was probably unconstitutional.4 The F.L.E.C. replaced the price controls of the earlier Act with a provision on how government may impose maximum prices: the president decides which 'basic necessity' products (not services) would be affected; a government agency sets the level of prices such that the cost structure does not lead to 'undersupply'; and the Consumer Protection Agency supervises compliance. The Federal Criminal Code still contains a provision addressing the old populist concern against 'starvers' that hide consumer goods in order to cause an 'artificial' price increase. However, records show that no prison sentences have ever been imposed stemming from such provisions. The F.L.E.C. did not include prison penalties, because judges would be reluctant to impose sentences based on vague antitrust formulas,5 and given the small size of the economy and the underdeveloped nature of the judicial system. The Mexican constitutional antitrust provision does not establish objectives of income redistribution or consumer protection. Rather, income redistribution falls within the ambit of tax revenue and general planning policies; consumer protection is an independent policy; and maximum prices for 'basic necessity' items are restricted to supply, which is what the Constitution expressly protects. Thus, from a regulatory perspective, the Mexican competition system is independent from alien interference stemming from domestic sources.
D. Antitrust Exemptions Several of the remaining paragraphs of Art. 28 exclude certain parties from coverage of the antitrust provisions: exclusive industries run by the State; 4
This statute had not only a 'shy away' effect on investments, but its bad economics impacted on several industries. For instance, canned sardines was a highly purchased product by low-wage workers and their families. Based on this statute, the government imposed a 'maximum' price that proved to be so low that it forced sardines to vacate the market for several years. This caused severe damage to that industry and to low-income consumers. 5 The Argentinean Law on Defence of Competition of 1980, which rests on the broad concept of 'damage to the general economic interest', provides a good example of how prison sentences may have a 'chilling' effect that prevents enforcement authorities from acting. Few, if any, convictions have been obtained on antitrust cases under this statute. See Reyes Oribe and Javier Iraola, Defensa de la Competencia: Sintesis del Sistema Argentino Actual y Proyectos de Reforma, 15 Revista del Derecho Industrial 5 (Ediciones Depalma, Buenos Aires 1993).
2 - Competition Policy Objectives
47
workers' associations; certain associations or cooperatives; inventions; and industrial property. The Constitution also exempts currency issuance and other central bank activities. The Constitution establishes that exclusive areas run by the State are 'strategic areas', including: mail; telegraph; radiotelegraph; oil and related products; 'basic' petrochemicals; radioactive minerals; nuclear energy generation and electricity, and others that Congress may designate in the future. The F.L.E.C. reiterates this provision, but states that the exemptions shall cover only ancillary activities related to these 'restricted areas'. It thereby warns that state enterprises in charge of such areas may not abuse the exemption by leveraging their power upon other markets. The F.C.C. has acted against P.E.M.E.X., the oil monopoly, in cases which confirm the damaging effects of such abuses.6 The F.L.E.C. acknowledges that such areas are legal monopolies, but the market power exerted by state entities as the result of such legal monopolies may not be leveraged to other markets, nor may such entities enjoy an open licence to engage in restrictive practices. This constitutes a departure from the standard 'state action' and primary jurisdiction doctrines of other legal systems. It also confirms a commitment to the goal of obtaining a workable balance between protecting strategic state activities and preserving a competition system with as few negative externalities as possible. The 'export cartels' exemption in the Mexican Constitution is restricted to associations or cooperatives that: cartelise their efforts to sell their products abroad; sell products the production of which is the main activity of their region, and which are not 'basic necessity' items; and are expressly authorised by local legislatures and monitored by federal or state authorities. These limitations make it difficult to qualify for the exemption. The export cartel provision reflects the old 'beggar-thy-neighbour' goal of regulating antitrust matters only for domestic purposes, and disregarding their external effects. However, the F.L.E.C. did not renew this sentiment, as it allows for the design of a fresh international approach with respect to export cartels and extraterritorial enforcement of competition statutes. An urgent need exists to create specific exemptions for horizontal conduct in agriculture and related sectors, as is also the case in the U.S., Canada and the E.U. The F.L.E.C. did not include any such exemption, in order to avoid pressure from other sectors. 6
The F.C.C. ordered, for instance, that P.E.M.E.X. eliminate a territorial restriction that prevented gasoline retailers from operating gas stations in areas already 'covered' by others. See F.C.C, 1993-94 Ann. Rep. 33. The F.C.C. has accepted only a restricted interpretation of such exemptions.
48
Working Paper II
E. Regulated Public Services The tenth paragraph of Art. 28 of the 1917 Constitution provides that the State may grant concessions for the rendering of public services and the use of state property, but it does not explicitly authorise exclusive concessions. It expressly requires that such services not involve anticompetitive concentrations that may contravene the public interest. This important provision confirms another goal of competition law and policy: public services and use of public property shall also be in accordance with competition principles. The F.L.E.C. has reaffirmed this principle, and has required other laws and regulations, such as those governing telecommunications, railroads, roads, ports, electricity and natural gas, to include competition devices in the concession, licensing and pricing of regulated activities. The F.L.E.C. also empowers the F.C.C. to intervene to ensure that this principle is followed.7 This further confirms the departure from the 'state action' and primary jurisdiction doctrines adopted by other legal systems to exclude regulated sectors from antitrust scrutiny.8
F. Broad Economic Objectives in the F.L.E.C. The F.L.E.C. was enacted following two years of study and consultations. Economic and legal policy goals which it incorporates are: to change previous economic policy towards optimal efficiency of markets; to protect the competitive process, not specific competitors; to foster efficiency; to cover all economic agents, including government entities; to limit antitrust exemptions strictly to those included in the Constitution, in order to prevent the creation of political shelters; to rely on analyses of long term market effects; to incorporate a merger policy which is not directed at size; and to promote liberalisation of interstate trade.9 7
These statutes contain cross references to the F.L.E.C. and to the F.C.C, especially regarding tariff regulation and mergers. In particular, all mergers and licence transfers must be pre-cleared by the F.C.C, irrespective of monetary thresholds. 8 The recent U.S. railroad merger of Union Pacific and Southern Pacific illustrates how it could be disastrous for an infant antitrust system, like the Mexican system, to empower sectoral regulatory authorities to rule upon competition issues. There, the U.S. Surface Transportation Board allowed the merger, notwithstanding the strong opposition of the U.S. Antitrust Division (and the Mexican F.C.C). This merger creates the possibility for a single company to control over 90% of the U.S.-Mexico crossborder traffic. 9 See O.E.C.D. Doc. DAFFE/CLP (93) 10, at 2 (1993); Gabriel Castafieda et al., Antecedentes Econdmicos para una Ley Federal de Competencia Economica, 39 El Trimestre Economico 230, 237 (1993).
2 - Competition Policy Objectives
49
The F.L.E.C. constituted a dramatic change of orientation from previous Mexican economic policy. Open trade, privatisation of large concerns, and deregulation of economic activity required enactment of a complementary vigorous competition law. The F.L.E.C. was intended to provide the starting point, with the goal of attaining equal opportunity, elimination of barriers to social mobility and protection of a competitive process that would, in turn, generate a reduction of costs, more and diversified products, better services and a reduction of prices. Consumer welfare was not in itself a goal of the F.L.E.C, but an expected effect of its proper enforcement. The F.L.E.C. was drafted with the goal of preserving the right to enter and stay in the markets, not protecting incumbent economic agents. Thus, few antitrust exemptions were provided, in order to minimise regulatory capture, and state enterprises were included within its reach, with the exception only of those expressly included in the Constitutional 'strategic areas'. Wealth redistribution was viewed as a by-product of proper antitrust enforcement, considering that tax, health and education policies are better equipped to address such social goals. During the process of drafting the bill, an attempt was made to include all possible economic agents within the F.L.E.C.'s reach. Particular emphasis was placed on including state agents, given the potentially distortive effects that can result from direct or indirect state intervention. It was apparent that proper antitrust enforcement in markets held hostage by state power would be nearly impossible, and there would be no incentives to enter such markets. The F.L.E.C. did not include 'crisis', 'structural', 'ecological' or similar competition exemptions in order not to allow them to become the rule. Instead, the new statute had an educational goal of teaching firms to abide by market rules. Gradualist approaches were viewed as hazardous for the new law's proper enforcement, and exemptions would have invited pressure towards a permissive set of rules. Other than the dogmatic per se rule on horizontal breaches, the F.L.E.C. was designed to be flexible. It applies the rule of reason to vertical arrangements. It thus adopts a market analysis approach which focuses more on future market developments with respect to elements such as market entry conditions, product substitutability, barriers to entry, consumer tastes and impact of new technologies. The F.L.E.C.'s approach does not follow populist thinking that big is always anticompetitive. Instead, rule of reason analysis of vertical conduct and proposed mergers take into account actual practice regardless of size. The same is true of predatory behaviour. Discrimination, efficiency and failing firm defences are not provided under the statute, in order not to allow inefficient
50
Working Paper II
firms to use such devices to exert pressure for special treatment. This also protects the new enforcement agency from political lobbying. The F.L.E.C. also addresses the wide array of anticompetitive devices local governments may employ, which threaten the competitive process by erecting various types of barriers to physical entry to their territories. It established a formal conduit for affected parties to seek limited remedies before the F.C.C. Thus, the F.L.E.C. also had a goal of reaffirming principles of economic federalism.
G. The Objectives of Competition Law in Theory There is general agreement that competition policy should strive to achieve economic welfare, but there is no agreement on the definition of economic welfare, how it can be promoted by antitrust policy, and how it should be measured. This debate seems similar to the familiar and unresolved question whether political considerations are more important than economic considerations in determining whether a country is able to function properly. It does not seem possible to arrive at a single definition of economic welfare. In any event, before economic welfare can be achieved, several conditions must be satisfied, some of which are political and some economic. As to the political conditions, elementary notions of economics feed the popular belief that market 'fairness' is a value to be legally protected. Incentives for entrepreneurs may also be a powerful objective of competition law and policy. As to the economic conditions, optimal efficiency may be an acceptable general goal of competition policy, but economic efficiency is a matter of degree. Moreover, achievement of both static and dynamic efficiencies must be considered. Markets are dynamic in nature. Microeconomic modelling is a useful tool for analysis, because it provides an economic framework for arriving at solutions. However, the need for legal certainty is problematic when using these models. For instance, the perfect competition model is only a target (and perhaps an impossible target) of classical antitrust thought. To resolve the static-dynamic debate, perhaps the main goal of competition policy should be viewed as providing incentives to make markets contestable, which is precisely what occurs within the competitive process.10 Market contestability is a matter of degree. The ideal degree of contestability is whatever 10
William Baumol & Janusz A. Ordover, Antitrust: Source of Dynamic and Static Inefficiencies?, in Antitrust, Innovation, and Competitiveness 82 (Thomas M. Jorde & David J. Teece eds. 1992).
2 - Competition Policy Objectives
51
level is established by the technical evidence and reasoning in each individual market or set of interrelated relevant markets in the case being considered. Economic competition also involves: providing opportunities for market entry; promoting of innovation; decentralising power; balancing political considerations; popular myths that generate votes; eliminating red tape needed to keep jobs and investments; promoting national champions; erecting of trade barriers; and so on. Antitrust analysis is not pure, but is one among many policy tools, and one that is always threatened by political power, especially in undeveloped or incipient systems vulnerable by pressure groups. The ideal is to have political aims coincide with optimal market contestability, but the ideal is only a relative and vague aim. Perhaps the reason that competition policy is basically determined by economic content is that populist and other goals are more volatile and difficult to measure than economic criteria, and almost impossible to write into legal instruments. Moreover, notwithstanding their vagueness, economic standards provide more legal certainty than populist and other elements, as well as the possibility of ex post empirical assessment. We must consider the kind of antitrust enforcement needed in developing economies and the extent to which antitrust is properly equipped to counter market power generated by government protection (including trade distorting non-tariff barriers).11 Developing countries are more vulnerable to protectionist government policies and the negative effects of highly concentrated markets than industrialised countries. Therefore, developing countries require more antitrust advocacy and indeed more aggressive competition policy enforcement. As to the trade and competition interface, as trade liberalisation progresses, private anticompetitive behaviour creates new barriers that should be addressed. Thus, international cooperation should be pursued, perhaps in two stages: first, harmonisation of enforcement and basic principles; second, entry of cooperation agreements by competition enforcement authorities. Full harmonisation in the near future appears to be overly idealistic.
Conclusions Mexican constitutional provisions include the following explicit or implied competition policy goals: achievement of level playing fields; preservation of 11
A. Rodriguez & Mark D. Williams, The Effectiveness of Proposed Antitrust Programs for Developing Countries, 19 N.C. J. Int'l L. & Com. Reg. 209 (1994).
52
Working Paper II
the right to market entry; protection of the competitive process rather than competitors; per se prohibition of cartels; rule of reason analysis of restrictions of competition; a strong prohibition of market privileges, whether originating from public or private sources; limited allowance of populist concerns; preservation of the 'public interest' to limit the concentration of power; control or limitation, through statutes, of the State's market intervention; allowance of optimal competition even in regulated services or use of public property; free interstate flow of economic activities; and creation of entrepreneurial incentives. The F.L.E.C. translates pro-competitive principles into more detailed regulatory devices that are standard in developed competition systems. In addition, it includes several native ingredients, such as a strong commitment to antitrust advocacy and against state intervention, whether direct or through regulations. The F.L.E.C. has caused statutes aimed at regulated areas to conform with its principles; it grants competence to the competition authority in such regulated areas, departing from state action and primary jurisdiction doctrines applicable in other systems. Mexican competition law was intended to be an umbrella legislation that would cover wide structural, regulatory and behavioural situations. There is no single goal that could be identified as the paramount value of competition law and policy in the 1917 Mexican Constitution. Instead, it has a complex mixture of broad values. Finally, a workable solution to the theoretical debate may be to regard market contestability as a general principle that leads to other goals. Economic efficiency criteria are better equipped to allow empirical assessment and legal certainty in comparison with other broad competition policy objectives.
Ill Allan Fels Chairman, Australian Competition and Consumer Commission Canberra, Australia
Geoff Edwards Australian Competition and Consumer Commission Melbourne, Australia
Introduction Over the last several decades, the microeconomic paradigm has, with increasing frequency, been considered by academics and competition officials as the source for defining the legitimate goals of competition policy. This is perhaps a narrow view of competition, which has, in many jurisdictions, been increasingly criticised by the business community and the political elite. Many varied positions exist on what should be the legitimate objectives and goals of competition policy. These include: • that competition law should uniquely concern itself with maximising competition among firms, irrespective of other considerations; • that competition is desirable only to the extent that it promotes consumer welfare; • that competition is desirable only to the extent that it maximises total welfare—so that practices which increase producer surplus by more than they reduce consumer surplus should be allowed; • that in a world characterised by widespread underemployment of resources, the legitimate goal of competition law is to foster economic development and, consequently, competition law should not be limited to static efficiency analysis, but should concern itself also with dynamic considerations; and • that efficiency considerations are not the only legitimate goals of competition law, and other considerations, whether political (such as protecting small and middle sized enterprises, ensuring fairness, contributing to the democratic process, etc.) or social (such as maintaining employment, contributing to geographically balanced economic development, etc.) are also legitimate goals of competition policy.
54
Working Paper III
The question of the legitimate goals of competition policy should also be considered in light of the increasing globalisation of world markets. The establishment of a level world playing field in the area of competition presupposes some agreement on what the legitimate goals of such a policy should be.
A. Competition is Desirable in Itself Irrespective of its Economic (Efficiency) Effects The Australian Competition and Consumer Commission (A.C.C.C.) believes the arguments for competition are very strong, and encompass both economic benefits and positive effects on political and social conditions in a country or region. These arguments are set out in this section of the paper. The opposite of competition is monopoly, or a high concentration of market power, the shortcomings of which are generally the converse of the strengths of competition, and therefore need not be separately listed. First, the benefits to both static and dynamic economic efficiency conferred by competitive processes must be acknowledged. A perfectly competitive economy would yield an efficient use of resources in both the allocative and productive senses.1 In contrast, monopoly is generally condemned on efficiency grounds as leading to an inefficient allocation of resources through underproduction. As a result, consumer demands are not satisfied as completely as possible, given the resources available.2 Also, in the absence of competitive pressures, firms may fail to produce with optimal cost (productive) and managerial efficiency. Resources may therefore be wasted further.3 Where anticompetitive conduct or structures may be shown to enhance economic efficiency, an authorisation process is available in Australia, and public interest arguments may similarly provide an exception for firms in the European context. Competition may also better foster dynamic efficiency, also referred to as 'progressiveness'. The forces of competition may provide strong incentives to entrepreneurs to seek and adopt cost-saving technological innovations, or new or improved products.4 '[I]f industry capacity is correctly geared to demand at all times, the only way competitive firms can earn positive economic profits is 1
See Phillip Areeda, John Solow & Herbert Hovenkamp, IIA Antitrust Law (1995), 3-4; Roger Blair & David Kasserman, Antitrust Economics (1985), 21-22; Fred Hilmer, National Competition Policy (1993), 4. 3 See Frederick Scherer & David Ross, Industrial Market Structure and Economic Performance 23 (3rd ed. 1990). 3 Id, at 28-9. 4 Id, at 20-21, 28.
2 - Competition Policy Objectives
55
through innovative superiority.'5 Thus, technological progress is likely to be more rapid, and dynamic efficiency more enhanced, in competitive industries than in sheltered ones. However, as shown below, the opposite has been argued, and has become known as the Schumpeter6 thesis. Second, there are political and social ends of competition, quite aside from efficiency effects, which make the fostering of competitive processes desirable. In the present timeframe, these aims are often neglected, as the focus is mainly on economic efficiency effects. However, it is these political and social arguments, rather than models demonstrating economic efficiency benefits, which have 'tipped the balance of social consensus towards competition'.7 The following is a brief outline of these non-economic effects of competition. • Competitive processes place limitations on the powers of both government bodies and private individuals. The atomistic structure of large numbers of independent buyers and sellers in competitive markets decentralises and disperses both economic and political power.8 Competition can be a means of reducing the concentration of economic power. Under competitive processes, resource allocation and income distribution problems are solved through the interaction of supply and demand forces in the market, not through the conscious exercise of power by private individuals (such as under monopoly), or government (such as under government enterprise or regulation).9 Without atomistic competition, consumers and small business may be vulnerable to unilateral action by big businesses with unassailable market power.10 Competition makes available alternative customers and suppliers, reducing the economic dependency of one agent on another.'' Also, without atomistic competition, a small number of private individuals and firms may have an undemocratic effect on the formation of government policy and even of public opinion.12 Concentrated wealth, concentrated economic power and concentrated control of economic activity are often seen as 5
Id., at 20; see also, Areeda, Solow & Hovenkamp, supra note 1, at 4. Joseph Schumpeter, Capitalism, Socialism and Democracy (1942). 7 Scherer & Ross, supra note 2, at 18. 8 Id., at 18; P.H. Karmel & Maureen Brunt, The Structure of the Australian Economy (1966), 62. 9 See Scherer & Ross, supra note 2, at 18-19. 10 See Karmel & Brunt, supra note 8, at 62. 11 See Maureen Brunt, Legislation in Search of an Objective, in Australian Trade Practices Readings (J.P. Nieuwenhuysen ed. 1970) 239. 12 See Karmel & Brunt, supra note 8, at 62-3; William Shepherd, The Economics of Industrial Organization (3rd ed., 1990), 175. 6
56
Working Paper III
incompatible with democratic government.13 Powerful and influential economic units in society can secure legislative and administrative policies in their own narrow interests, which are very different from the interests of the bulk of the electorate.14 The preservation of, or reversion to, a situation of numerous, individually small and relatively powerless economic units, may be a desirable political end in itself.15 If limiting the power of both government bodies and private individuals to make decisions that shape people's lives is a desirable goal, then the fostering of competitive processes is a means to that end. • Competition laws afford protection to small businesses from the economic power of large firms.16 One may query, however, whether favouring small businesses and affording them protection is a desirable objective of competition law. • Competitive market processes can solve economic problems impersonally, rather than having control of such matters reside with entrepreneurs or bureaucrats. This may be a politically desirable situation, as no identifiable individual or group can then be seen as responsible for market outcomes resulting from the impersonal interplay of competitive market forces.17 • Competition also may provide a more efficient means of economic organisation than central planning, or than monopoly (whichfitssomewhere between the two).18 Competition arguably provides the best means of discovery and utilisation of the enormous knowledge of specific circumstances in the economy held by many different individuals. It thereby co-ordinates the actions of these many individual economic actors. An alternative to this market driven system of organisation is central planning, which requires the gathering and processing of enormous amounts of situation-specific information, initially dispersed among many individuals. It is simply not feasible for all the useful knowledge held by economic actors to be conveyed to a central authority. By its nature, central planning based on statistical information cannot take direct account of the particular circumstances of time, people and place. Competition and market forces allow society to dispense with centralised control. With rapid adaptations to changing circumstances necessary for effi13
See Joe Bain, Industrial Organization 36 (2nd ed. 1968).. Id. at 36. 15 Id. at 37. 16 See Maureen Brunt, Lawyers and Competition Policy, in Australian Lawyers and Social Change 278 (A.D. Hambly & J.L. Goldring eds.). 17 See Scherer & Ross, supra note 2, at 19; Karmel & Brunt, supra note 8, at 62. 18 See generally Frederick Hayek, The Use of Knowledge in Society, 35 American Economic Review 519 (1945). 14
2 - Competition Policy Objectives
57
cient organisation of the economy, ultimate decisions should be left to the individuals who are familiar with those particular circumstances. The problem cannot be solved by communicating all these bits of knowledge to a central body, which issues orders after it has integrated all the information. Individuals need only an effective price system to make good economic decisions. Prices in competitive markets co-ordinate the separate actions of many individual actors, allowing them to direct resources to their most efficient uses. • Competitive markets provide freedom of opportunity. At least when the 'no barriers to entry' condition of perfect competition is satisfied, individuals are free to choose their preferred trade or profession, limited only by their own talent, skill and ability to raise capital.19 Hence, a competitive society provides opportunities for new firms, new talent and new ways of doing things.20 Big businesses, if allowed to benefit from an absence of competition laws, may take advantage of their position as market leaders to establish a position of impenetrable strength, raising barriers to entry against other domestic firms.21 • Competition is associated with the development of initiative, self-reliance and flexibility of mind.22 Competitive firms are also more likely to be able rapidly to adjust to unforeseen changes in their environment.23 • Competition can lead to a greater freedom of choice, by creating a range of alternatives and variety for consumers, workers and firms.24 • Competition arguably promotes equal treatment of workers according to race and sex, as competitive firms cannot afford to discriminate.25 However, competition may also more rapidly marginalise less able workers. • Fostering competition and avoiding concentration in industries also arguably limits wealth transfers from consumers to capital owners.26 However, competition policy will often benefit capital interests. • Competitive markets also arguably foster economic security, that is, the steadiness of employment. Tight oligopoly can increase the severity of recessions.27 The costs of this include: more widespread unemployment; greater 19
See Scherer & Ross, supra note 2, at 19. See Brunt, supra note 11, at 239. 21 See Karmel & Brunt, supra note 8, at 64-5. 22 See Brunt, supra note 11, at 239. 23 See Hilmer, supra note 1, at 4. 24 See Brunt, supra note 11, at 239; Shepherd, supra note 12, at 164. 25 See Hilmer, supra note 1, at 5; Shepherd, supra note 12, at 172-5; K e n n e t h Elzinga, The Goals of Antitrust: Other than Competition and Efficiency, What Else Counts?, 125 U. Penn. L. Rev. 1191, 1202 (1977). 26 See Areeda, Solow & H o v e n k a m p , supra note 1, at 2; Shepherd, supra n o t e 12, a t 20
166-7. 27
See Shepherd, supra note 12, at 176.
58
Working Paper HI
government expenditure for unemployment compensation; and a greater level of general insecurity in the community. • Competition has the capacity to dampen price inflation, generally regarded as an economic harm. • Competition can also foster community beliefs in the virtues of diversity, tolerance, flexibility, individual initiative, the atomisation of power and freedoms of opportunity and choice. Thus, there are strong political and social arguments for competition policy, quite aside from the contribution of competitive markets to economic efficiency. Political factors are thus not necessarily antithetical to competition policy. They may even be complementary.
B. Political Aspects of Competition Policy Section A has outlined legitimate rationales for competition policy other than economic efficiency. Further, in implementing competition policy, political considerations often override economic efficiency considerations. The adoption of a competition law is a political act, and, as such, political considerations should be paramount. However, this does not imply that political arguments are necessarily against the fostering of competitive processes. As shown in Section A, there are many convincing political arguments in favour of effective competition law. Further, competition policy may be politically unpopular, as it disturbs established economic interests (property rights) and affects the distribution of wealth. A tradeoff must often be made between the efficiency and equity effects of economic policies.28 Policies that offer only minor efficiency gains and have undesirable effects on equity should tend not to be implemented. However, competition policy has a sound efficiency justification, and other tools of social policy, such as taxation and government spending, may provide more appropriate and directed solutions to distribution problems. Many of the 'political' challenges made to competition law and policy are not, in reality, based on genuine public interest concerns. Rather, they are challenges advanced by sectional groups with political clout, advanced in their own narrow sectional interests, rather than in the true public interest.
28
See generally Arthur Okun, Equality and Efficiency: The Big Tradeoff (1975).
2 - Competition Policy Objectives
59
C. Demand Side View of Efficiency The A.C.C.C. endorses a concept of efficiency that is much broader than just supply side cost minimisation. Efficiency also has demand side elements, which are sometimes overlooked. A truly competitive market would offer consumers a good choice of both price and product. Greater choice allows for more exact matching of consumer wants and enhanced allocative efficiency. Consumers not only want a good price for a product, they also want the possibility to choose from a range of quality, service, credit, etc. Competition should be encouraged in all these dimensions. For instance, when competition in the medical profession is suggested, fears are expressed that price-cutting doctors will emerge, offering below acceptable standards of service. Two arguments can be made in response. First, in the market for medical services, quality is not negotiable for consumers. A competitive health market should instead deliver improved quality, as what consumers want when they visit a doctor is, above all, quality. There may still be a range of choices along the price/quality plane, but this range should be in the high quality end. Second, it may be incorrect to assume that suppliers who cut price necessarily offer lower quality. Price cutters are often more efficient or technologically progressive, and therefore able to provide good quality at lower prices.
D. Static and Dynamic Efficiencies If economic efficiency is a legitimate objective of competition laws, the question arises of how we should handle the possible tradeoff between the static efficient allocation of resources and economic development. If competitive processes offered the best prospects for innovation and dynamic efficiencies, no tradeoff would be necessary, because fostering competition would be the best policy to enhance both static and dynamic efficiency. Only if dynamic efficiency is better served under uncompetitive market structures does the question of trading off static for dynamic efficiencies arise. Accordingly, we must first ask whether dynamic efficiency is better served under competitive or uncompetitive market structures. If dynamic efficiency is better served under uncompetitive market structures, we must then consider what value should be placed on dynamic, as opposed to static, efficiency. How important for future incomes is the rate of technological progress? Technological progress can contribute significantly to consumer welfare, not only because it leads to new and better
60
Working Paper III
products more accurately reflecting consumer tastes, and existing products at lower cost, but also because it increases the productive base of the economy, creating new industries and new jobs.29 In the long run, technological progress is likely to contribute far more to consumer welfare than the elimination of allocative inefficiencies is likely to do.30 It is the rate of technological progress, not the efficiency of resource allocation at any one moment in time, that in the long run determines whether real incomes will be high or low.31 The Schumpeter thesis holds that competition is inferior to concentrated scenarios in terms of fostering innovation and dynamic efficiency.32 First, Schumpeter argued that firms need supranormal profits, or at least the expectation of future supranormal profits, in order to have the funds, or expect sufficient returns, to invest in research and development.33 Thus, in a perfectly competitive market, there would be inadequate innovative activity. This is so not only because research and development is a high risk activity, but also because in a world of perfect knowledge, the full value of an innovation could not be captured by the innovator, since once disclosed, it could be used by others at little or no cost.34 The imitators' prices would not reflect the development costs, and competition would prevent the innovator from recovering its investment in the innovation. Firms need some protection from competition before they will bear the risks and costs of invention and innovation. Only within big sheltered units may long term views towards investment be possible.35 Therefore, Schumpeter argued that monopolies are necessary to provide the funds or the incentives for investment towards innovation. Second, research and development often require large scale, which may only be possible within firms of large size. These firms are more likely to exist in monopolistic markets than in atomistic competitive markets. Thus, the thesis of Schumpeter suggests that progressiveness may be inconsistent with static efficiency, and the competitive market structures required by the latter may impede the former. Several arguments refute the Schumpeter thesis. First, would a monopoly, with no threat of rivalry in the near future, need to spend on innovation, and bother to do so? Monopolists, secure from the threat of competition, may dis29
See Hilmer, supra n o t e 1, a t 4. See Scherer & Ross, supra n o t e 2, at 3 1 , 613-60. 31 See Scherer & Ross, supra n o t e 2, at 31, 613. 32 See Schumpeter, supra n o t e 6, at 106. 33 See id.; R i c h a r d Caves et al., Australian Industry: Structure, Conduct, Performance 118 (2nd ed., 1987). 34 See Areeda, Solow & H o v e n k a m p , supra note 1, at 29-30. 35 See Schumpeter, supra note 6; Areeda, Solow & Hovenkamp, supra note 1, at 32-3. 30
2 - Competition Policy Objectives
61
play a 'lethargic attitude towards technological innovation'.36 Second, would not a competitive market structure better foster innovation investment? The forces of competition give strong incentives to entrepreneurs to seek and adopt cost-saving technological innovations, or new or improved products.37 Thus, technological progress may be more rapid, and dynamic efficiency more enhanced, in competitive industries than in sheltered ones. Would notfirmsbe more likely to seek innovations under competitive pressures, especially if effective patent laws offer innovative firms sufficient rewards from innovation? Third, firms in competitive markets may be well able to achieve the necessary scale for research, by combining with other firms in joint ventures, or outsourcing research and development programs to dedicated researchfirms.(See below.) Areeda, Solow and Hovenkamp suggest that oligopolistic structures may best facilitate innovative activity.38 Supranormal profits earned through oligopoly may provide access to sufficient funds for research and development. Unlike monopolists, oligopolists are vulnerable to the innovations of existing rivals, providing greater incentives to themselves to innovate.
1. Evidence Scherer and Rossfindstatistical support for the proposition that greater rivalry (approximated through lower concentration indices) invigorates research and development investment up to a point. However, an excessively atomistic market structure discourages research and development investment. This is because the share of the benefits from such investment that any one firm in such an industry could appropriate would be too small to confer net gains.39 They also argue that new entrants, without any commitment to established technologies, have been responsible for a very substantial share of revolutionary new industrial products and processes.40 Indeed, actual and potential entry appear to be important, both as a source of innovative products and processes, and as a stimulus for existing firms to innovate.41 36
Scherer & Ross, supra note 2, at 28; see also Areeda, Solow & H o v e n k a m p , supra note 1, at 34. 37 See Scherer & Ross, supra note 2, at 2 0 - 1 , 28; Caves et al., supra n o t e 3 3 , a t 119; Hilmer, supra note 1, at 4. 38 See Areeda, Solow & H o v e n k a m p , supra note 1, at 34. 39 See Scherer & Ross, supra note 2, at 637, 646. 40 See id., at 653-4; Caves et al., supra note 33, at 123-4. 41 See Scherer & Ross, supra note 2, at 653-4; Areeda, Solow & H o v e n k a m p , supra note 1, at 35; Caves et al., supra note 33, at 123-4.
62
Working Paper HI
2. The Significance of Size on Incentives to Innovate Larger companies, which are more likely to exist under less competitive market structures, have several advantages in research and development.42 First, their size allows them to exploit scale economies, which are often necessary for research and development projects.43 Second, larger companies have the ability to maintain a diversified portfolio of research and development projects, enabling them to hedge their risks against any particular project failing.44 Third, larger entities are generally able to attract more capital,45 and at lower cost, making them better able to finance research and development projects.46 However, one disadvantage of large size is that small firms may be more adept at risk taking than large firms. Scherer and Ross suggest that there is a bias against really imaginative innovations in large firms.47 Ideas must be approved by several layers of management, which increases the likelihood that they will not be approved. The inability to get ideas approved by higher management may thereby drive creative individuals out of large organisations. Also, firms with little or no market share have greater incentives to force the innovative pace, where the potential of gaining first-mover advantage and substantial market share exists.48 Most importantly, small firms may be able to achieve the necessary scale for research that only large firms might otherwise possess, by combining with other firms in joint ventures, or outsourcing research and development programs to dedicated research organisations. Both large and smallfirmsthen have certain advantages and disadvantages in innovation. From a study of empirical evidence, Scherer and Ross conclude that the superiority of one over the other varies with the nature or stage of the particular innovative activity.49 Thus, there is a place for firms of all sizes in technological innovation. 'Technological innovation thrives best in an environment that nurtures a diversity of sizes'.50 42
See Scherer & Ross, supra note 2, at 652. See Scherer & R o s s , supra note 2, at 652; A r e e d a , S o l o w & H o v e n k a m p , supra note 1, at 31. 44 See Scherer & Ross, supra note 2, at 652; Areeda, Solow & Hovenkamp, supra note 1, at 31. 45 See Areeda, Solow & H o v e n k a m p , supra n o t e 1, a t 30. 46 See Scherer & Ross, supra note 2, a t 652; A r e e d a , S o l o w & H o v e n k a m p , supra note 1, at 3 0 - 1 . 47 See Scherer & R o s s , supra note 2, at 6 5 2 - 3 . 48 See id., at 651. 49 See id., at 651-4. 50 Id, at 654. 43
2 - Competition Policy Objectives
63
There appears to be no conclusive answer on the accuracy of the Schumpeter thesis. There are as many arguments for as against the view that monopolies are best for dynamic efficiency. Certainly, a satisfactory rate of innovative activity requires significant departures from the model of perfect competition. The development of patent laws recognises this. Only a situation between monopoly and perfect competition could be optimal for progressiveness.51 Scherer and Ross suggest that a blend of monopoly and competitive processes, with greater emphasis in general on the latter, would be ideal.52 Also, a blend of large and small firms seems desirable.53 Our view at the A.C.C.C. is basically agnostic, but slightly favours competition as the better means of encouraging innovation and achieving enhanced dynamic efficiency.
E. Use of Competition Law to Achieve Economic Results which are Not Strictly Related to the (Narrowly Defined) Concept of Economic Efficiency Generally, competition law should not be used directly to achieve economic objectives which are not strictly economic efficiency objectives, such as preventing price increases, maintaining employment levels, etc. That is, competition law should not be distorted to achieve these objectives. Indirectly, these other economic objectives should be aided by an effective competition law. For example, competitive processes should lead to greater price restraint than uncompetitive structures and conduct. Moreover, an improved employment situation should be another result of fostering competition, as lower prices should lead to increases in demand, increases in firm outputs and corresponding increases in jobs. The assumption that competition is bad for jobs is clearly suspect in theory. In the long run, greater competition can only enhance employment opportunities. F. The Integration of National Markets is a Legitimate Economic Objective of European Union Competition Law, with Political Benefits The integration of national markets is a legitimate objective of competition law, both from an economic perspective and from a political perspective. It is 51 52 53
See Caves et al., supra note 33, at 130. See Scherer & Ross, supra note 2, at 660. See Caves et al., supra note 33, at 122.
64
Working Paper III
a legitimate economic objective of European Union competition law if the E.U. wants a more efficient overall economy and greater production of goods and services. Integration of national markets may also have political benefits for the European Union and its members. The integration of markets within the E.U. will also have worldwide benefits, as long as it is not accompanied by new or increased restrictions on trade between Member States and third countries. Raising barriers to trade between an economic union and non-union countries is a dangerous potential side effect of integration, as internal pressure will arise to restrict competitive pressures that emanate from outside the union. Politicians within the union may easily bow to such pressure, given that the main losers from such restrictions are foreigners holding no voting power.
G. Competition Laws and Policies in Different Countries should be Broadly Similar While their forms may differ, competition laws and policies across countries should be broadly similar. Indeed, a convergence can be detected around the world. Competition laws and policies in different countries may vary, not as the result of different views about economics, but as the result of different cultural inheritances. Between Europe and the United States, differences in cultural, moral and political history have resulted in 'differences in attitudes toward private economic power,54 the grant of broad discretionary authority to governmental bodies,55 the place of morality in economic affairs56 . . . and the emphasis to be placed upon distributive justice as compared with enhanced productivity'.57 European restrictive practices legislation reflects little distrust of concentrated business power.58 Further, where controls are placed on monopolies and cartels, broad discretion is usually entrusted to government officials.59 The acceptance of such private and bureaucratic power contrasts sharply with the distrust in the United States of both powerful business and powerful government.60 54
See C o r w i n E d w a r d s , C o n t r o l of Cartels and Monopolies: An International C o m p a r i s o n (1967), 22. 55 See E d w a r d s , supra note 54, at 22. 56 See id., a t 2 1 . 57 See id. a t 15, 2 1 - 2 . 58 59 60
Id. at 16. Id. Id.
2 - Competition Policy Objectives
65
Different attitudes to the role of government in Europe and the United States have therefore conditioned their law and policy.61 United States antitrust law was born out of hostility towards big business, with a presumption that big is necessarily bad. In contrast, in Europe, there is a more neutral attitude towards big business, with no such presumption.62 Monopoly or large size are not, in themselves, seen as undesirable. Rather, each firm and industry is judged on its behaviour and performance. Rather than opposing market power as such, only the uses to which monopoly power may be put are controlled.63 United States antitrust policy, in contrast, tends more towards an attack on positions of market power as such, and on conduct byfirmsdirected at acquiring and maintaining that power.64 For instance, Arts. 85 and 86 of the Treaty of Rome are concerned primarily with the protection of consumers and of a dominant firm's purchasers and suppliers. Neither Article prohibits highly concentrated market structures, nor conduct aimed at retaining such structures. Thus, the main concern is not the survival of the competitive process, and it is not the acquisition and retention of monopoly power, but rather the abusive exploitation of existing power.65 The consequence of such differences in attitude is that even if similar views are held among nations as to the economics of competition, different views will exist as to the appropriate form of implementation of competition law. 61 62
See id., at 15-23.
See Joliet, Monopolization a n d Abuse of D o m i n a n t Position: A Comparative Study of the American a n d European Approaches to the Control of Economic Power (1970), 8-16. 63 See id., a t 8-9; Edwards, supra note 54, at 22. 64 See Rene Joliet, supra note 62, at 9 - 1 1 . 65 See id., at 11.
IV Anna Fornalczyk Vice President, Competition Development Center Ltd. Professor Lodz, Poland
Introduction This paper discusses legitimate and desirable competition policy objectives in transition countries. Development of competitive societies in transition countries is a substantial challenge for proponents of democracy and a free market economy. After many decades of operating under a communist system without political and economic freedom, it is difficult for individuals to adapt to the new requirements resulting from democratic and market-friendly reforms in the Central and Eastern Europe countries (C.E.E.C.). Following efforts of reformers, the public understanding of the advantages of the new political and economic order gradually increases. However, in the early years of reform, many are disappointed with the negative effects of political and economic liberalisation. Freedom creates difficulties because it requires individuals to take responsibility for their own lives. Under the communist system, an individual would work in the same enterprise and perform the same functions for his or her entire working life. A society whose mentality has been formed by such living conditions is not prepared to handle the demands of democracy and a free market economy. Nevertheless, it is impossible to return to the communist system. Reforms, therefore, must continue. From an economic perspective, liberalisation of prices and trade resulted in a radical change of income distribution. Social groups that were in the best position under the previous distribution structure are now worse off. Often, such groups feel that they are victims of re-allocation of resources under pressure of consumers' demands demonstrated in a market. Many enterprises face bankruptcy due to the pressure of market forces. Unemployment is the painful result of structural adaptation in emerging market economies. In the early years of economic reforms, some individuals profit dishonestly, which harms public opinion of business culture under normal market conditions. Thus, in transition countries, there are many reasons that promotion of competition is very important but controversial and difficult to achieve. The reforms needed in the C.E.E.C. may be compared to a surgical operation that is necessary and will be painful, and therefore should be completed as quickly
68
Working Paper IV
as possible. If the reforms are postponed, their negative effects will prevail over the positive ones, and more individuals will lose the chance to achieve a better standard of living. The promoters of competition must clarify all of these controversies for the public, and undertake to increase public understanding of the advantages of market reform. These reforms must be supported by the majority, which may exercise its democratic rights in favour of or against politicians engaged in transformation from command to market economy. Populists point out the negative effects of reform in their fight against reformers. This fight for political power may substantially impede the progress of the transition process. Accordingly, in transition countries, competition policy plays a far more important role in promoting the activity of market forces than it does in mature economies. In emerging markets, it is important to build a commitment of society to competition in all areas of public activity. At the beginning of market reform, competition seems like a strange and undesirable phenomenon. Many individuals view it as the root of difficulties which they face. Promotion of competition should be pursued by consistently enforcing competition law, encouraging state authorities to consider competition when taking their decisions, and advocating competition to society. Each of these activities is an important element of bringing about progressive change in the mentality of the people and their behaviour as consumers and producers, as well as improvement in the efficiency of the economy. Competition policy should therefore be understood in a broader sense than competition law enforcement. Given that all transition countries inherited a monopolised industrial structure from the command economy, competition policy must cover many areas of economic activity. It is closely tied to the main economic tasks of reform, which are: decreasing the rate of inflation and the budget deficit, improving the convertability of the domestic currency, liberalising pricing and trade, restructuring and privatising, and developing the financial sector. Each of these tasks is directly or indirectly related to the concerns of competition policy. However, the tradeoff to be made between pursuit of competition policy and the aforesaid tasks is not as simple as it may seem. The remainder of this paper will address some questions and dilemmas of competition policy in transition countries.
A. Legitimate Objectives of Competition Policy in Transition Countries Since the inception of market-oriented reforms in transition countries, reformers have viewed competition law as a necessary element of a successful process
2 - Competition Policy Objectives
69
of transformation. Competition policy consists not only of the enforcement of competition law. In addition, the power of enforcement authorities to bring competition cases provides them with leverage over other government departments to consider competition issues with respect to economic policy and decisions concerning privatisation and restructuring of enterprises and entire sectors of the economy. In general, the legitimate objectives of competition policy in transition countries are the same as the objectives well known in Western countries. However, under the laws of some of the Commonwealth of Independent States (C.I.S.), the competition agency, which is a state administrative body, has the power to take action against anticompetitive decisions of the state administration or local authorities. The reason for granting this power appears to be the lack of credibility of the commitment of public authorities to competition, following decades of fighting against the freedom of individuals. Old communist customs are still alive in the former Soviet Union. Under the communist regime, government economic policy was formulated based upon numerous compromises between various social, business and political interest groups. Such differing interests also exist in mature market economies and mature democracies. Differences may be overcome through the use of a democratic institutional framework and the willingness of people to compromise on problems that arise in all areas of society's political and economic activity. One goal of reform in transition countries is to construct an effective democratic institutional framework and to enable individuals to develop negotiating skills. However, the legitimate power of competition agencies to take action against the anticompetitive acts of other state bodies may not be workable in the long run, because it will lead to legal conflicts within state administration. The following areas are generally covered by competition law in transition countries: abuses of dominant position, anticompetitive agreements, merger control, and, in some countries, unfair competition. The majority of these countries have merger provisions that extend to privatisation of state owned enterprises and so-called mass privatisation projects. Thus far, the experience with enforcement of competition laws in transition countries has been relatively poor. Some of the main pitfalls and questions concerning promotion of competition in emerging markets have become apparent.
1. Pitfalls of Regulation As stated above, transition countries inherited monopolised markets from their command economies. Thus, competition agencies were originally flooded
70
Working Paper IV
with complaints about abusive behaviour of incumbents in numerous product and geographic markets. The public did not understand the concerns of competition law, and thus complaints often did not take into account the market position of the firms in question. The agency was thereby forced to spend time on inquiries even when thefirmsdid not have a dominant position in a relevant market. In addition, price liberalisation in the early years of reform resulted in an increase in the inflation rate. This placed enormous pressure on the competition enforcement agency to mitigate the price-making policy of monopolies. In the short run, many of even the most liberal reformers were interested in the agency's price control function. As a result, competition agencies were broadly involved in regulation of the behaviour of enterprises, which was at odds with the proscriptive character of competition law. The law establishes what actions are prohibited from the perspective of development of competition, but it does not address how market operators should run their business. Enforcement of the law to control the pricing policies of monopolies placed the agency in the role of administrative price-maker, a role well known under the command economy. Moreover, it is especially difficult to implement a test of production costs in transition countries, since prices of almost all commodities in a market are in a state of dynamic change. The majority of competition agencies in these countries were able to avoid taking on the role of price-maker or to give priority to price control. Instead, they focused on counteracting market foreclosures, which involved them in various aspects of government economic policy.
2. Abusive Conduct of Bankruptcies The next dilemma concerns abusive conduct of enterprises that face bankruptcy. Often, an enterprise's profits follow from its monopolistic position and price-making policy. While the enterprise is large and politically powerful, enforcement of the competition law against it is difficult, notwithstanding the public support for it. From the perspective of consumer interest, monopoly profits are not justified, but they allow the inefficient enterprise to survive. The bankruptcy of such a large enterprise may cause significant unemployment and social unrest, which is often stimulated by political opposition and trade unions. Since the number of such enterprises is considerable in a transition country, this problem may substantially impede progressive changes in the economy. The problem concerns declining industries, which should be the objects of the government economic policy rather than the targets of competition law enforcement.
2 - Competition Policy Objectives
71
3. Consumer Test and Efficiency Test A contradiction between the results of implementation of the consumer interest test and the efficiency test arises in many areas of competition law enforcement. Such contradiction is characteristic not only for transition countries; competition agencies in mature market economies also deal with it. However, in transition economies, the number of dominant and inefficient enterprises is much larger, and an insufficient level of financial and human resources has been devoted to solving the problem. For instance, the level of foreign capitalflowinginto an emerging market is related to the degree of concentration in that market. The evidence suggests that investors are not interested in investing in competitive markets. Rather, they are willing to pay higher prices for productive assets held by monopolies than for those held byfirmsin competitive markets. Asset owners are interested in receiving the highest price possible from investors. Most assets in transition countries are owned by the State. The lack of domestic capital, modern technology and qualified management imposes pressure on the government to sell the assets under conditions required by the investor.
B. Methodological Problems Competition law enforcement requires comprehensive market analysis, both to assess an enterprise's market position and to analyse the substance of its behaviour. It is difficult to apply the law in transition countries because determinations must be made which are difficult to make without rich precedent. Industrial organisation theory and experience gained by Western countries are useful for a rapid development of knowledge on methodological problems, but some sophisticated methods and interpretations are difficult to apply in emerging markets. For instance, how should the notion of 'substantial restriction of competition' be interpreted? What should be the balance between arguments related to existing market structure versus those related to potential competition? How much weight should be given to 'soft' information collected from consumers when attempting to define the relevant market? Finally, how should social reasons for market foreclosure be weighed in the assessment? In transition countries, the answers to all of these questions are more complicated than in Western countries as a result of characteristics of the transformation process. From a methodological point of view, market definition requires much detailed information about both the supply and demand characteristics of the relevant market. The supply-side analysis appears easier than
72
Working Paper IV
the demand-side analysis, unless problems exist with gathering information on imports to the relevant market. Trade liberalisation in transition countries, huge numbers of very small firms, and unreliable statistical sources distort information on imports. In addition, information on the substitutability of products and methods of production are difficult to obtain from producers who are not accustomed to such analysis. Thus, market definition is based more on existing industry structure than market structure. The one-sided definition of the relevant market may distort the substance of the inquiry. Demand-side analysis of markets is not well developed in transition countries for two main reasons. First, competition agencies are not able to perform time-consuming and expensive consumer surveys, and there are few experts outside the agency that are specialised in such research. Random information collected sporadically does not constitute convincing evidence in formal cases. Tests developed in Western countries may not be appropriate for emerging markets. For instance, the 5 per cent test, which is proposed in the American Merger Guidelines, is not appropriate for emerging markets, which have a high and volatile rate of inflation. The social impetus for market foreclosures concerns the role of the competition agency as an advocate of economic policy, rather than enforcer of competition law. Competition law should be applied to promote what has legally been defined as the public interest. Other aspects of the public interest must be established by state officials who are responsible for various elements of government economic policy. Competition concerns should be taken into account in such deliberations. It is the responsibility of the competition agency to provide the information needed to allow competition concerns to be properly considered in such discussions. In general, the application of competition policy in transition countries must take into account the dynamic changes which are occurring in their economies. The results of change are difficult to predict, which impedes the ability of the competition agency to appreciate potential competition or to apply an efficiency test. Accordingly, it is desirable for the competition agency to be involved in policy development by the government. Involvement in policy development may provide a useful source of information for the competition agency, as well as a good opportunity for it to advocate competition concerns at the government level. A tradeoff must be made between effective fulfilment of legitimate competition policy objectives and other objectives of government policy. For instance, a domestic dominant producer may be protected from foreign competition by tariffs and quotas established by the government. Such protection has the effect of strengthening the firm's dominant position in the domestic
2 - Competition Policy Objectives
73
market. Thus, it would be useless for the competition agency to counteract monopolistic practices of the protected dominant firm. There are many examples of market foreclosures in transition countries which both impede development of competition and support inefficient producers facing bankruptcy.
C. Desirable Competition Policy Objectives Competition policy in transition countries should cover many areas of public activity. Some aspects of the transition process should be covered by competition policy, such as: restructuring of enterprises and sectors; regulated sectors and activities; state aid monitoring; and competition advocacy in public.
1. Restructuring of Enterprises Restructuring of enterprises and entire sectors of the economy causes a reallocation of productive resources according to demand requirements. The reallocation may occur naturally as the result of market forces. It leads to many social and political difficulties. In the case of a declining industry, the government must be involved in liquidation of over-capacity, closing down enterprises because of environmental pollution, and decreasing unemployment by creating new jobs and financial incentives for unemployed individuals who would like to set up new businesses. Restructuring also may involve breaking up large enterprises into small, viable units that compete with each other in a market. It is the responsibility of the competition agency to promote such procompetitive restructuring of firms. Demonopolisation of the economy is highly controversial. In transition countries, many critics argue that such activity lessens competitiveness of the entire economy in the world market. Big business in Western countries is given as the argument against breaking up domestic giants. Many large enterprises in transition countries were set up to meet all of the domestic economy's requirements. The highly monopolised industrial structure facilitated the work of central planners to control the activities of enterprises. However, the level of concentration did not result from economic forces. The decision to break up these dominant firms will hopefully result in creation of a competitive market that will be more efficient, to the benefit of consumers and the economy as a whole. Demonopolisation must be implemented with a long-run perspective. The divestiture of enterprises and the prevention of anti-competitive mergers must be accomplished taking into account existing and prospective market
74
Working Paper IV
structure. The competition agency must balance the effects of implementation under both the consumer interest test and efficiency test. The outcome of the efficiency test is often unpredictable; therefore, the consumer interest test should have priority. Moreover, in transforming countries, consumer interests are not represented as forcefully as are the interests of the business community, which also supports giving priority to the consumer interest test. Accordingly, the competition agency should represent the interests of consumers. The strategy pursued to accomplish demonopolisation must recognise that dynamic changes in many emerging markets provide an impetus for mergers and acquisitions, which are covered by the competition rules. Trade liberalisation also allows domestic producers to become sellers in the world market, which radically alters their market position. This should be recognised by the competition agency in making its analysis. Development of competition in a domestic market prepares domestic producers for competition with foreign firms. This should be the main argument used by the competition agency to persuade the government not to intervene on behalf of inefficient large producers. In addition, the agency should argue that such intervention, including state aid granted to unprofitable market operators, will lead to consumer losses. The large number of inefficient enterprises and sectors in transition countries, as well as the political aspects of solutions to this problem, should make the competition agencies sceptical of crisis cartels and reluctant to grant them exemptions from the competition laws. Historically, the governments of these countries did not employ market-friendly regulation of the economy. These countries also suffer political instability, which results in frequent changes of governments. Such an environment does not secure the continuity of restructuring projects commenced by the current government. The considerable resistance to implementation of restructuring projects is apparent when a new government takes control, as it will postpone or eliminate projects commenced by the previous government in order to encourage social stability. If enterprises which are being restructured are exempted from competition law enforcement, it will be difficult to eliminate such exemption at a later time.
2. Regulated Sectors Restructuring of enterprises is often carried out jointly with substantial deregulation of certain sectors, as well as privatisation of state owned assets. In market economies all over the world, some sectors and economic activities are regulated, because market failures would otherwise occur. However, regula-
2 - Competition Policy Objectives
75
tion should be opened to competition concerns as much as possible. Promotion of pro-competitive changes in highly regulated sectors seems to be significant, desirable objective of competition policy. Regulated sectors often involve public utilities which are traditionally captured by strict administrative control in transition countries. In the process of deregulation, the competition agency should be involved in preparatory work, with respect to both restructuring projects and legal rules through which deregulation will be effected. The main task of the competition agency should include promotion of third party access and transparent standards of market access. Since assets in public utilities sectors are owned by the State, the agency should also be involved in establishment of regulatory bodies independent of state administration, and monitoring the fairness of the conduct of these bodies with respect to new entrants. The evidence shows that regulation by the owner of the service provider has protected the incumbent against new entrants. The best solution would be to privatise service providers, but in transition countries, the lack of capital and the conservative approach to public utilities make the changes of ownership impossible in the current timeframe.
3. State Aid The weak economic state of public enterprises is a significant impetus for the government to support them by financial and administrative measures. The same measures may be applied to restructuring and privatisation projects of state owned enterprises to provide incentives to investors. This group of government activities is called state aid; all types are detrimental to the development of competition, which requires equal access to markets and equal conditions for market operators. State aid may be granted to enterprises through subsidies, public procurement (where the price paid by the government is higher than the market price), mitigation of charges which must be paid under normal conditions (such as taxes and social security payments), preferential credits, credit guarantees and capital injections. Whatever the form of state aid, it benefits the firms that receive it and discriminates against their competitors. The rationale for state aid is provision of temporary support to the enterprises, sectors or regions which allows them to survive economic problems. State aid exists in market economies all over the world. In the European Union, state aid is compatible with the common market and accepted by the Commission under certain conditions. Even the most free market-oriented
76
Working Paper IV
countries, like the United States, make public expenditures. A major difference exists between industrial and transforming countries with respect to their general approach to state aid. In industrial countries with mature market economies, state aid is treated as an exception from the normal rule that market forces govern, since the state-owned sector is small. In contrast, the stateowned sector is significantly larger in transforming countries, and state-owned enterprises treat state aid as a continuation of the rules which governed in the command economy. In addition, state aid is granted mainly to ensure the survival of inefficient, large, state-owned enterprises. The scarce financial resources used for state aid impede progressive, market-oriented changes in these misdeveloped economies. In the centrally planned economies, priority was given to the development of intermediate commodities, which distorted the entire structure of the economy in favour of heavy industries. Public spending to maintain an existing structure may secure social rest. However, in the long run, state aid may increase the budget deficit and the inflation rate, both of which are relevant to stabilisation of these emerging markets. Particularly in transition countries, state aid which delays structural changes necessary to render enterprises or sectors economically viable, and which allows inefficient market operators artificially to survive, should be avoided. The public should understand that long run negative effects are likely to outweigh short run benefits of maintaining a certain level of employment. Accordingly, a state aid monitoring system is a desirable element of competition policy. The rationale for including state aid in competition policy is that a competition agency is not substantially interested in granting aid as other government bodies may be, and that the agency is responsible for securing equal conditions for all market operators, including both incumbents and new entrants. State aid cannot be eliminated completely, because it is a significant element of government economic policy, especially in misdeveloped transition countries. However, state aid can be made transparent and discussed in public, so that taxpayers know how much of their money is spent for noneconomic reasons. Finally, the effectiveness of such expenditures in accomplishing progressive and market-oriented changes in the economy should also be known. Transparency of state aid is also required by the European agreements signed by the majority of Central European countries. In this way, the association process may support market reforms in these countries.
2 - Competition Policy Objectives
11
4. Competition Advocacy The last desirable objective of competition policy is advocacy of competition concerns to society. Given the relatively low level of public awareness about competition issues, this is a significant task for the competition agency. This advocacy can be performed formally or informally. Promotion of competition will support progress in the transition process. It is especially important to educate and inform the business community, which must adapt its activity to new market requirements. Advocacy may be effected through the following activities: publication in newspapers of decisions on competition matters, preparation of guidelines, organisation of conferences and workshops for people interested in competition matters, and public comments concerning the current situation and the effects of government decisions and policy. The agency should publish bulletins with its most important decisions and comments by the case-handlers. In the majority of transition countries, the competition agencies are involved in competition advocacy. However, a large number of cases and many methodological problems do not allow competition authorities to engage in advocacy, given its significance for development of public awareness about competition concerns. All of the competition policy objectives described above are interdependent. Effective enforcement of the competition rules by the competition agency strengthens its actual authority in the government and in public. Competition policy objectives generally influence both government economic policy and public awareness about competition concerns. Both are relevant to construction of developed market forces and a new business culture in transition countries.
Conclusions Development of competition in transforming countries is important not only for progressive changes in the economy, but also for changes in the mentality of the people, which has great significance with respect to building a political environment for society's new activities. Effects of competition must be evaluated with respect both to the short term and the long term. In transforming countries, the shortcomings of competition are felt by citizens through the negative effects of reallocation of resources and changes in income distribution. However, restructuring of these economies requires radical reforms; postponing them may cause
78
Working Paper IV
larger losses in the future. Development of competition in transforming countries can be viewed as a process relevant to successful economic change. Effects of competition can be evaluated by implementation of the efficiency test. For purposes of competition law, economic efficiency can be defined at both a microeconomic and macroeconomic level. Elimination of inefficient firms will lead to an improvement of efficiency in the economy as a whole, because it will facilitate reallocation of resources and the more efficient use of resources. However, such a narrow definition of economic efficiency can result in price increases or growth in the rate of unemployment, both of which would be detrimental to public acceptance of market-oriented reforms. In transforming countries, the alternatives of postponing reforms or implementing other economic policy instruments may be available, and could mitigate the negative effects of such reforms. Effects of economic decisions made by state administrations or managers of firms should be evaluated from the perspective of dynamic efficiency, since transition from a command economy to a market economy involves the dynamic process of reallocation of resources. Apart from that, transition economies are a part of international markets, and domestic markets change very quickly. In this situation, static evaluation of efficiency may be misleading. For instance, profit gains of a firm may follow from its dominant position in a market, due to market foreclosures established by state authorities. It would be worthwhile to conduct a dynamic evaluation of efficiency, which means evaluation of existing profit gains and taking into consideration future competitive markets. Political and social goals, such as protecting weak firms or maintaining levels of employment, should not be competition policy objectives. State aid and protection of domestic producers are the most frequently employed methods for realisation of political and social goals. Both practices should be subject to scrutiny for their effect on competition in the relevant market. It may be justified to distort competition in a market for non-economic reasons, but such distortions should be temporary and aimed at solving actual problems within a defined period. Dynamic economic efficiency criteria must be applied to evaluate the effects of government projects. The main concern should not always be to contest the projects, but to measure the level of competition distortion which must be considered as externalities of these projects. Association of some transition countries with the European Union will support market-oriented reforms in these countries. Harmonisation of their competition law with E.U. competition law should facilitate the progressive change of emerging markets. This does not imply rapid and comprehensive adoption of European competition provisions by these countries. Differences between mature market economies and emerging markets are especially rele-
2 - Competition Policy Objectives
79
vant to deciding which methods of competition law enforcement are most appropriate. In general, it is desirable to have similar competition policy objectives in countries which are trade partners, and which are increasing the level of cooperation. The degree of harmonisation needed depends on the intensity of economic cooperation.
V Hideaki Kobayashix Deputy Secretary-General Japanese Fair Trade Commission Tokyo,Japan
When it was introduced 50 years ago, competition law was something totally unknown to political leaders, government officials, and the general public in Japan. Thus, the competition law of Japan, the Anti-Monopoly Act (A.M. A.), included a detailed and complex provision on objectives. Discussions on the objectives of the Act have led to important distinctions between (i) the objectives and the tools of competition policy and (ii) its two types of objectives: direct and ultimate, where the former are instruments to obtain the latter. The A.M.A. provision on objectives states: Art. 1. This Act, by prohibiting private monopolisation, unreasonable restraints of trade and unfair trade practices, by preventing excessive concentration of economic power and by eliminating unreasonable restraints on production, sale, price, technology and the like, and all other unjust restrictions of business activities through combinations, agreements and otherwise, aims to promote free and fair competition, to stimulate the creative initiatives of entrepreneurs, to encourage business activities of enterprises, to heighten employment and people's real income, and thereby to assure the interests of consumers in general, and also to promote democratic and wholesome development of the national economy.
The provision on objectives can be divided into several parts. Thefirstpart lists the instruments that the Act employs to attain its purposes: prohibition of private monopolisation, unreasonable restraints of trade and unfair trade practices, and prevention of excessive concentration of economic power. The second part specifies the tools referred to in the first part: elimination of unreasonable restraints of production, sale, price, technology, and the like and all other unjust restrictions of business activities through combinations, agreements and otherwise. The third part indicates the direct objective of the Act: to promote free and fair competition. The fourth part explains the significance of the Act's competition policy: to stimulate the creative initiative of entrepreneurs, to encourage business activities of enterprises, and to heighten the level of employment and people's real income. The fifth part clarifies the ultimate 1
The views expressed here are personal, and do not necessarily reflect the position of the Japanese Fair Trade Commission (J.F.T.C.).
82
Working Paper V
objectives of the Act: to assure the interests of consumers in general, and to promote the democratic and wholesome development of the national economy. Several questions arise with respect to objectives. First, should competition be pursued irrespective of its economic effects? Under the A.M.A., it should not, because the Act assures free and fair competition and not unqualified competition. 'Fair competition' is generally considered to mean 'competition by means of intrinsic elements of business such as price and quality'.2 Since the A.M. A. also states the value and the ultimate objective of competition policy, it does not endorse the kind of competition that would be contrary to them. Second, is 'efficiency' the sole benchmark of the legitimacy of competition policy? The objective provision of the Japanese competition law does not specifically mention efficiency. However, the provision indicates that creative initiative of entrepreneurs stimulated by free and fair competition results in encouraging business activities of enterprises and in heightening the level of employment and people's real income. One may deduce that an increase in efficiency is presumed. However, an increase in efficiency would not, for instance, render legal an otherwise illegal merger. Third, should maximisation of consumer welfare be given the highest priority? There has been long-standing discussion on this question in Japan in relation to the last clause of Art. 1, which states: 'to assure the interests of consumers in general, and also to promote the democratic and wholesome development of the national economy'. Various theories have been advanced: that 'consumer welfare' is subordinate to 'development of the national economy'; that 'consumer welfare' itself is not the objective of the competition policy, but competition policy results in an increase in consumer welfare; and that competition law is consumer protection legislation. The second theory finds the largest number of supporters. The first theory has given the theoretical background to industrial policy makers when they attempted to promote A.M.A. exemption legislation. However, the J.F.T.C. has been delegated some consumer protection responsibility. Specifically, the Act against Unreasonable Premiums and Misleading Representation provides that the J.F.T.C. will regulate the offer by companies of excessive free gifts as accompaniment to sales, or of excessive prizes through lotteries and competition, and false or misleading representations, such as in T.V. or newspaper advertisement or in labeling. This is considered to be an extension of the regulation of unfair trade practice. Fourth, can protection of the weak be considered an objective of competition policy? The Act is silent on this matter. This is natural because competi2
Imamura, Dokusenkinshiho 5-6 (1978).
2 — Competition Policy Objectives
83
tion law is designed to eliminate obstacles to free and fair competition, but not to guarantee protection of the weak. The provisions of competition laws related to protection of the weak usually prevent socially undesirable effects of competition. They often take the form of exemptions, such as that for cooperatives of small or medium sized enterprises or of consumers. Fifth, should increasing national income or employment be considered an objective of competition policy? Art. 1 of the A.M.A. refers to heightening the 'level of employment and people's real income'. However, as has been mentioned, this is usually interpreted as a statement of the value or significance of competition policy, rather than an objective. In practice, one of the strongest motivations for governments (including the Japanese government) to step up enforcement of competition policy is to improve international competitiveness through increased efficiency, and thus to increase employment and national income. Sixth, can promotion of regional integration be viewed as a legitimate objective of competition policy? For instance, strict territorial allocation schemes are subject to the rule of reason test in Japan. Attainment of an objective is in question here, which is outside the normal scope of competition policy objectives. Accordingly, it is not an objective of competition policy. Seventh, how should diversification of business conglomerates be handled? One of the A.M.A.'s objectives is the prevention of excessive concentration of business controlling power. Accordingly, Art. 9 of the A.M.A. prohibits the establishment of a holding company or transformation of any company into a holding company. This provision has two objectives: a political objective, to prevent reemergence of pre-war zaibatsu which were considered a driving force of the war effort, and were dissolved after World War II; and a competition policy objective, to prevent the emergence of a dominant economic power that could restrict competition in certain markets. In recent years, the Japanese business community has been requesting the abolition of this prohibition, stressing that increased competition in the domestic market has eliminated any risk of re-emergence of zaibatsu, and that the restriction unduly limits the alternatives of business organization for Japanese enterprises. The position of the J.F.T.C. is that since the risk of excessive concentration of business- controlling power still exists, restrictions on holding companies are still needed, although total prohibition may not be necessary. In March 1997, the J.F.T.C. proposed a bill to the Diet that would allow holding companies to the extent that they do not result in excessive concentration of businesscontrolling power. This is an example of the need for tailoring the provisions of the competition act to the specific condition, economic or otherwise, of each country.
84
Working Paper V
Eighth, how should below cost sales be controlled? The fairness element of the A.M.A.'s objectives implies that competitors must compete on a business basis. Moreover, some schools of thought hold that 'free competition' requires that competitors not be unreasonably excluded from competition. Below cost sales are problematic from the both standpoints. Under the Japanese competition law, below cost sales are actionable only when: goods or services are sold at a price far below their costs; over a substantial period of time, such sales are likely to create difficulties for other enterprises; and there is no justification for such sales. It would be difficult to outlaw all below cost sales under the competition laws. On the other hand, regulating below cost sales in order to prevent money laundering is a political decision. In any event, care should be taken to avoid any adverse effects on competition policy. Finally, will globalisation require the convergence of competition law among different countries? The most basic objective of competition law is to eliminate obstacles to free competition. This objective should be common among various jurisdictions. Moreover, from the viewpoint of creating a level playing field, it is desirable that the basic objectives of competition policy are the same all over the world. However, realising this objective is not easy. Accordingly, a more realistic objective would be to make the conditions of competition, or the degree to which obstacles are present, similar in each country. On the other hand, the economic conditions in each country (such as the degree of economic concentration, the presence of long term business relationships among enterprises, etc.) differ considerably. Accordingly, the specific instruments used to achieve competition policy objectives must be designed in accordance with the specific conditions in each country. Hence, it would not be desirable or necessary for each country to have the same instruments. It is difficult for countries to recognise the differences in the relevant economic conditions. Accordingly, the second best alternative may be agreement on a minimum standard for those instruments.
VI Francine Matte, Q. C. Senior Deputy Director of Investigation and Research, Canadian Competition Bureau Quebec, Canada
Introduction Globalisation, technological change and trade liberalisation have increased the integration of markets. In this context, competition policy must be flexible enough both to foster competition and, in full measure with the needs of the private sector, to make the adjustments necessary to respond to contemporary market demands. A framework that permits and encourages robust competition is a key component of a government's national economic policy. A. The Objectives of Canadian Competition Policy Historically, three major objectives of Canadian competition policy have been identified by the courts, ministers of the Crown, interested officials and academic analysts. They are: maintaining free competition; preventing abuses of market power, including the protection of consumers; and promoting economic efficiency. In addition, a number of supplementary objectives have been identified. These include codifying the common law doctrine of restraint of trade; fighting inflation; protecting small business; preserving the free enterprise system; and ensuring fairness and honesty in the market-place.1 In the 1970s, there was vigorous debate over the primacy of these objectives in government, business and academic circles. A consensus emerged that competition policy should focus on the goals of protecting and preserving competition and promoting economic efficiency. Subsequently, concern with economic efficiency as an objective of competition policy in Canada and elsewhere has further evolved to encompass the goal of international competitiveness. Increased efficiency is now widely acknowledged as the key to improved Canadian performance in world markets. Conversely, foreign competition via imports can play a key role in limiting the potential for abuse of market power in the domestic economy.2 1
For a fuller elaboration of these objectives, see Paul K. Gorecki & W.T. Stanbury, The Objectives of Canadian Competition Policy 1888-1983 (1984). 2 Bureau of Competition Policy, Canadian Competition Policy: Its Interface With Other Economic and Social Policies (1989).
86
Working Paper VI
1. The Evolution of Canadian Competition Policy a. Origins In May 1889, Canada became the first western industrialised nation to enact legislation designed to prevent firms from forming agreements in restraint of trade.3 However, enforcement was difficult due to cumbersome administrative procedures. Canada's traditional antitrust law, the Combines Investigation Act, was enacted in 1910, and significantly amended in 1976. The 1976 amendments: • extended the enforcement of Canadian competition law to service activities; • established civil enforcement against restrictive practices through the Restrictive Trade Practices Commission; • strengthened the prohibitions against collusive agreements, bid rigging and resale price maintenance; and • made civil suits available to private litigants injured through antitrust violations.4 Even with these amendments, the Act remained a relatively inflexible statute with too much reliance on criminal sanctions to deter anticompetitive conduct. Criminal sanctions were seen as particularly inappropriate to merger review and monopolies/abuse of dominant position, where a more flexible, case-by-case approach under a civil standard balance of proof was deemed desirable. b. The Competition Act of 1986s Canada's competition law was substantially overhauled and modernised in 1986, when the law was renamed the Competition Act (the Act). The Act is a framework law which establishes basic rules for the conduct of business in Canada, including both criminal and non-criminal provisions.
3
An Act for the Prevention and Suppression of Combination formed in Restraint of Trade, S.C. 52 Vic, c. 41 (1889). 4 See Paul K. Gorecki & W. T. Stanbury, The Administration and Enforcement of Competition Policy in Canada, 1889 to 1952, in Historical Perspectives on Competition Policy (R. Shyam Khemani & William T. Stanbury eds., 1991). As shown below, a private right of action is not available with respect to the non-criminal provisions of the Act. 5 Competition Act R.S.C., ch. C-34 (1985) (as amended).
2 - Competition Policy Objectives
87
The purpose clause of the Act summarises its objectives: to maintain and encourage competition in Canada, in support of four specific objectives: (1) to promote the efficiency and adaptability of the Canadian economy; (2) to expand opportunities for Canadian participation in world markets (while at the same time recognising the role of foreign competition in Canada); (3) to ensure that small and medium sized enterprises have an equitable opportunity to participate in the Canadian economy; and (4) to provide consumers with competitive prices and product choices. The Act applies to all sectors of the economy, with the exception of activities that are specifically exempted, such as collective bargaining and amateur sport, those subject to Crown immunity,6 or those that fall within the so-called regulated conduct defence.7 The rules under the Act differ fundamentally from direct economic regulation by a board or tribunal, in that they establish a broad framework within which businesses are largely self-governing. A high degree of voluntary compliance with the rules of the Act is relied upon, thereby precluding the need for ongoing monitoring of the behaviour and performance of firms. The Director of Investigation and Research (the Director) is an independent law enforcement official responsible for the administration and enforcement of the Act. He or she is appointed by, and serves at the pleasure of, the Cabinet. Employees of the Competition Bureau are on the Director's staff. The Director's role is to investigate, not to adjudicate, and he or she is empowered to conduct investigations with respect to both criminal and civil provisions of the Act. Evidence of criminal matters is referred to the Attorney General of Canada for possible prosecution in the criminal courts. With regard to civil matters, the Director applies to the Competition Tribunal for remedial orders designed to preserve competition. The Director can commence an inquiry when he or she believes there are reasonable grounds for making an order under a civil law provision, or where an offence has been or is about to be committed. The Director is also obliged to commence an inquiry if the responsible Minister so directs, or when six Canadian residents make an application pursuant to the Act. The discontinuance of an inquiry must be reported to the Minister, who can order further 6
The Supreme Court of Canada confirmed the application of Crown immunity for competition law offences in R. v. Eldorado Nuclear Ltd., 2 S.C.R. 551 (1983). However, s.2.1 of the Competition Act provides that the Act is binding on those considered as acting as an agent of the federal or provincial government 'in respect of commercial activities engaged in by the commercial corporation in competition, whether actual or potential, with other persons to the extent that it would apply if the agent were not an agent of Her Majesty.' 7 See part B.I, infra.
Working Paper VI
inquiry. The Minister, however, cannot order the Director to discontinue an inquiry or to conduct the inquiry in a specified manner. Traditional enforcement actions are complemented by a comprehensive programme to encourage voluntary compliance with the Act. This programme encompasses extensive public information activities to inform the public about the application of the Act, and the provision of advisory opinions upon request. In addition, the Competition Bureau publishes guidelines and bulletins to foster public understanding of the Bureau's enforcement and compliance policies. The Act also provides explicit statutory authority for the Director to make interventions before federal and provincial regulatory agencies that make decisions affecting competition in particular markets. The Competition Bureau also has the opportunity to provide input into the design and implementation of other government policies that affect the competitive market system, such as financial sector reform. The Bureau also represents Canada's interests in international competition policy fora. In addition to the core provisions of the legislation dealing with agreements in restraint of trade, mergers, abuse of dominant position, deceptive marketing practices and other matters, the Act includes provisions dealing with specialisation agreements, which allow two or more firms to reorganise their production and product lines in order to achieve the efficiency gains made possible by specialisation and longer production runs. Unlike the other non-criminal provisions of the Act, private parties may apply to the Competition Tribunal to authorise the registration of a specialisation agreement. This authorisation of a specialisation agreement exempts it from application of the conspiracy, tied selling, exclusive dealing and market restriction provisions of the Act. The Competition Tribunal Act, which was also adopted in 1986, provides for the establishment of the Competition Tribunal which adjudicates civil matters on application by the Director.8 The Tribunal operates at arm's length from the Director. It includes judges drawn from the Federal Court, Trial Division, and lay members, who are typically economists or individuals with business experience. Decisions of the Competition Tribunal cannot be appealed to the Cabinet of the federal government on public interest or other grounds. This differs from other economic framework legislation in Canada, such as the Communications Act, where decisions of the Canadian Radio-television and Telecommunications Commission ('C.R.T.C.') may be appealed to the Cabinet. 8
The Restrictive Trade Practices Commission was abolished with the establishment of the Competition Tribunal.
2 - Competition Policy Objectives
89
Similarly decisions of the Director have been characterised as administrative by the courts, with very limited scope for judicial review.
2. Overview of Key Provisions a. Abuse of Dominance The abuse of dominance provision is one of the key elements of the Competition Act. This non-criminal provision replaced the all but unenforceable criminal law on monopoly in 1986. This provision applies if one or more persons substantially control a class of business throughout Canada, and have engaged in, or are engaging in, anticompetitive acts that have the effect of preventing or substantially lessening competition. In these circumstances, the Tribunal may issue an order prohibiting or otherwise remedying the anticompetitive conduct at issue. However, in assessing the effects of anticompetitive acts, the Competition Tribunal may consider whether the practice is a result of superior competitive performance. This provision recognises that consumers benefit when product innovation or improved distribution systems result in a firm prevailing over its rivals in the market-place. The legislation also provides that the legitimate exercise of intellectual property rights is not an anticompetitive act. b. Mergers The merger provisions apply to all mergers in Canada, irrespective of the size or nationality of the parties. Firms are required to fulfill mandatory prenotification requirements when those transactions exceed particular threshold levels.9 The assessment of mergers by the Bureau focuses on the question whether a proposed transaction is likely to prevent or lessen competition substantially. The assessment is not based strictly on market share or concentration, but also may include a number of qualitative factors including, inter alia, the extent of effective foreign competition, whether one of the merging parties is a failing business, the existence of any barriers to entry (including regulatory barriers), and the nature and extent of innovation in the markets under examination. These qualitative considerations are discussed in detail in the Merger Enforcement Guidelines issued by the Director in March 1991. 10 9
The law requires prenotification of large merger transactions, where the combined revenues or assets of the parties exceed $400 million, and the general value of the acquisition target exceeds $35 million in revenue or assets, apart from amalgamations. 10 See Director of Investigation and Research, Merger Enforcement Guidelines (1991), 450-2 & Appendix 2 .
90
Working Paper VI
The Act also provides for an efficiency exception in the event that a substantial prevention or lessening of competition is found likely. The merging parties must demonstrate that the gains in efficiency which will likely arise from the merger will be greater than, and will offset, the effects of any likely substantial prevention or lessening of competition, and that these gains would not likely be attained if the merger were prohibited. In this context, the Bureau requires the provision of meaningful data and verifiable information by the parties to back up their claims of efficiency gains, expanded exports and import substitution predicted to result from a merger.11 In specific circumstances, the Director shares jurisdiction for mergers with two other agencies of the federal government. The C.R.T.C. has jurisdiction over certain industries regulated under the Communications Act, while the Minister of Industry is responsible for reviewing large merger acquisitions involving nonCanadians under the Investment Canada Act. The Director's role in merger review is primarily to investigate and to apply to the Competition Tribunal for orders. The C.R.T.C. and Investment Review Division of Industry Canada perform essentially administrative law functions. Until 1996, the National Transportation Agency had the power under the National Transportation Act to review mergers in federally regulated transportation industries, such as marine shipping and air transportation. In 1996 this power of review was removed. c. Conspiracy Another key provision of the Act relates to conspiracy, which is the oldest existing provision of Canadian competition law. Section 45 of the Act makes it an indictable (i.e. criminal) offence for any person to conspire, combine, agree or arrange with another person to prevent, limit or lessen competition unduly. The Act lists several exceptions, including cooperative arrangements relating to research and development, the exchange of statistics and product standards. These exceptions do not apply in circumstances where an agreement is likely to lessen competition in respect of prices, the quantity and quality of production, markets or customers, or distribution channels. If the conspiracy relates only to the export of products from Canada, the conspiracy section does not apply—provided that the conspiracy does not result in a reduction or limitation of the real value of a product, or restrict any person from entering into or expanding the business of exporting products from Canada. Section 46 of the Act also proscribes participation in foreign business practices that give effect to a conspiracy, combination agreement or arrangement 11
For elaboration, see Margaret Sanderson, Efficiency Analysis in Canadian Merger Cases (1996).
2 - Competition Policy Objectives
91
entered into outside Canada, that if entered into in Canada, would be in contravention of section 45.12 d. Deceptive Marketing Practices
The misleading advertising and deceptive marketing practices provisions of the Act are intended to promote the honest and efficient functioning of the market by prohibiting representations which are false or misleading in a material respect. The rationale underlying the misleading advertising provisions, which are criminal, relates to market information to both consumers and businesses. Unsubstantiated performance and durability claims, as well as misleading warranties and misrepresentations as to regular price, fall into this category. The provisions also cover multi-level marketing and pyramid selling plans, baitand-switch-selling and deceptive promotional contests. Enforcement of the misleading and deceptive marketing practices provisions of the Act is geared to ensuring that no competitor gains or retains market share by deception. The overall policy goal is to allow competitive market forces, within the framework provisions of the Competition Act, to determine winners and losers in the market-place.
B. The Interface Between Competition and Economic Regulation 13 1. The Regulated Conduct Defence In Canada, the interplay between competition law and regulation is governed to a large extent by a jurisprudential doctrine known as the regulated conduct defence. The essence of this doctrine is that specific activity which is authorised or carried out pursuant to a valid scheme of regulation is deemed to be in the public interest. As such, it cannot violate the criminal provisions of the Act. There is no conclusive jurisprudence with respect to the civil provisions of the Act, and it remains for the courts and the Competition Tribunal to determine the full scope of the application of the regulated conduct defence.14 12
See R. v. Sumitomo Canada Limited, Ottawa T-2687-93 (F.C.T.D. 1993). For a fuller elaboration, see Robert D. Anderson et al., Regulatory Reform and the Expanding Role of Competition Policy in the Canadian Economy, 1986-1996, in Review of Industrial Organization (forthcoming); Donald Mercer, The Regulated Conduct Defence and the Telecommunications Sector (1995). 14 This issue has been touched on in two recent cases: Canada v. Bank of Montreal et al., Comp. Trib. CT 9502/92,20 June 1996 (unreported)(Interac); Law Society of Upper Canada v. Canada, No. RE5349/95 (Ontario Court (General Division) 1996). 13
92
Working Paper VI
2. Advocacy Sections 125 and 126 of the Act provide the Director with the independent authority to advocate competition in certain circumstances. Section 125 enables the Director to make representations in respect of competition before any federal board, commission or other tribunal where such representations are relevant to the matters under consideration. Section 126 empowers the Director, with the consent of a provincial regulatory board or tribunal, to make similar representations before any such provincial regulatory boards, commissions or tribunals. The Bureau has made formal and informal interventions in support of competition ranging from presenting evidence and participating in regulatory proceedings to providing working level commentary and advice to colleagues in other departments or agencies. The Bureau focuses its advocacy work on hearings, reviews and other proceedings in which it can make a significant contribution to enhancing or protecting competition in a market. For instance, the Bureau has intervened in proceedings regarding licensing or other such entry restrictions, unnecessary impediments to competition created by the regulatory, legal and institutional structure of industry, anti-dumping duties and other impediments to import competition and establishing the terms for competitive access to essential facilities. Since 1976, the Competition Bureau has made approximately 200 written submissions.15 The Bureau views competition advocacy and enforcing the Act in particular sectors to be complementary. Effective advocacy which results in liberalisation of regulations governing particular sectors must be followed by vigorous enforcement to prevent private anti-competitive practices that could nullify the gains from regulatory reform. As a result of regulatory reform in past years, the Bureau has seen a substantial increase in its responsibilities in terms of the amount of economic activity subject to the Act. The focus of the Bureau's advocacy activities has evolved over time in response to shifting priorities and opportunities for effective intervention. The first half of the 1980s witnessed a large number of interventions (in the range of 20-25 per year) in the transportation and telecommunications sectors, as the ground work was laid for deregulation of those sectors. Electricity generation and financial services are currently at the same stage in Canada that telecom15
See Joseph Monteiro, Regulatory Interventions by the Bureau of Competition Policy, in Canadian Competition Law and Policy at the Centenary (R. Shyam Khemani and William T. Stanbury, eds., 1991).
2 - Competition Policy Objectives
93
munications and transportation were ten years ago. The Bureau has begun to advocate competition infinancialmarkets and in the energy sector. The Bureau is likely to renew its advocacy of competition in the agriculture sector before marketing boards and related institutions. a. Telecommunications Competition in the Canadian telecommunications markets has been a priority of the Bureau for more than a decade, beginning with terminal attachment, then long distance competition and currently local telecommunications services. The passage of the Telecommunications Act16 in 1993 highlights the convergence of telecommunications regulation and competition law in Canada. A primary policy objective of the Telecommunications Act is increased reliance on market forces for the provision of telecommunication services. Under this Act, the Canadian telecommunications regulatory body, the C.R.T.C., is required to forbear from regulation where itfindsthat a service or a class of services is or will be subject to competition sufficient to protect the interests of consumers. The Director has been a frequent intervenor before the C.R.T.C. in numerous proceedings relating to long distance and local telecommunications competition, the regulatory framework for telecommunications and the distribution of broadcast services, as well as forbearance from regulation in a number of areas, including most recently long distance services. The Bureau has also made representations with respect to the review of Teleglobe's statutory monopoly on the carriage of international traffic, and competition in the distribution of broadcast services by cable, direct broadcast satellite or other technologies. The Bureau has advocated that certain fundamental principles should govern the regulatory framework for the development of competition in communications services. In summary, these principles are: • Maximise reliance on competition and market forces at the outset; • As a corollary, minimise regulation for incumbents and avoid imposing economic regulation on new entrants; • Adopt market-based pricing as soon as possible in local telecommunications and, if necessary, introduce specific, targeted mechanisms to address social policy objectives; • Establish clear rules governing incumbents' obligations to provide access to their networks by competitors, and adopt appropriate pricing principles to induce efficient competition; 16
Telecommunications Act, S.C. 1993, Ch. 38.
94
Working Paper VI
• Establish timely and effective dispute resolution mechanisms to ensure that incumbents do not attempt to delay access to their networks; and • Liberalise foreign ownership rules for communications networks to assist in the rapid construction and development of communications networks. Progress has been made to opening Canadian telecommunications markets to greater competition and reducing the level of regulation. Thus, the Bureau is shifting its emphasis away from regulatory and policy interventions, toward greater enforcement activity, including merger and other restructuring issues which are emerging in the transition period to a competitive market. b. Electricity The energy sector in general and the electricity industry in particular pose new challenges that require re-evaluation of existing regulatory structures. The Canadian electricity industry is only starting down the road toward open competition. This sector is a high priority for the Bureau, which can have a significant impact in establishing a competitive framework for the long term. The Bureau has been extensively involved in the Province of Ontario's deliberations on reforming the Ontario electricity system through the provision of comments and analysis to the relevant department of the Ontario government. It has provided input to the Advisory Committee on Competition in Ontario's electricity system, with respect to legislation and regulation restricting competition, particularly the generation and retailing of electricity. Issues that the Bureau addressed included the need for effective separation of the natural monopoly parts of the sector (transmission and distribution) from competitive parts (generation, retailing and electricity-related markets such as energy services), restrictions on international and interprovincial trade and competition in electricity, and regulatory and structural approaches for minimising the threat of cross-subsidisation from monopolistic to competitive parts of the sector. The Advisory Committee's recommendations are currently under consideration by the Ontario Government.17 While no specific restructuring proposals have been forthcoming from the provincial government, there appears to be strong support within the province for major pro-competitive reforms. c. Agriculture Between 1975 and 1996, the Bureau made thirteen interventions in the agriculture or agri-food sectors. A majority of these interventions occurred after 1986, 17
See A Framework for Competition: The Report of the Advisory Committee on Competition in Ontario's Electricity System to the Ontario Ministry of Environment and Energy (1996).
2 - Competition Policy Objectives
95
when the sector was initially exposed to competitive forces. The interventions covered the poultry, dairy, and fruit and vegetable sectors. Seven representations were made to various federal boards, six to provincial boards. The issues addressed ranged from licence and access applications, merger agreements, dumping, etc. There have, however, been two overriding issues in these interventions: supply management and price setting.18 In its submissions, the Bureau has opposed marketing board schemes to determine the level of output and prices based on cost of production formulae, on the grounds that they would reduce output, raise prices, cause income distribution problems and use inefficient prices in costing formulae. The Bureau opposed government price controls in interventions on price setting issues, arguing that market based arrangements are more consistent with maintaining and promoting an efficient and competitive processing sector. d. Antidumping In Canada, the Special Import Measures Act19 governs the application of antidumping and countervailing duties to imports of dumped or subsidised goods that injur domestic producers. Since the early 1980s, the Competition Bureau has intervened in seven cases before the Canadian International Trade Tribunal and its predecessor organisations to address questions of material injury and public interest. The Bureau intervenes in proceedings that address significant economic sectors where important competition issues are at stake, particularly on material injury and public interest in these dumping cases. These interventions have involved the following industries: sugar, cars from Korea, fibreglass pipe insulation, beer and home canning materials. The Bureau has also played an important role in the reform of anti-dumping and countervail policy in Canada. This role has ranged from interdepartmental development of positions for trade negotiations, to making submissions and an appearance before a special House of Commons' Committee examining the Special Import Measures Act.20 A significant number of the Bureau's 18
See Gerald Robertson et al., Competition Policy, Trade Liberalization and Agriculture (a paper prepared for the Third Policy a n d T r a d e Disputes W o r k s h o p , Tucson, Arizona, M a r c h 1997). 19 Special I m p o r t Measures Act, R.S.C. 1985, Ch. 5-15. 20 See Submission of the Competition Bureau, Industry C a n a d a , t o the SubCommittee on the Review of the Special Import Measures Act of the Standing C o m mittee on Finance and the subcommittee on T r a d e Disputes of the Standing Committee on Foreign Affairs a n d International Trade, regarding a Call for C o m m e n t s concerning the Review of the Special Import Measures Act (20 N o v e m b e r 1996).
96
Working Paper VI
proposals have been taken up in the Committees' Report,21 and have been included in government's response to the Report.22
C. Competition Policy and Economic Restructuring Competition policy has a particularly important role in an era of globalisation, trade liberalisation and increased mobility of production factors. While trade liberalisation, privatisation and deregulation can invigorate markets, they are not a substitute for the sound application of competition law and policy.23 Moreover, the evidence indicates that vigorous rivalry in domestic markets for goods and services promotes the competitiveness of domestic firms in export markets by fostering innovation and productivity improvement.24 Competition law and policy defines the rules and the environment within which Canadian companies can develop the strength needed to compete. The Act is fully compatible with the Canada-U.S. Free Trade Agreement and the North American Free Trade Agreement. By providing an efficiency gains defence in certain cases for otherwise anticompetitive mergers, it explicitly recognises the need for industry to adjust. Prior to the mid-1980s, Canada's economy was small, regulated and protected, and had inefficient manufacturing capacity and high-cost distribution systems. The 1986 Act facilitates the move by Canadian industry to a more open and deregulated economy. Vigorous enforcement of competition law provides incentives for investment in continual innovation and the systematic upgrading of products and production processes, necessary to keep pace with competitors. This serves the twin objectives of enhancing companies' ability to compete successfully in world markets and maximising benefits to the Canadian economy.
21
T h e Reports of the Sub-Committee was tabled in the House of C o m m o n s on 11 December 1996. See http://www.parl.gc.ca. 22 T h e Government Response to the Report on the Special Import Measures Act was tabled in the H o u s e of C o m m o n s on 18 April 1997 by the Minister of Finance, the H o n o u r a b l e Paul M a r t i n . See http://www.fin.gc.ca. 23 See R o b e r t D . A n d e r s o n a n d S. D e v K h o s l a , Competition Policy as a Dimension of Economic Policy: A Comparative Perspective, Occasional P a p e r N o . 7 (Competition Bureau, Industry C a n a d a 1995). 24 See Michael Porter, The Competitive Advantage of Nations (1990), 117.
2 - Competition Policy Objectives
97
1. Enforcement of the 1986 Competition Act The Act, which came into force on 19 June 1986, differs considerably from previous legislation. Since the Act took effect, court decisions have clarified its meaning and assisted the business community in understanding and complying. The priority areas of enforcement have been abuse of dominance, mergers and conspiracy. In addition, direct mail promotions, telemarketing and marketing practices promoting the sale of telecommunications services are priorities in the enforcement of the deceptive marketing practices provisions.
2. Abuse of Dominance Prior to passage and implementation of the Act, convictions under the monopoly provision of the old law were rare. Between 1935 and 1985, seventeen cases were initiated, in only two of which convictions were obtained. Prohibition orders were issued in six of the cases (including one where a conviction was obtained), defendant firms were acquitted in six cases, and in the remaining four, charges were withdrawn. Remedial orders were issued in each of the six cases brought forward by the Director.25 Under the new Act, considerable experience has accrued with respect to abuse of dominant position provisions on such issues as market definition, contract exclusivity, allocation of markets, the concept ofjoint dominance, directed sales and the creation of barriers to product innovation. In addition to gaining experience in investigating and litigating these cases before the Competition Tribunal, several procedural issues, such as the role of intervenors, confidentiality and others, have been resolved by the Tribunal and the courts. In Canada v. Laidlaw Waste System Ltd.,26 the Director requested the Tribunal to consider a number of practices designed to create and protect a dominant position. These included, inter alia: acquiring competitors in the relevant market; entering a market by acquiring the sole competitor, then excluding that competitor through restrictive covenants; maintaining dominance by 25 See C a n a d a v. D & B Co. (Canada), 64 C.P.R.(3d) 216 (1996)(Neilsen); C a n a d a v. Bank of Montreal et al., supra note 14; C a n a d a v. NutraSweet C o . , 32 C.P.R.3d 1 ( C o m p . Trib. 1990); C a n a d a v. Laidlaw Waste System Ltd., 40 C.P.R.3d 289 ( C o m p . Trib. 1992); C a n a d a v. Tele-Direct (Publications) Inc., N o . C T - 9 4 / 3 (Comp. Trib. 26 February 1997); C a n a d a v. A G T Directory Limited et al., N o . CT-94/2 ( C o m p . Trib. 18 N o v e m b e r 1994). 26 Supra note 25.
98
Working Paper VI
signing customers to long term exclusive contracts, which included right to compete clauses that gave Laidlaw the right to match the price of any competitor or any provision, and required the customer to accept Laidlaw's offer.27 The Tribunal ordered amendments to the customer service agreements and required Laidlaw to refrain from enforcing those terms in existing agreements that were found to be anticompetitive. In Canada v. The NutraSweet Co., the Tribunal addressed a number of issues relating to the role of intellectual property rights in the sweetener industry.28 The Tribunal accepted the Director's argument that an allowance which NutraSweet offered to its customers to encourage them to display the NutraSweet logo constituted an anticompetitive act, in view of the associated requirement that customers displaying the logo must use exclusively NutraSweet brand aspartame. In the Tribunal's view, this created an 'all or nothing' choice for purchasers, which impeded entry by competitors. The Tribunal's remedial order prohibited NutraSweet from continuing to offer discounts for the display of its trade mark or logo, and from engaging in other acts, such as offering its customers meet-or-release or 'most favoured customer' clauses, which were deemed anticompetitive. The decision suggests that the Tribunal viewed NutraSweet's marketing strategy as an attempt to extend artificially the duration of the company's Canadian patent protection. The Tribunal's decisions in these cases are particularly significant with respect to enforcement policy. Exclusionary and restrictive clauses are a frequent occurrence in certain industries' contracts. Where such clauses do not have a proper business justification, and serve only to protect a dominant position by creating barriers to entry, the Director will seek corrective action.
3. Mergers The merger provisions of the Act have generated the greatest interest over the past ten years. Previous criminal merger provisions penalised mergers that were detrimental to the public, and were an inappropriate standard to evaluate complex economic arguments and evidence. Since 1986, mergers have been examined under non-criminal standards, according to a merger's impact on competition and economic efficiency. Since the passage of the Act, a number of key cases have been decided, providing guidance in the merger area. For instance, in Canada v. Hillsdown 27 28
40 C.P.RJd 289 (1992). Supra n o t e 25.
2 - Competition Policy Objectives
99
Holdings (Canada) Ltd.,29 the Tribunal noted that while market share can give a prima facie indication whether a merger is likely to lead to an enhancement of market power, the Director is also responsible for adducing some evidence regarding barriers to entry. Specifically, the Tribunal observed that 'in the absence of entry barriers, it is unlikely that a merged firm could maintain supra-competitive profits, regardless of market share or concentration'.30 The Hillsdown case also provided the backdrop for a discussion on the treatment of efficiencies under the merger provisions. Specifically, the Tribunal indicated that it had difficulty accepting the Director's position that pure transfers of wealth from consumers to producers were not 'effects' of a lessening of competition within the meaning of the Act. It offered as an alternate test: the weighing of inefficiencies against the likelihood that detrimental effects (both the deadweight loss and the wealth transfer, or a portion thereof) will arise from the substantial lessening of competition. The Tribunal thereby apparently suggested that the larger the price effects of a merger, the less weight should be given to efficiencies. The Tribunal's comments have been criticised as having the potential to erode the efficiency orientation of Canadian merger policy, which facilitates business restructuring. A pure consumer price test would effectively offset the efficiency exception of the Act. However, since the Tribunal's observation on this point was obiter dictum, it is not necessarily binding on future cases. The Director, therefore, elected not to appeal this aspect of the decision. In Canada v. Southam Inc.,31 the Federal Court of Appeal affirmed the conceptual approach to the delineation of relevant markets which is used by the Bureau and set out in the Merger Guidelines. The Court found that the Tribunal had given too much weight to the lack of evidence of price sensitivity as an indicator of the lack of competition, and had ignored other evidence that revealed that competition was present. This decision was recently overturned by the Supreme Court of Canada, largely on the administrative law grounds that as a specialised, quasi-judicial body, the Tribunal's decision should be accorded considerable deference by the appellate courts.32 The question before the Supreme Court was the application of tests for product market definition in the case before it. The Court observed that the weighting and application of certain factors should be 'subtle and flexible', not mechanical, and that 'as a 29 30 31 32
41 C . P . R J d 289 (Comp. Trib. 1992). Qd. at 324. 63 C.P.R. 3d 67 (F.C.A. 1995); Southam et al. v. Canada, No. A-1668-92 (C.A.). C a n a d a v. Southam Inc., [1997] 1 S.C.R. 748.
100
Working Paper VI
matter of law, the Tribunal should consider each factor, but according of weight to the factors should be left... to the Tribunal'.33 However, the Supreme Court upheld the Tribunal's and Federal Court of Appeal's remedy order in the Southam case, which involved the print real estate advertising market in the British Columbia lower mainland. The Supreme Court held that the correct test is whether the proposed remedy will effectively eliminate the substantial lessening or prevention of competition. A remedy need not cause the market to return to its pre-merger state of competition, as the Tribunal originally required. Canada v. Asea Brown Boveri Inc.2"* involved restructuring in response to the lowering of trade barriers that resulted from commitments made in the Canada-U.S. Free Trade Agreement (C.U.S.T.A.). In 1989, the Tribunal issued a Consent Order under section 105 of the Act in relation to the acquisition of the electric power transmission and distribution business of Westinghouse Canada Inc. (Westinghouse) by Asea Brown Boveri Inc. (A.B.B.). The order required that A.B.B. divest certain assets acquired from Westinghouse in the event that it was unable to obtain specific tariff relief measures for imports of medium and large power transformers. Tariff protection was viewed by the Director as a significant barrier to entry in relation to competition in the Canadian market-place. On 24 May 1990, the Privy Council issued an order giving effect to the implementation of accelerated tariff reductions under the C.U.S.T.A., as required by the consent order. Revitalisation of the consent order process is a significant recent development with respect to merger review in Canada. Three consent orders are currently before the Tribunal. In Canada v. Dennis Washington et al. (Seaspan)35 and Canada v. Canadian Waste Services Inc.,36 the Tribunal issued consent orders in February and April 1997, respectively. The Director filed an application for a consent order in Canada v. ADM Agri-Industries, Ltd.31 in March 1997. Behavioral provisions, such as supply contracts, may be included in the remedial order, along with structural relief, divestiture offixedassets and customer contracts.
33
2d a t 770. Unreported decision (Comp.Trib., 18 December 1989). 35 CT-96/01 (unreported). 36 CT-97/1 (unreported). 37 CT-97/02 (unreported). 34
2 - Competition Policy Objectives
101
4. Conspiracy A central element of the conspiracy section is the test of 'undueness' which is applicable to agreements that lessen competition. The decision of the Supreme Court of Canada in R. v. Nova Scotia Pharmaceutical Society ('PANS')38 confirms that the test of undueness embodies a partial 'rule of reason' standard. This requires that the seriousness or significance of the effect(s) of a particular anticompetitive agreement be considered. The test does not go so far as to permit potential efficiency-related benefits that derive from cooperative arrangements among competitors to be considered. In its decision, the Court also elaborated on its views regarding the major elements of the examination mandated under the conspiracy provisions. With respect to market structure, the court clearly emphasised that the main objective is to determine the degree of market power enjoyed by the parties to the agreement. Market share, while a relevant factor, is not determinative and must not be considered in the context of other relevant factors. Effective enforcement in a global economy requires cooperation between antitrust enforcement agencies. In the fax paper case, the cooperative mechanisms available under the Canada-United States Mutual Legal Assistance Treaty in Criminal Matters39 were employed. They enabled the Bureau and the Antitrust Division of the U.S. Department of Justice to undertake a joint investigation and secure convictions in both countries against four foreign suppliers of thermal fax paper. In 1994, Kanzaki Specialty Papers Inc.40 pleaded guilty to having conspired to lessen competition in Canada. Mitsubishi Corporation and its Canadian subsidiary Mitsubishi Canada Limited41 also pleaded guilty to having conspired to lessen competition as well as having engaged in price maintenance activities in Canada. Following further investigation in Canada and the United States, guilty pleas, fines and prohibition orders were secured against Rittenhouse Ribbons and Rolls Ltd.,42 New Oji Paper Company Ltd.,43 and
38
43 C.P.R.3d 1 (S.C.C. 1992). M u t u a l Legal Assistance in Criminal Matters Act, R.S.C., C h . C - 3 0 (4th S u p p . 1985). 40 C a n a d a v. Kanzaki Specialty Paper Inc., 82 F.T.R. 63, 56 C.P.R.3d 467 (1994). 41 C a n a d a v. Mitsubishi C a n a d a Limited a n d Mitsubishi C o r p o r a t i o n , N o . T-1825-94 (8 August 1994) (unreported). 42 C a n a d a v. Rittenhouse Ribbons a n d Rolls Ltd., N o . T - 2 6 6 3 - 9 5 (20 D e c e m b e r 1995) (unreported). 43 C a n a d a v. New Oji Paper C o m p a n y Ltd., N o . T - 1 6 9 3 - 9 6 (16 August 1996)(unreported). 39
102
Working Paper VI
Mitsubishi Paper Mills.44 Fines in Canada in the fax paper matter have totalled $3.45 million.
5. Deceptive Marketing Practices The proliferation of sophisticated communications technologies has enabled illicit businesses to communicate rapidly and cheaply with their victims, no matter where they are located. Free trade agreements and deregulation within certain sectors have also helped to facilitate international consumer transactions. However, statutory and regulatory jurisdictions have remained fragmented and unco-ordinated. Consequently, in a number of cases, domestic markets are not well protected, and victims have little recourse when misleading advertising or deceptive marketing practices originate outside the country. The Bureau is especially concerned about the potential for on-line ('Internet') deceptive marketing practices. Cybermarketing is a fast growing activity bringing to consumers on-line magazines, catalogues, shopping centres and interactive advertisements. Canada is a member of the International Marketing Supervision Network. The Network, established in 1992, is an alliance of organisations from member and observer countries of the O.E.C.D. involved in the promotion and enforcement of fair trading practices. Its primary objectives are to share general information on cross-border deceptive marketing activities and to explore opportunities for cooperative enforcement. Canada currently serves as chair of the organization. At the last full Network conference, held in Ottawa in September 1996, examples of deceptive telemarketing schemes and successful cross-border enforcement initiatives were discussed. Delegates also examined their different regulatory regimes and how these can be effectively reconciled to achieve the common goal of promoting fair trading practices. Fiscal restraints and globalisation of markets will challenge the Bureau's abilities to enforce the misleading advertising and deceptive practices provisions of the Competition Act efficiently and effectively. The Bureau will continue to opt for a compliance-oriented approach, under which enforcement initiatives are directed selectively to have major economic impact on the Canadian marketplace, maximum deterrent effect, or optimal development of case law.4S 44
C a n a d a v. Mitsubishi P a p e r Mills, N o . T - 1 2 5 - 9 7 (17 F e b r u a r y 1997)(unreported). F o r disposition of m o s t recent cases, see D i r e c t o r of Investigation a n d Research, Misleading Advertising Bulletin. 45
2 - Competition Policy Objectives
103
D. The Interface Between Competition Policy and Trade Policy Trade liberalisation and competition policy share broadly similar goals. Both seek to ensure that artificial barriers to the competitive process are removed to the greatest extent possible in order to encourage efficiency in the production and allocation of goods and services. One specific goal, market access, represents an important confluence between trade and competition policies. Measures that generally foster global contestability of markets will be consistent with objectives of both trade and competition policies.46 There is a growing acceptance of the linkage between strong rivalry in the domestic market, supported by vigorous antitrust enforcement, and the realisation of gains from trade liberalisation agreements. Domestic competition laws complement trade liberalisation agreements by ensuring that the benefits of such agreements are not negated by private restraints of trade. Strong competition in domestic markets also helps to smooth the structural adjustments that are the inevitable result of any trade liberalisation accord.47 The close relationship between competition policy and international trade is directly reflected in several provisions of the Competition Act. As noted above, the purpose clause refers specifically to the importance of Canadian participation in world markets as well as the role of imports in strengthening competition in the domestic economy. The role of imports as a competitive alternative is also included in the list of factors to be considered under the merger provisions of the Act. This view was reinforced by the Competition Tribunal in its Hillsdown decision, which focused on the potential competition offered by U.S. facilities that were not yet strongly active in the Canadian market. Competition policy has already begun to make an appearance in various international agreements. While the Canada-U.S. Free Trade Agreement Trade did not contain any specific chapter on harmonisation or convergence of antistrust laws between Canada and the U.S., consideration of competition policy did arise in the context of the convergence of anti-dumping and antitrust laws. In the North American Free Trade Agreement (N.A.F.T.A.), there is an explicit provision on competition law: Chapter 15 reflects a commitment in principle to maintaining and enforcing national competition laws and to promoting effective competition law enforcement in the North American free 46 See R. Shyam Khemani, The Complementarity Relationship Between Trade Policy and Competition Policy, Symposium Paper N o . 10, prepared for the O . E . C . D . T r a d e C o m m i t t e e Symposium on Regulatory Reform and International M a r k e t O p e n n e s s (9-10 July 1996). 47 See Anderson and Khosla, supra note 23.
104
Working Paper VI
trade area. In the Canada-Chile Free Trade Agreement, the parties agreed to mirror the language of N.A.F.T.A. on competition. The parties also agreed to the mutual elimination of anti-dumping duties within a maximum of six years. In December 1996, following the World Trade Organization (W.T.O.) Singapore Ministerial meeting, a working group was established to study issues raised by members relating to the interaction between trade and competition policy, including anticompetitive practices, in order to identify areas that may merit further consideration in the W.T.O. framework.48 A second group on trade and investment was also established. The Organization for Economic Cooperation and Development (O.E.C.D.) has done extensive work in the area of trade and competition, which will undoubtedly contribute to the efforts of the WTO working groups.
E. Future Directions and Emerging Issues Given the internationalisation of markets, competition policy must be in full measure with the needs of the private sector to make the essential structural adjustments in response to evolving market demands. This reflects the shift in economic philosophy over the past decade away from government intervention and towards a new vision of an economy discussed in global terms, in which competitive market forces are the key factor in this global economy. The Competition Act of 1986 has now been in force for almost eleven years. It is the principal piece of framework legislation aimed at maintaining and strengthening the role of competitive forces in Canada. The Act successfully anticipated the competitive challenges arising from trade liberalisation agreements, such as the Canada-U.S. Free Trade Agreement and N.A.F.T.A. Furthermore, trade liberalisation has been reinforced on the domestic side by an increasing trend towards privatisation and deregulation. An Agreement on Internal Trade (A.I.T.) amongst the Canadian provinces has also been entered, the goal of which is to eliminate interprovincial barriers to the free flow of goods, services, labour and capital.49 The government is cognizant of the need to ensure that the Competition Act keeps pace with the corporate strategies of global firms. In their search for greater economic efficiency, firms are increasingly using new combinations of 48
World Trade Organization, Singapore Ministerial Declaration, N o . 96-5315, (9-10 December 1996). 49 See First Ministers of Canada, the ten provinces and the two territories, Agreement on International Trade (Ottawa, 18 July 1994).
2 - Competition Policy Objectives
105
investment, trade and collaborative agreements to expand internationally, enter new markets and exploit technological and organisational advantages as widely as possible. Bill C-67, introduced before Parliament in November 1996, proposed changes to the Act, essentially aimed at more effective and flexible enforcement. The amendments are intended to provide quicker and more effective resolution of misleading advertising and deceptive marketing practices; address the recent proliferation of deceptive telemarketing practices; improve the administration of the merger prenotification process while reducing the regulatory burden on business; and expand the tools to address criminal conduct through consent resolutions and corrective orders. In recent years, several important sectors of the Canadian economy have been affected by privatisation and deregulation, including transport, telecommunications, energy and financial markets. This trend has expanded the scope of the Competition Act as the only available lever to discipline anticompetitive conduct at a time when fiscal restraints put pressure on all government agencies to do more with less. At present, only the Director may launch proceedings before the Competition Tribunal in respect of civil reviewable matters. Opportunities may exist for improving the effectiveness of the law and enhancing compliance by allowing private parties to initiate proceedings. Further consideration should be given to the potential for such a mix of public and private enforcement.
1. Cost Recovery With continued fiscal restraint, the Bureau has had to search for methods of generating revenue. In this context, cost recovery for the Bureau's services is an attractive option. Proposals include partially recovering the costs of reviewing Advanced Ruling Certificate50 requests and merger prenotification filings, rendering advisory opinions, and providing copies of seized documents prior to statutory requirements for their return. Charging fees for our service will enable us to increase our resources, improve our service standards, and make the Bureau more accountable in the delivery of those services.
50
Sec. 102 of the Competition Act, R.S.C. 1985, Ch. C-34, states that "Where the Director is satisfied by a party or parties to a proposed transaction that he would not have sufficient grounds on which to apply to the Tribunal under Section 92, the Director may issue a certificate to the effect that he is so satisfied.'
106
Working Paper VI
2. Cooperation Among Firms Firms successfully competing in global markets must typically be highly innovative, keeping pace with technological change and other market developments. At the same time, the global market-place is increasingly characterised by interactions across borders, both between firms and within firms. These interactions involve all levels of activity, including research and technology development, design, production, sourcing and marketing. New methods of doing business are evolving, and we need to examine how best to ensure that competition legislation can distinguish between efficient responses to market developments and those which are clearly anticompetitive.51 Members of the business and legal communities have suggested that a review of the legislation may be needed to avoid discouraging firms from entering into beneficial forms of cooperation, which have no anticompetitive effect and are increasingly necessary in sophisticated global markets.
3. Enhanced Compliance Programmes Deterrence will continue to require selective litigation, but compliance programmes are highly cost effective in preventing anticompetitive behaviour. The Competition Bureau already has an active compliance programme, and will increasingly rely on compliance to fulfill its mandate.52 Fostering compliance with the Competition Act always has been, and will continue to be, the Bureau's overriding objective. We have an open door policy, under which parties can discuss proposed mergers and other business conduct in private. In fact, there is considerable saving in money and time to be gained by taking advantage of the Program of Advisory Opinions or by approaching the Bureau early in the merger planning process. The Bureau will be exploring ways to improve certainty for business and reduce anticompetitive conduct by expanding its compliance programmes.
51
F o r a fuller discussion, see Director of Investigation a n d Research, Strategic Alliances Under the Competition Act (Information Bulletin, 26 August 1994). T h e preface to the Bulletin observes that 'it is the Bureau's experience that most forms of strategic alliances d o n o t raise issues under the Competition Act': id. at 1. 52 T h e Competition Bureau expects t o release its C o r p o r a t e Compliance Bulletin in
the near future.
2 - Competition Policy Objectives
107
4. Intellectual Property The interface between competition policy and intellectual property rights has been receiving increasing attention from competition law authorities, including the Competition Bureau, as knowledge-based industries become increasingly important. It would appear that the goals of competition policy and the statutory power to exclude competitors through intellectual property rights are at variance. However, further analysis reveals that this divergence is not as great as it may initially appear. In a significant number of industries, intellectual property rights do not confer market power to the extent of raising concerns for antitrust enforcers. Moreover, in many instances, potential abuses involve extending intellectual property rights beyond their immediate domain, such as by incorporating tying arrangements or territorial restrictions to licensing arrangements, which are business practices subject to conventional antitrust prescriptions. In the long run, however, intellectual property rights may strengthen competition in the economy by enhancing innovation and providing incentives for the development and production of new products or methods of production.
5. International Cooperation International cooperation is part of the Bureau's comprehensive approach to a modern, effective and relevant competition policy. The 'nationality' of corporations has become more difficult to ascertain in the wake of far-reaching investment and capital market liberalisation. Hence, at present, the relative economic importance for the host country of national versus foreign firms is much less clear-cut than in the past. The Bureau recognises that globalisation and the emergence of large, 'stateless' corporations are placing new pressures on the international antitrust system. These pressures take many forms, such as transnational mergers and acquisitions which are reviewed by several antitrust jurisdictions; international strategic alliances and other forms of international cooperative arrangements; and international cartels which are sometimes promoted by trade measures such as voluntary export restraints. In September 1996, the Bureau and the U.S. Federal Trade Commission signed an agreement establishing a Canadian-U.S. Task Force on Cross Border Deceptive Marketing Practices. The primary purpose of the Task Force is to provide a framework to promote cooperation between law enforcement agencies in Canada and the U.S. with respect to deceptive marketing practices
108
Working Paper VI
with a transborder component. The Task Force will operate within the confines of the laws, policies and practices of each country. With the globalisation of business activity, advances in technology, and greater potential for international deceptive marketing practices, this type of proactive, cooperative approach to enforcement has become increasingly necessary. Deceptive telemarketing is an area with international dimensions where we will focus attention. In this regard, Canada and the U.S. recently announced the formation of a binational law enforcement group to combat cross-border telemarketing fraud. Law enforcement agencies are to report back by September 1997 on joint progress in fighting this type of crime. The impact of competition policy is increasingly felt across borders, with a corresponding increase in the interaction between antitrust agencies. Canada has been at the forefront of greater cooperation with its trading partners in antitrust law enforcement. The 1995 Canada-U.S. Agreement Regarding the Application of their Competition and Deceptive Marketing Practices Laws, and the Mutual Legal Assistance Treaty in Criminal Matters between Canada and the U.S., facilitate close cooperation between Canadian and U.S. competition law enforcement agencies, in circumstances where this is warranted. The Bureau believes in increasing cooperation among antitrust enforcement agencies. International cooperation will become more significant as competition policy comes under closer scrutiny in other multilateral fora, such as the W.T.O. A sustained and determined effort to cooperate in the enforcement of existing laws over the next few years is the best way to determine whether new multilateral rules are required, and what form they may take.
Conclusion The legislative changes embodied in the 1986 Competition Act have enabled the government to open the economy to greater competition and have facilitated the structural adjustment of Canadian industry to global change. The Act expressly recognises the role of foreign competition and the objective of enhancing Canadian competitiveness. The Act has fostered stronger competition and more intense rivalry in the Canadian market-place, making Canadian companies more efficient and better able to compete abroad. In the long run, an open, competitive environment will serve to promote productivity, growth, and a rising standard of living. The experience to date with respect to the enforcement of the Act lends support for the objectives of Canadian competition policy, as stated in the Act's purpose clause. The Supreme Court expressly recognised in the PANS case
2 - Competition Policy Objectives
109
that the Act is a central feature of Canadian public policy in the economic sector. The merger provisions and their enforcement are fundamentally economic in nature, and are designed to respond to the dynamic nature of a modern industrial economy. The cases brought by the Director and the case law indicate the flexibility and balance of the Act in dealing with anticompetitive abuses by dominant firms. The 'total welfare' approach to merger analysis is an example of the forward-looking, flexible approach needed for an open and small economy. In the global economy, competition authorities will pursue a growing number of cases in which cooperation and co-ordination will be necessary to ensure healthy and competitive markets. The challenge to antitrust enforcement in a global market-place will be to secure greater enforcement cooperation and harmonisation of competition policy. The work of the O.E.C.D. in the development of bilateral agreements has contributed to a better understanding of the need and merits of closer cooperation. The impact of the Act goes beyond the Bureau's enforcement activities. In most cases, businesses voluntarily design their practices, or alter existing arrangements, to ensure that they comply with the Act. Furthermore, the Bureau actively promotes voluntary compliance through a number of initiatives besides enforcement activities, notably programmes of information and public education. Finally, forums such as this workshop foster a deeper understanding of the issues facing competition authorities, improve international antitrust relations and thereby help us all to maximise our impact in our efforts to promote dynamic, competitive economies in our respective jurisdictions.
VII Damien Neven Professor, University of Lausanne Lausanne, Switzerland
Introduction This paper will focus on how to achieve the objective of economic efficiency (including how we should address possible tradeoffs between various aspects of economic efficiency) and on whether a conflict exists between economic efficiency and alternative objectives, with particular reference to market integration. Several more general comments willfirstbe made. Regarding the legitimacy of objectives, the adoption of a competition policy and associated legal and institutional framework seems to be a political act. What is legitimate is effectively determined by sovereign democratic institutions. Whenever a policy has been adopted by these institutions, its legitimacy cannot be questioned. A discussion of the relative merits of the various objectives, and of the possible tradeoffs between them may, nonetheless, be both useful and necessary to fuel the democratic debate about them. This paper adopts such a positive (rather than normative) approach. A distinction should be made between the objectives being pursued by the policy and the objectives that are assigned to the institutions in charge of implementing the policy, as reflected in the legal framework under which these institutions function. The objectives embodied in the legal framework may differ from ultimate objectives of the policy for at least two reasons. The first reason is that the constraints under which the institutions operate must be explicitly recognised. For instance, if an institution has limited resources, it may be wise to exclude a time-consuming task from its mandate. More generally, the political economy of such an institution should be considered; its mandate should be sufficiently limited so that one can be reasonably confident that the institution can fulfil it. For instance, if broad discretion in implementation could be abused, it may be desirable to formulate the institution's mandate in terms of simple rules. The consequence of doing so may be loss of precision in implementation with respect to the ultimate objective, but this could be less damaging than leaving implementation of general objectives open to capture by particular interests or by the implementing institution itself. Evaluation of efficiency benefits in merger cases illustrates this tradeoff: even if the efficiency benefits of mergers should be considered, pure economic efficiency may be better served if this consideration is excluded.
112
Working Paper VII
The second reason involves strategic interactions that may occur between the institution in charge and other agents directly concerned with the policy (principally the firms subject to the statute but also, for instance, the authors of complaints). The theory of delegation provides the important insight that a particular objective may best be achieved indirectly, by delegating responsibility to achieve it to an agent with a different objective. For instance, shareholders will often enjoy greater profits by choosing managers who maximise sales rather than profits. This is because competitors will usually find it most profitable to accommodate the behaviour of afirmled by an aggressive manager by reducing their own sales. For instance, assume that two firms merge, and competitors will be in the best position by accommodating a change of strategy by the merged entity, so that a reduction in output by the merged entity does not trigger an aggressive response by competitors. The incentive of the merged entity to reduce costs is, in turn, enhanced by the degree of rivalry in the product market. Under a rule based on economic efficiency, which takes account of the cost reduction achieved by the merger, the merger is allowed, because the welfare loss associated with the reduction in output after the merger is compensated by efficiency benefits. In contrast, under a rule based purely on aggregate market power, the merger would be allowed only if it is structured in such a way that rivalry in the output market is not affected. However, this enhances the incentives for the merged entity to invest in cost reduction. Accordingly, this alternative rule may lead to lower market power and lower costs, and thus higher aggregate efficiency, than the rule based directly on efficiency.
A. Competition and Economic Efficiency In this section, it is assumed that the ultimate objective of competition policy is economic efficiency. Following the approach discussed above, this section will consider what 'intermediate' objective could be assigned to the institution in charge of implementing the policy when the ultimate objective is economic efficiency, and how narrow that objective should be. The European legal framework provides a clear illustration of alternative options in this respect. On the one hand, the merger regulation focuses exclusively on competition, and at least formally disregards efficiency benefits. On the other hand, the structure of Art. 85 can be viewed to recognise a primary concern for competition (in Art. 85(1)) which can, pursuant to Art. 85(3), be balanced against considerations of productive efficiency (improvement in the production and distribution) and dynamic efficiency (technological progress).
2 - Competition Policy Objectives
113
An argument in favour of a narrow objective focusing on competition is that the evaluation of efficiency benefits is a difficult exercise, requiring broad discretion.1 Firms will always have an advantage over the competition agency in arguing their case for several reasons: they will have access to privileged information on cost issues, which is difficult to verify; and they are likely to have resources at their disposal to argue their case which far exceed those available to the monitoring agency. In addition, whenever an agency has broad discretion, accountability is difficult to implement institutionally, creating a higher risk of capture. The tradeoff between a narrow competition objective and a broader one that encompasses efficiency benefits will also hinge on the extent to which competition can be expected to promote economic efficiency. If the additional considerations that could be taken into account with broader objectives are relatively insignificant, then the case in favour of narrow objectives could be quite strong. Most economists and antitrust lawyers probably share the belief that competition enhances efficiency. This belief is based on the classical paradigm of perfect competition, which achieves Pareto efficiency, and the presumption that more competition will tend to move the industry closer to this benchmark.2 Workable competition is then viewed as the acceptable approximation to perfect competition that should be pursued. This belief should be refined in order to evaluate whether competition can reasonably be assigned as the sole objective of the agency in charge of implementation. This discussion is aided by the discussion which follows, clarifying what is meant by 'competition' and what is meant by economic 'efficiency'.
1. Productive rather than Allocative Efficiency Economic efficiency can be considered from several perspectives. Traditionally, a distinction is made between three aspects: productive efficiency (products are produced at least cost); allocative efficiency (appropriate quantities of output are produced); and dynamic efficiency (adequate investments are undertaken).
1
Damien Neven, et al., Merger in Daylight (1993). Indeed, one of the most popular models of imperfect competition, the Cournot model, has the property that as competition increases (as measured by the number of firms), prices converge to marginal cost. 2
114
Working Paper VII
Traditionally, allocative efficiency has been the focus of antitrust analysis (as reflected, for instance, in its emphasis on price distortions3). However, this emphasis is somewhat odd in light of the available empirical evidence regarding the magnitude of welfare losses associated with price distortions. Typically, estimated welfare losses do not exceed 2 per cent of GDP. This is not surprising, since the loss of consumer surplus which results when firms exercise market power is largely compensated for by profits. The net decrease in welfare (the Haberger triangles) is zero if demand is inelastic, and increases in line with the elasticity. Since elasticity of demand at the industry level is generally low (often below four), it is no surprise that net welfare losses are small. Regarding productive efficiency, the idea that market power may significantly raise the cost of production has been much explored in the last decade, both theoretically and empirically. It appears that the welfare losses associated with absence of productive efficiency will typically be much larger than the Haberger triangles of conventional theory. Regarding dynamic efficiency, important new insights have been developed in the theoretical literature, but the evidence is not as conclusive as that gathered with respect to productive efficiency. It appears, however, that at least in some circumstances, dynamic efficiency losses associated with lack of competition may far outweigh those associated with allocative efficiency. These developments underscore the need for a more precise definition of competition. The vague association between increase in competition and a move towards the state of perfect competition simply does not suffice to understand how competition affects productive and dynamic efficiency.
2. Competition is a Process and not a State In The New Palgrave, Stigler defines competition as 'a rivalry between individuals (or groups or nations) and it arises whenever two or more parties strive for something that all cannot obtain'.4 This definition is formulated in behavioural terms, in contrast with the traditional approach, under which competition is associated with a particular state of affairs which is characterised as perfect competition.5 Moreover, this definition is potentially quite broad, to the extent 3
See, e.g., Frederick M. Scherer & Daniel Ross, Industrial Market Structure and Economic Performance (1990). 4 George J. Stigler, Competition, in The New Palgrave (John Eatwell et al., eds. 1987). 5 It is paradoxical that little competition, in a behavioural sense, takes place in a perfectly competitive state.
2 - Competition Policy Objectives
115
that it encompasses many different instruments of rivalry (including both price and non-price instruments, such as advertising or R&D) and different objectives (including profits, efforts, sales or market shares). Vickers observes that saying 'more competition' exists in a given situation can have several meanings: (i) lower barriers to entry (and more generally lower constraints for rival firms), (ii) an increase in the number of rivals, (iii) a move away from collusion and (iv) larger stakes (namely greater rewards from achieving what firms are striving to achieve and larger penalties for failing to achieve it).6
3. Competition, Productive and Dynamic Efficiency The links between competition and productive and dynamic efficiency are as follows. There are at least three reasons that competition may enhance productive efficiency. First, to the extent that managers and workers can control the disposition of the fruits of market power, the presence of substantial rents creates a temptation to inflate production costs rather than to work hard for the sole purpose of handing these rents over to the firms' owners. As Hicks remarked, 'the best of all monopoly profits is a quiet life'. Thus, there is a risk that market power will translate into higher costs rather than higher profits. As rivalry increases, the scope for increasing costs will fall. Second, competition may improve incentives for efficiency by providing a benchmark against which performance can be measured (sometimes referred to as 'yardstick competition').7 Such comparison improves the precision with which performance can be assessed, and thereby allows for a more efficient contract between managers and firm owners. Third, competition has a selection effect: efficient firms gain larger market shares than inefficient ones, and some may be driven out of the industry altogether. These themes have been much developed in the recent theoretical literature. Not surprisingly, results are often ambiguous, proof of the positive link between competition and productive efficiency will rest, in part, on empirical evidence. The available evidence is, however, comforting, as it8 suggests that 6
John Vickers, Concepts of Competition, in Oxford Economic Papers (1995), 1, 4. For a detailed discussion of this issue, see id. 8 See, e.g., Steve Nickell, Competition and Corporate Performance, 104 J. Pol. Ec. 724 (1996), Donald Hay & Georges Liu, The Efficiency of Firms: What Difference does Competition Make, (Mimeo, Institute of Economics and Statistics, Oxford 1994); Damien Neven & Lars-Hendrik Roller, Competition and Rent Sharing in the Airline 7
116
Working Paper VII
when rivalry increases,firmsbecome more efficient. Moreover, it confirms that the effects are quantitatively large, and that the selection effect is present. The link between competition and dynamic efficiency is more delicate. The theoretical results are much less robust, and the empirical evidence is sparse, although it points in the right direction. For instance, Nickell finds that rivalry reduces the cost level, but enhances the rate of growth of productivity over time.9 Here again, the effects are quantitatively large. Until now, this paper has addressed the link between competition, productive and dynamic efficiency in isolation. It is worth asking whether a tradeoff is made between allocative, productive, and dynamic efficiency. Von Weizsacker argues that such a tradeoff is made at a very general level.10 He distinguishes three different levels of economic activity: consumption, production and innovation. He then argues that the efficient provision of activity at one level depends on the restriction of competition on another (where competition is defined as freedom of access). Incentives for production will require some restriction of competition at the level of consumption (since a firm has little incentive to produce if everyone can compete to consume its output). Such restrictions of competition take the form of property rights. Similarly, to provide incentives for innovation, some restriction of competition at the level of production will be required, which will typically take the form of intellectual property rights. Even if the terms of the tradeoff will presumably differ on a case-by-case basis, it is unclear whether competition law should consider these tradeoffs explicitly, given the difficulty of evaluating them. This approach may thus suggest that the balance that competition strikes between productive, allocative and dynamic efficiency should be determined essentially by property rights, independently of competition law. 4. A Tentative Conclusion The evidence tends to confirm that it may be reasonable to assign a narrow objective, formulated solely in terms of competition, to the institutions in charge of implementation. The presumption that competition will drive effiIndustry, 1996 European Economic Review 933; S. Olley & Adrian Pahes, The Dynamics of Productivity in the Telecommunications Equipment Industry (NBER Working Paper 3977, Cambridge, Mass. 1992); Paul Seabright & Francis McGowan, Deregulating European Airlines, 9 Economic Policy 282 (1989); Paul Geroski, Innovation, Technological Oportunity and Market Structure, 42 Oxford Economic Papers 586. 9 Nickell, supra note 8. 10 C. von Weizsacker, Barriers to Entry: A Theoretical Treatment (1980).
2 - Competition Policy Objectives
117
ciency sufficiently well so that possible divergence can be ignored has recently gained some credence. However, the foregoing discussion has also highlighted the need for a more refined definition of competition. What we mean by competition must be made more precise and substantially broader, not only in theory, but also in antitrust practice. Especially in the European context, the concepts of competition and its restrictions are currently very narrow11 (and biased toward the evaluation of allocative efficiency). Thus, it may make sense to eliminate Art. 85(3), but this will require a fundamental rethinking of Art. 85(1). B. Economic Efficiency and Market Integration This section abandons the economist's abstract world and explicitly recognises that objectives other than economic efficiency will be assigned by democratic institutions. In the context of the European Union, market integration is probably the most important objective, which should be expected to remain in place in the short to medium term, according to Mr. Schaub. The question then arises whether the goals of market integration and economic efficiency conflict with each other and, if so, whether the tradeoff is severe. The Community's market integration objective does not necessarily promote market integration! Indeed, as Korah has observed with respect to vertical restraints, the Commission's insistence on the absence of territorial restraints for the sake of market integration might have the effect of deterring the distribution of commodities across Member States, since such distribution is profitable only in the presence of territorial exclusivity. By insisting that a restriction on parallel imports is practically a per se restriction of competition, the Commission tends to confuse symptoms with causes. The existence of price differentials between Member States may well be a symptom of market power in those markets, but it is by no means clear that, conditional on the existence of market power, parallel trade enhances economic efficiency. Indeed, when firms sell across segmented markets, they charge different margins in different markets (and typically higher margins in domestic markets where the elasticity of lower). An increase in parallel imports will reduce the scope for discrimination, but it does not necessarily enhance efficiency12; domestic consumers will typically gain sales at the expense of 11
See Jonathan Faull's paper for this conference, ch. 8; Damien Neven et al., Trawling for Minnows, Agreement in EC Competition Law (CEPR 1998), for an illustration from the recent case law. 12 See e.g. the ex post internal exercise published by DG II, European Economy (1997).
118
Working Paper VII
consumers in export markets (where sales fall). Conditions under which less discrimination enhances efficiency are also rather stringent. Thus, there is not even a presumption that in most cases preventing arbitrage will increase efficiency.
Conclusion This paper has developed three arguments. First, a distinction should be drawn between the ultimate objectives of competition policy and the objectives assigned to the institutions in charge of implementing them; the latter should explicitly recognise the constraints under which the institution is operating, as well as its strategic interactions with other agents. Second, from the narrow perspective of economic efficiency, it may be reasonable to assign competition as the sole objective of a competition agency. This would however require a more refined definition of competition, and of what constitutes a restriction of competition. Third, the market integration goal is often in conflict with the economic efficiency goal. It may be advantageous to reduce its scope.
VIII Alexander Schaubx Director General for Competition European Commission Brussels, Belgium
Introduction The debate on what the objectives of competition policy ought to be is not new. Issues such as monopoly exploitation, the protection of the weak against the strong, and the compatibility of economic power with democracy have been debated over centuries, if not millennia. One might think of the Code of Hammurabi or the ancient ruling in Greece to prevent monopolistic exploitation. The subject is, indeed, much older than the first modern competition law. Moreover, it has not been the 'exclusive territory' of economics or law, but also has philosophical and political elements. What are the objectives of competition policy? The promotion of efficiency to the benefit of consumers, i.e. economic objectives? The support of pluralism and democracy, as well as the protection of small and medium sized enterprises? Or is competition an end in itself which must be protected without looking further towards ultimate objectives? Normally these issues are debated when a political decision on the adoption of a competition law needs to be taken. One may ask why we are having this debate now, since we are not in the middle of adopting a completely new law, and the decision on the basic principles has already been made. Could there be disagreement on what the fundamental objectives should be? And, if so, does such a disagreement cause significant divergence in competition law enforcement? Moreover, what do we imply when we talk of competition policy? Do we mean classical antitrust exclusively or do we go beyond this subject? In other words, do we agree that a consistent competition policy should include, or at least be in line with, other fields such as state aid control and deregulation of public monopolies in order to achieve our objectives? Being an optimist, I hesitate to believe that there are fundamental discrepancies. This would be a serious problem, not only for a coherent competition policy within the Community, but also for further enlargement and international 1
I wish to express particular gratitude to Elke Graper of the Directorate General for Competition, who made an essential contribution to the preparation of this paper. Opinions expressed are personal.
120
Working Paper VIII
cooperation in general. Thus, I tend to believe that we agree on the basic principles, and that a reflection on our objectives will help us to shape competition policy for the twenty-first century. It will also help us to carry out specific projects, such as the Commission's Green Paper on vertical restraints, the F.T.C.'s Joint Venture Project, and the German reform of competition law. Competition is the key element of our economic system. It forces companies to run a tight ship, to adjust to changes and to innovate. It thereby benefits the consumer and promotes the welfare of society in general. This is why we protect competition, in Europe and abroad. In the Community, competition has one additional fundamental objective: to promote market integration. In this context, it is particularly important to remember that Community competition law not only protects competition against private action, but also against governmental measures.
A. Objectives of Competition Policy—Two Levels of Discussion Before entering the issue of objectives, one should always distinguish two levels of the debate: the level of political decision making and the level of law enforcement. This distinction is important because thefirstlevel concerns more fundamental or ultimate goals, while the second concerns operational criteria for enforcement practice.
1. Political Decision Making A debate on objectives in the context of a political decision, such as the adoption of a competition law, focuses on fundamental or ultimate objectives. These ultimate objectives are economic prosperity or the welfare of citizens, employment and social cohesion. A decision must be made whether these are better achieved with or without protecting competition. For instance, if the opinion prevails that the market mechanism should function without interference, adoption of a competition law does not make sense. The assumption that major players with market power are good for employment and social cohesion, as well as the acceptance of'freedom to act' as the ultimate aim, including the freedom to form cartels, also support this conclusion. On the other hand, if competition is viewed as the genuine element of an open market economy which must be protected against cartels, market power etc., then a law protecting effective competition and controlling market power does make sense.
2 - Competition Policy Objectives
121
At the level of political decision making, economic arguments, as well as political, social and philosophical arguments, are considered. For instance, one may debate whether competition rules could also be an instrument to serve pluralism and democracy, and whether the provisions should be limited to protecting the efficient functioning of markets (by means of the control of market power, prohibition of cartels etc.) or extended to controling economic power (e.g. including divestiture, control of size). Social issues such as the possible conflict between competition and employment in the short term, may also be debated.
2. Law Enforcement Once agreement on the ultimate objectives is reached, it would then be necessary to translate them into provisions and criteria for law enforcement. For instance, if protection of effective competition as an instrument for the efficient functioning of markets were accepted as the ultimate goal, provisions such as the prohibition of cartels or the control of market power are the likely means of achieving outcome. Operational criteria for enforcement must be established, as well as the means to achieve the right balance between 'as much market mechanism as possible, but interference only where necessary'. One cannot go back to social or political considerations as direct criteria for law enforcement because it would change the basic concept, which is the—indirect— achievement of fundamental goals via the mechanism of competition. Job creation or security cannot be a criterion for law enforcement because, in the long run, this could have exactly the opposite effect or, even worse, as several state aid cases show. Moreover, it would change the basic concept of our economic system and move in the direction of economic planning. A debate on provisions and enforcement criteria may, however, include the classical question whether market structure, conduct, or performance should be the main test. Once protection of competition as a process and as an indirect means to achieving a proper performance is accepted, performance as a test can almost be excluded. Furthermore, a forecast as to market performance can hardly be carried out in an individual case. In practice, however, one does not follow a pure and ideal concept, but a 'mix'. Thus, most competition laws have both control of structure and of conduct as criteria; even rules related to market performance are used (prohibition of exploitative abuses, efficiency defences etc.).
122
Working Paper VIII
B. A Comprehensive Approach to Community Competition Policy Community competition policy, as established in the Treaty of Rome, protects competition as a means of achieving our ultimate objectives, such as establishing a common market, promoting a high degree of convergence of economic performance, and raising the standard of living and quality of life.2 Ensuring that 'competition in the internal market is not distorted'3 is the intermediate or direct objective of competition policy, which also is reflected in the competition rules. Prosperity, employment and social cohesion are the ultimate objectives, which are the reasons that the political decision for a Community competition policy has been taken. This is clearly set out in Art. 3a of the Treaty, which stipulates that economic policy is, among others, conducted in accordance with the principle of an open market economy and free competition. How are these basic principles translated into provisions of law? Classical antitrust or 'competition rules for undertakings', as we call it in the E.U., constitutes only one of the areas covered. However, control of private action is not enough. Governmental measures such as subsidising industry or protecting rigid market structures by regulation can also cause a substantial degree of competition distortion. The best antitrust law can accomplish little if vast amounts of state aids are handed out to ailing industries. Why should we prevent the creation of market dominance by means of merger control if, at the same time, some monopolies are protected by regulation? We do not promote dynamism and we do not prevent the existence of the 'lazy monopolist' by antitrust alone. That is the reason that Community competition policy covers rules for Member States as well. The Commission is the only authority that can adequately control or terminate these governmental measures, even if it is not an easy task. It would not make sense to leave it to the Member States to control themselves, or simply to call for discipline. As the most recent state aid report has shown, the Member States exercise only limited discipline: particularly the richer ones have increased aid to industry during the period from 1992 to 1994 to an average level of 43 billion ECU per year. Problems with deregulation measures in certain sectors, such as postal services, is another example. Within the Member States, one may ask how antitrust enforcement coexists with these interventionist measures. For instance, Germany is well known for the strict enforcement policy of the Bundeskartellamt. However, Germany is generous with state aid, which is partially explainable by the restructuring of east German industry. Each job in the coal mines is subsidised with more than 2 3
See Treaty of Rome, Art. 2. Id., Art. 3g.
2 - Competition Policy Objectives
123
50.000 ECU, and further large amounts of aid have gone to other ailing industries such as shipbuilding. In addition, Germany is not a leader with respect to liberalisation measures. Thus, I question whether Germany has a coherent overall policy towards competition. Are there different objectives with respect to antitrust and other policy fields concerning competition? Or does the division of tasks between different authorities lead to these seemingly different approaches?
C. Coherence Between Different Policies The Treaty requires that the variousfieldsof economic policy be coherent. For example, measures to promote the competitiveness of European industry must be in accordance with a system of open and competitive markets.4 Consequently, subsidising specific industries or other interventionist measures cannot be the Community policy to promote competitiveness. Rather, the Community has the task of creating the essential conditions for the development of an efficient industry and providing a stable economic environment. In that sense, competition and competitiveness policy serve the same objectives; competition is a key element for achieving competitiveness. The internal market is another example of coherence of the various policyfields.Opening markets by means of public procurement directives or other measures increases competition, and therefore complements competition policy. Moreover, market integration creates essential conditions to promote dynamism and the efficiency of industry. Critics of the Community's approach argue that requiring coherence between different areas of policy leads to political interference with competition law enforcement. This criticism reflects a deep misunderstanding. Competition and other fields of policy are interdependent. Competition will bring about an environment that creates competitive industries only if the different policy fields are based on a coherent concept. In contrast, in a country where state aid to industry is an important aspect of economic policy, competition policy is in conflict with industrial policy, and the maintenance of effective competition would not seem to be the overall objective of economic policy.
D. Competition as an Objective in Itself Competition is not an 'ultimate' objective in itself, but an instrument or intermediate objective to achieve economic prosperity and employment. It is 4
Id, Art. 130.
124
Working Paper VIII
generally agreed that the welfare of our citizens is the ultimate objective. It would make little sense to have a policy and law protecting competition if we did not believe that it would help us to achieve this goal. Thus, competition is always an instrument or intermediate objective, though a fundamental one for the functioning of markets and for the achievement of these ultimate goals. Whether competition is an objective in itself was debated in Germany during the 1960s, and has recently resurfaced in the context of the German 'harmonisation debate', i.e. the envisaged reform of German competition law. Reports of the German Monopolkommission and the Advisory Council to the Ministry of Economics state that in the Community, competition policy is regarded only as an instrument for the fundamental objectives of the Treaty, whereas in Germany, competition policy is an objective in itself. Accordingly, German policy protects the freedom to compete (Wettbewerbsfreiheit). The logic of these arguments is questionable. The motivation for having a competition policy in Germany must also be economic prosperity. Indeed, the success of German competition policy has often been described as one reason for the smooth functioning ofsoziale Marktwirtschaft and for the Wirtschaftswunder. Thus, competition is not protected for its own sake, but helps to achieve these fundamental objectives. The meaning of the term Wettbewerbsfreiheit (freedom to compete) is elusive. Does any restriction to this freedom by means of cooperation or distribution contracts put effective competition at risk? Some restrictions may, but what is decisive is the effect on the market in view of market structure and the strength of the partners. Consequently, the term 'freedom' provides little guidance, as the once heated debates in Germany and the U.S. some decades ago showed. These debates occurred because competition policy and theory partly have their origins in free trade laws and the general liberalisation of professions/trade about 200 years ago. However, this historical background cannot serve as an interpretative tool for today's legal terms and economic reality. Modern competition theory and case law have provided better tools based on the assumption that the protection of a system of workable competition is at issue. In this context, restrictions on the freedom to compete are not always relevant. Moreover, a restriction of competition may not be caused by a restriction of freedom to compete. Rather, it may be the inherent effect of cooperation between competitors which cannot be characterised as a restriction of freedom (e.g. joint selling or buying without exclusivity, information exchanges, inherent effects of joint ventures/minority shareholdings between competitors). The subject appears to be a mere philosophical debate. Whether competition is defined as an end in itself or as a means to an end does not seem to result
2 - Competition Policy Objectives
125
in significantly different enforcement practice. It does not seem decisive, either for the mix of provisions (i.e. structural and behavioural control) or for application of the law in individual cases. For instance, it is difficult to detect differences in the results of enforcement practice in Germany, the Community and the U.S. In all three laws, merger control plays an important role, and cartels and the abuse of market power are prohibited. Differences exist with respect to details. For instance, Germany, which defines in general the freedom to compete as the major issue, is, with respect to vertical restraints, less strict than the Community, which must take account of the market integration goal. In contrast to both, the U.S. places greater emphasis on efficiency justifications, as the new U.S. guidelines demonstrate in the area of merger analysis. This may be explained by their lower threshold for the identification oTprima facie cases. However, until now, efficiency defences have not played an important role in U.S. enforcement practice and, under the carefully drafted new guidelines, it is doubtful that they will in the future. It appears that differences occur mainly at the 'philosophical level', but are often not significant in practice. Thus, seemingly different objectives can lead to similar rules and case handling, as a comparison between Germany and the Community demonstrates. On the other hand, similar objectives and even the same set of rules can result in substantial differences with respect to enforcement. Thus, in the 1970s and 1980s, the Chicago School defined efficiencies to the benefit of consumers as the main objective, but concluded that cartels, but not mergers, could threaten achievement of this objective. Consequently, merger control in the U.S. was defacto non-existent in this period, although the rules basically did not change.
E. Economic, Social or Political Objectives Adoption of a competition law is a political act. Thus, a debate on objectives cannot be limited to economic arguments. However, economic reasoning dominates the debate, since competition and economic power are at the heart of economic life. Social and political considerations also play an important role. For instance,fiercepolitical debate can occur because maintenance of competition in the short run may conflict with the objectives of employment and social cohesion. If short term thinking dominates, the adoption of a competition law might be put into question altogether. This may occur, for instance, in regions facing structural unemployment where political campaigning occurs. If, however, competition is viewed as a mechanism essential to achievement of fundamental long term objectives, such issues should not be considered.
126
Working Paper VIII
Basing enforcement on social criteria would contradict the basic concept of competition, by attempting directly to achieve ultimate objectives through application of competition rules. Moreover, political, social or moral criteria are not operational, and may be applied arbitrarily and subjectively. What is fair? What is social justice? Where is the borderline between 'the weak and the strong'? We must bear in mind that competition is the essential mechanism of our economic system, and we must protect the process of competition. Competitors may thereby be protected, and in that sense, we also protect the 'weak against the strong'or help small and medium enterprises. Economic criteria and a proper economic analysis are essential in competition law enforcement. In case evaluation, however, this does not mean that we should focus exclusively on the 'microeconomic paradigm'. Microeconomic models are only one tool to express certain concepts. However, they cannot sufficiently take account of all important considerations, and they are not 'the absolute truth', as the various trends and schools have shown over the decades. Competition policy and enforcement criteria should not be overloaded with too many objectives, because this would render competition policy ineffective. Competition is essential, but many types of problems are not resolved through competition. In certain exceptional circumstances (e.g. training/education, basic research, environmental polution), measures such as state aid, regulation, taxation etc. can be more appropriate than a mere market mechanism.
F. Market Integration—Both a Political and an Economic Objective Market integration is both a political and an economic objective of Community competition policy. It is a legitimate objective for two reasons: first, it increases competition; second, effective competition promotes and facilitates market integration. Although it is a rather unique objective among antitrust authorities, national authorities are also concerned that national markets are not artificially divided by agreements between companies. In the Community, market integration has not only been an 'ultimate' or indirect objective of competition policy, but also a direct enforcement criterion, as absolute territorial protection is prohibited. This is an exception to the general principle that competition is maintained as an indirect measure to achieve ultimate objectives. It may even be criticised as a violation of the general rule, because it mixes enforcement criteria with ultimate objectives. In the current timeframe, markets have been largely integrated. This raises the question whether we should continue to use competition rules as a direct mechanism to ensure market integration, or whether we can now rely on the
2 - Competition Policy Objectives
127
indirect mechanism, i.e. protecting the process of effective competition in order to create market structures which contribute to the objective of market integration. The latter would be more in line with the general principle. However, I believe that despite the progress towards the realisation of the internal market, a direct signal to companies is still required for its psychological and pedagogical effects, at least for the next years.
G. Different Economic Situations—Different Approaches to Competition Different countries have different economic environments. Eastern Europe is in a different situation from Western Europe, and developing countries face yet another set of problems. These specific circumstances must be recognised by laws concerning the economy. If, however, these countries follow a similar economic approach—an economy based on the market system and thereby on effective competition—the basic concept of competition policy should be the same. Without this, a level playing field will not exist, and cooperation between competition authorities will be impossible, be it within the E.U., between the E.U. and C.E.E. countries, or bilaterally or multilaterally with other countries. Therefore, we encourage harmonisation of competition laws within the Community, as well as the direct application of Community rules by Member States. We are also working to ensure approximation of legislation with respect to the C.E.E.C. countries. Applying the same principles and basic concepts does not mean having exactly the same set of enforcement rules. We do not even have this within the Community. It also does not mean that the priorities must be the same. For instance, a country just developing a market economy may focus on concrete liberalisation measures; in contrast, in the U.S., merger control is one of the cornerstones. In the Community, a merger control system would have been less important in the early years as compared to the present, where markets have become much more integrated and cross-border operations are frequent. A competition law must be flexible enough to address specific circumstances, but it is not a tool to solve all types of problems. If industry is already characterised by monopolies or conglomerates, divestiture measures should be considered before establishing competition rules, such as a per se prohibition of diversification which may affect dynamism and effective competition in general. If big drug dealer rings use reselling below cost as a measure to enter markets and thereby launder money, it is doubtful whether a competition rule
128
Working Paper VIII
prohibiting sales below cost in general is an appropriate measure. First, it will not solve the problem of laundering money because dealer rings will find other means of entering, such as takeovers. Second, such a measure could lead to rigid structures at retail level and less competition. Direct measures against drug dealers, subsidising farmers, etc., appear to be better methods.
Conclusion The main points argued in this paper are as follows. First, the ultimate objective of competition policy is the economic welfare of society. This includes the objective of low prices and high quality for the consumer, as well as dynamism and job creation of industry. Community competition policy also has the objective of market integration. Second, competition is not an end in itself, but an indirect means to achieve ultimate goals. Third, competition policy and enforcement are not appropriate instruments to achieve ultimate objectives such as prosperity and job creation directly. On the contrary, social objectives may conflict with the maintenance of competition in the short run. Therefore, the maintenance of effective competition must be the direct objective of enforcement. To achieve this objective, indications such as market strength or concentration ratios must be employed. Fourth, different countries may have different competition rules with respect to particular aspects. However, the basic concept should be the same. Otherwise we cannot reasonably cooperate, and a level playing field will not be achieved.
IX Dieter Wolf President of the Bundeskartellamt Berlin, Germany
Introduction This paper deals with the question of what competition, competition law and competition policy should or even can achieve. The subject induces me to be— moderately— provocative: Is competition merely an instrument or a means of achieving any number of aims? Is competition the economists' magic wand which alone brings efficiency, but does not produce all the expected results and therefore has to be guided by courageous politicians who are of course more creative than the merely mechanistic market process? Can competition be compared to a computer which is condemned to stupidity until politicians of all fields of policy and party affiliation have programmed it—or made it intelligent? Can competition then also be programmed for the future? And if competition is programmable, could it not take on a few more objectives than just efficiency alone? One could also set it goals such as achieving full employment or monetary and social policy aims. And since competition, when introduced, is like the hypothetical machine that once set going continues for ever, one could assume that it will not break down under the huge burden! It might be no surprise to learn that the president of an antitrust agency thinks otherwise. In this paper I will put forward an argument in favour of the primacy of competition and argue that competition policy is not merely one among many policies of equal rank. What is the case for a theory of the primacy of competition? I am from a country which had to start from scratch when I was very young. Not only did Germany lie in ruins in the literal sense of the word, she also had to cope with the ruins of a totalitarian political and economic regime. We were very fortunate in having the opportunity to start anew in political as well as economic terms. A decisive factor for this new beginning was the teachings of the members of the Freiburg School. Their central theory is that the political and economic systems are interdependent. That is to say, whoever desires political freedom must also allow economic freedom and vice versa since the freedom of economic activity is a vital constituent of the basic rights of personal and political freedom. Or in other words, whoever enjoys democratic freedoms in business life will assert the same rights in the political sphere—a fact which the rulers of the socialist "reforming" countries had to learn, sometimes the hard way, in the 1980s.
130
Working Paper IX
The consequence of this is that free societies are always interrelated through open markets, and thus also through the basic steering principle of competition. The interdependence of economic and social freedom is the decisive argument in support of the view that competition policy in Europe and elsewhere is not merely one among many policies of equal importance but should have priority. The advocacy of the primacy of competition thus results from reflections on how individual freedom and economic freedom interact. However, that does not mean that this evaluation is merely hypothetical. On the contrary, it is of great practical relevance. This may be illustrated by considering three issues: the relationship between competition policy and the Community's goal of achieving the single market; whether competition policy should concern itself with economic efficiency; the relevance of non-competitive objectives to this concept?
A. The Relationship Between Competition Policy and the Single Market First, the relationship between competition policy on the one hand and the goal of achieving the single market and the integration of national markets on the other has to be considered. Without doubt the integration of markets, the removal of government barriers to trade and of customs duties constituted an ambitious goal which greatly intensified competition. In this respect, competition policy and single market policy shared a common aim and were in synch with each other. But, this does not rule out the fact that, from the perspective of market integration, competition policy then had to be interpreted differently from now. For example, the ban on restrictive vertical agreements in Art. 85(1) of the E.C. Treaty made good sense at that early stage of integration, since distribution agreements are an excellent means of sealing off or allocating markets, thereby thwarting any efforts at integration. Now that the single market has been legally completed and has become an economic reality in many areas, the dangers of vertical restraints of competition have diminished considerably. Thus, it is only logical that the Commission considers in the pertinent Green Paper whether the scope of Art. 85 of the E.C. Treaty should remain as broad as it used to be. At this early stage, competition policy was not subordinated to the goal of integration. Rather, the two policies were complementary. Therefore, as a logical consequence, both policies have been pursued consistently by the Commission. A different situation has arisen now that the single market has
2 - Competition Policy Objectives
131
become a reality. It is time for the emancipation of competition policy. The goal of integration has largely been reached, but the goal of competition can only be reached in an on-going process.
B. Competition Policy and Economic Efficiency As far as the efficiency question is concerned, the basic assumption is of course that competition itself is the best means of ensuring efficiency. However, in particular cases, restraints of competition may have efficiency-enhancing effects. For instance, cartel agreements are generally banned, irrespective of the effects they have on the market. But certain forms of cooperation between firms are efficiency-enhancing. Thus, we also have to provide for exemptions to ensure that the possibility of efficiency-enhancing cooperation between companies is not ruled out altogether. This is the purpose of Art. 85(3) of the E.C. Treaty or the German provisions for instance on rationalisation cartels. The crucial point, however, is that such agreements must not eliminate competition with respect to the products in question. Things are different as far as merger control is concerned. Under German and European law, mergers are allowed in principle. Companies may fully realise merger-induced efficiency gains. A merger has to be prohibited only if a market-dominating position is likely to be created or strengthened, in other words if competition is in danger of being eliminated completely. This leaves no room for efficiency considerations, because nothing can justify relinquishing competition. The advantages of dynamic competition as a discovery process are far too great to be given up for static efficiency considerations as they are laid down in the famous Williamson trade-off.
C. The Relevance of Non-Competition Objectives Another question is to what extent economic and social policy goals should play a part in competition analysis. To put it bluntly, I would say, not at all. A clear distinction must be made between competition as a system and the results of competition. Protecting competition as a system does not mean having to accept the economic results of the competitive process at all times. Correcting the distributional results of the market process in favour of the poor is in fact the "social" ingredient of the market economy. But: Social considerations ought to be taken into account only after competition has been able to perform its allocative function. But I do not wish to advocate that competition should
132
Working Paper I
take place in a vacuum. Considerations other than competition concerns can be legitimate, and therefore should be included. In particular cases, it may indeed be appropriate that overriding public interest arguments prevail over the competition principle. But the public should be well informed when this occurs. It is only the public discussion that makes sure than industrial or social policy concerns remain transparent and are not used under the guise of apparently competition-based arguments. If competition is to take second place to other goals, somebody must clearly assume the political responsibility for this decision. If the decision-making process in the context of competition law is not transparent, the cost of political intervention is no longer identifiable.
Conclusion In conclusion I wish to address the question as to whether competition policy and competition law ought to be, or even should be, similar in all countries. If there are in fact legitimate objectives of competition policy all over the world, one could expect to find similar laws all over the world that serve to reach these goals. But the situation is not that easy. The pillars on which modern competition laws are based should be similar in all cases, but the details of the legislation must depend on the economic situation prevailing in the country concerned. In a highly concentrated and integrated economy, merger control is likely to be the prime concern. In contast, in an economy characterised by small and medium-sized firms the focus will be on the ban on discrimination in order to protect small enterprises from an abuse of market power by dominant firms. This is the underlying reason why it is more important to establish a competition culture than to place particular emphasis on the wording of laws. Competition has no lobby, and one of the foremost tasks of competition policy is to make people aware of the competition principle. As a German I know what I am talking about—the German economy was dominated by cartels and large industrial groups in the first half of the century. There was no competition awareness whatsoever. Today, Bundeskartellamt decisions generally enjoy public backing. Securing freedom was at the forefront of German competition policy after the war and its current standing is the result of a long process and hard work. Persistence in enforcing the law is probably the most important aspect of establishing a competition culture.
PANEL DISCUSSION
COMPETITION POLICY OBJECTIVES IN THE CONTEXT OF A MULTILATERAL COMPETITION CODE
GENERAL RAPPORTEUR:
Eleanor Fox, Professor, New York University School of Law, New York, New York, U.S.A.
PARTICIPANTS:
Roderick Abbott, Head of Delegation of European Commission, Geneva, Switzerland Jacques Bourgeois, Professor; Partner, Akin Gump Strauss Hauer & Feld, Brussels, Belgium Ulrich Immenga, Professor, University of Gottingen, Gottingen, Germany R. Shyam Khemani, Group Manager, Competition and Strategy Group, World Bank, Washington, DC, U.S.A. Connie Robinson, Deputy Assistant Attorney General, Antitrust Division, United States Department of Justice, Washington, DC, U.S.A. Mitsuo Matsushita, Professor; Member, Appellate Body, World Trade Organisation, Geneva, Switzerland
134
Petros Mavroidis, Professor, Universite de Neuchatel, Neuchatel, Switzerland Frangois Souty, Conseil de la Concurrence, Paris, France
Panel Discussion
3 - Competition Policy Objectives in a Multilateral Competiton Code
135
• PROF. FOX—This session addresses objectives of competition policy in view of the prospect of internationalisation of antitrust . . . or one might say it is about internationalisation of antitrust in view of the divergence of objectives of antitrust policy in various countries. Would the divergence in antitrust goals of various countries undercut any initiative to internationalise antitrust? The answer to this question is a function of what we mean by internationalisation. Three possible concepts of internationalisation are: (1) at one extreme, more and better cooperation of the antitrust agencies, more positive comity, and possibly inclusion of competition policy in various sectoral trade agreements; (2) at quite the other extreme, a detailed international code; (3) the middle ground, a world antitrust policy linked with liberal trade policy, which would address private restraints of trade that bar access to markets, along with public restraints of trade that are now already governed by the W.T.O. system. This last option might involve an international instrument linking trade and competition systems in some way, and covering mainly procedural safeguards such as rights of foreign nations to bring actions in harming nations, transparency, non-discrimination and national treatment, and choice of law rules. That type of agreement would present fewer problems than a full international code, because it would not be necessary to mediate among diverse objectives. I shall summarise some of the positions of the panellists in their papers on the question whether differences in values impacts upon the effort to internationalise. Interestingly, the panellists who were more sympathetic to internationalisation tended to believe that the disparity of goals does not present a problem, while those who were less sympathetic towards internationalisation tended to believe that disparity of goals was a real problem. Several panellists expressed specific ideas about the diversity of values. Mr. Khemani's paper sets out a possible scheme for linking a competition system with world restraints, and states that substantive harmonisation is neither necessary nor realistic. Prof. Immenga believes that clear objectives can be achieved by an international competition policy, and that recognising the imperative of internationalisation does not require an initial analysis of national differences. Prof. Bourgeois states that nations that have different objectives on various policy issues commonly come to agreement on certain rules that they have a reason to support, although their reasons may differ. Similarly, Mr. Souty suggests that some rules seem to be broadly accepted among trading partners, and that we should begin with an international law against cartels. Mr. Klein is greatly sceptical of the international project, because he is concerned that differences in antitrust values will undermine antitrust, causing movement towards a lowest common denominator, and perhaps forcing those with stronger antitrust rules to accept weaker, fairness rules. Prof. Matsushita
136
Panel Discussion
suggested that we should begin thinking about trade-linked concepts for internationalisation, and that we should take into account the goals, values and voices of less developed countries. Prof. Matsushita's perspective seems to feed into the concerns of Mr. Klein, who is concerned that those who believe in strong competition rules might be pushed towards weak rules by countries that want protection against efficient multinational companies that might come to invade the economies of the former. Most of those who wrote on the issue felt that any conflicts among nations in antitrust values would not interfere with internationalisation for the following reasons. Prof. Immenga argued that certain antitrust principles are likely to develop in an international environment that are consistent with the goals of various nations. For example, nations that want to be part of a world trading system may agree that they must prohibit anticompetitive restraints that block market access. Differences among nations may also be addressed by allowing the countries themselves t o formulate whatever they believe is an anticompetitive restraint impairing market access, as long as it is a credible antitrust principle. Mr. Abbot, coming from the trade side, stated that we in the antitrust community are always discussing harmonisation, or cooperation through bilateral agreements, and that is not what the trade community had in mind when they sought progress in international rules related to trade and competition policy. He proposed that we should discuss the values of competition policy versus the values of trade policy, and whether and where there is convergence of liberal trade principles and competition principles regarding market access restraints. Mr. Klein discussed the progress that has been made through bilaterals. Prof. Mavroidis asked whether we really want and need internationalisation, and whether negative spillovers impel us to create international rules. Prof. Mavroidis' paper provided a sceptical a counterpart to Mr. Khemani's paper, which also addressed negative spillovers and how they should be handled in an international environment. To conclude, I shall frame some questions for discussion. First, to what extent do differences undermine the effort for internationalisation? Many possibilities for internationalisation exist, both substantive and procedural. We should focus on anticompetitive blockage of market access and, maybe more generally, cartels. In this regard, does it matter whether the form of cooperation is an agreement that incorporates subsidiarity principals (natural law with non-discrimination and choice of law rules), as opposed to a complete code? Second, do differences in goals and in cultures, such as the difference between statism versus liberalism, undermine the effort for internationalisation? This is particularly important because internationalisation could involve
3 - Competition Policy Objectives in a Multilateral Competiton Code
137
disciplines on governments not to allow export cartels or perhaps require governments to redraft antitrust laws so that they recognise harms caused by their firms anywhere in the world, rather than just consumer harms in their own territory. Finally, should we have internationalisation? • MR. ABBOT—This panel should consider competition policy objectives in the context of a multilateral code, but we must first define the sort of code we are considering. Trade policy advocates are concerned with how competition principles dealing with certain market access restraints reinforce the disciplines on governments imposed by trade policy. Internationalisation of antitrust is something else, and covers cartels, boycotts, and exclusive agreements. For instance, the proposed Boeing-McDonnell Douglas merger may present a problem, because it goes beyond a single jurisdiction. However, that is not what trade policy advocates are concerned with. Moreover, it may not even present a problem, because it can probably be addressed through existing and future bilateral agreements. Such agreements could be developed into a network that will address jurisdiction problems, especially with respect to mergers and dominant positions. Any multilateral agreement would have to be much simpler. It is worth spelling out what the E.U. now seeks at the multilateral level, within the W.T.O., because it is really quite modest, compared with the position taken by Sir Leon Brittan in a 1992 speech, calling for a complete framework and an international authority. The E.U. seeks a multilateral accord that countries should be encouraged to adopt a competition law and a proper means of enforcing it which would promote free trade principles and free competition. This should reflect some common principles, especially in areas like cartels and boycotts, but not in more complex areas. The areas where the greatest degree of consensus exists should be identified for this purpose. The E.U. also seeks a government commitment that a competition authority would investigate a complaint by another country of a practice within the jurisdiction of that competition authority. Thereafter, the competition authority would prepare a report, on the basis of which some action might be taken by that authority. That is a fairly modest agenda, but even that is going to take a long time, because only 40 or 50 countries presently have any form of competition law. To conclude, trade policy advocates are concerned about cartels, boycotts and exclusive agreements of various kinds that result in market access restraints. I believe these should be regarded as illegal per se because they block trade. • PROF. BOURGEOIS—An initial agreement on international competition rules
138
Panel Discussion
should include consultation and cooperation obligations. Such rules should not, however, attempt to pursue a consensus on objectives because this would kill the negotiation. There are many examples of international negotiations where states could agree on rules without agreeing on the objectives that they pursue. For instance, in the negotiations on a subsidies code, there were two distinct views on the objectives to be pursued, the U.S. view and the E.U. view. Nonetheless, an agreement was reached on concrete rules. Second, a consensus should be built based on the nuisance value of the absence of common rules, that is, adverse or counterproductive effects of conflicting requirements. For instance, conflicting requirements have resulted when one country has cleared a merger, but another country has not cleared the same merger. The practical result is that the merger does not go through. Consequently, these conflicting requirements put into question the efficiency of each nation's competition system, and the interests of the companies that a given competition system is supposed to protect. Third, the purpose of international rules should be to prevent anticompetitive conduct that restricts international trade. This may not be the optimal solution, but it is, at least, a feasible solution. International rules preventing governments from imposing quantitative restrictions on imports and exports already exist. Agreement on similar rules applying to private companies does not seem like a very large step. Finally, two sets of rules are needed: one applying to conduct that significantly affects international trade, and a second applying to conduct not affecting international trade, that continues to occur on a national level. This would avoid the danger referred to by Mr. Klein, that international antitrust rules may weaken national rules. • PROF. IMMENGA—Divergences in national competition law are not an important obstacle to achieve an international understanding. Four categories of differences may be identified: those related to competition policy; those related to the impact of politics on competition policy; those related to national enforcement practices and institutions; and those related to cultures and traditions. These differences are an impediment, but they are not insurmountable. They can be overcome by considering the objectives of internationalisation, and balancing those obligations against national differences. Achieving the ideal of an international code does not require an initial analysis of national differences. A general understanding of the need for internationalisation of competition policies, which goes beyond the general need to maintain competition, may be achieved. On the basis of such understanding, a discussion of the substance of an agreement will be possible. Only then, will national differences become relevant.
3 - Competition Policy Objectives in a Multilateral Competiton Code
139
A major argument in favour of internationalisation was underscored by the success of the Uruguay round, which substantially set aside state barriers to cross border trade. This process of trade liberalisation will continue, as confirmed by the recent W.T.O. agreement to liberalise the telecommunications sector. However, absent internationalisation of competition laws, private firms will re-establish trade barriers through agreements or through the use of economic power. Thus, internationalisation is inevitable. By its nature, this argument goes beyond national divergences. Furthermore, the linkage of trade and competition policy, as for instance in thefieldof market access and antidumping, supports creation of an international competition regime. Finally, arguments regarding the need for a level playing field for global players are not closely related to national competition policy objectives. If we could agree on international objectives, an understanding on substance might be achieved, and the national differences might be overcome. An international code might be useful, but we cannot start with a comprehensive code, such as that presented by the Munich group. In the end, we might arrive at such a code. We must start with an international agreement on the most important types of cross-border restrictions, on the plurilateral level of the W.T.O., and not on a multilateral level. • MR. KHEMANI—At the World Bank, we have been searching for a cohesive framework to deal with international competition conflicts. There are conflicts between trade and competition policy in many areas, but one area where there is confluence is market access. Market access fosters competition, and competition policy can be used to foster market access. However, not all market access issues are necessarily competition issues. Therefore, the substantial lessening of competition screen must be used to determine when market access issues are competition issues. Moreover, means must be focused to distinguish between different types of such situations. Since the focus is on trade issues, the following should be considered: restrictions imposed by importing countries, restrictions imposed by exporting countries, and restrictions that may arise from strategic application of trade or competition policy. There are differences in competition policy objectives in different countries, and indeed, not all countries have a competition law in place. The number of countries with competition laws has grown considerably since 1990, and this trend is likely to continue. It is not necessary to dwell on the differences in competition policy objectives, because there is agreement on the core principles. Most, if not all, of the statutes that I have come across seek to maintain and encourage competition, although not necessarily to achieve the same ends. These differences in goals probably do not reflect differences in culture, since
140
Panel Discussion
competition is an area where there is a cultural confluence. Indeed, if one were to identify an area where neo-classical economists, Marxist economists, Islamic fundamentalists, Christian clergy, Hindu priests, Jewish rabbis and monks of other religious orders would agree, it would be on the evils of monopolistic exploitation. Thus, core principles of agreement exist, but it is necessary to address the differences in approach that can be accomplished by redesigning some of the institutional elements. Where international competition conflicts are involved, nations should be required to enact a national competition law. It need not be a specific law. Some jurisdictions may not have a specific competition law in place. However, their legal/economic system contain provisions against monopolistic exploitation. This is exemplified by the case of Mexico, whose constitution contained a provision against monopolistic exploitation, although it was not applied. International standards should allow private actions, so that the enforcement burden does not fall solely on the government, especially if the government is likely to be captive to various pressures. Political independence of the administrative mechanism to deal with competition conflicts should be fostered. In order to address trade-related issues, the law should establish standing of private parties, foreigners and exporters, including those without subsidiaries. Non-discrimination principles, i.e. national treatment, should also be provided. In an ideal world, competition principles should replace anti-dumping rules. However, if abolition of antidumping is not possible, then other screens should be established where antidumping suits are launched by firms that have a record of engaging in anticompetitive practices. • Ms. ROBINSON—An important agenda in international antitrust enforcement is partially relevant to international trade issues, but the concerns of international antitrust are much broader than those of international trade. That agenda can be realised soon, based on many of the efforts currently underway. It is important that thefirststep of an international antitrust agenda achieves meaningful cooperation in the prosecution of global cartels. Cartels are unambiguously viewed as bad. They are straightforward. The second step should be market access. Procedures currently underway to achieve bilateral agreement on positive comity are progressing, and could achieve much success in the near future. It is critical that these ongoing efforts continue as the W.T.O. work programme progresses. We urge a cautious approach to the W.T.O. agenda on competition policy, allowing us to observe what happens in these other areas. The differences in antitrust policy among the various countries may prevent
3 - Competition Policy Objectives in a Multilateral Competiton Code
141
any kind of affirmative result in the forseeable future in the W.T.O., but these other efforts are likely to succeed notwithstanding the differences. In the cartel example, the problem of gathering confidential information on an international basis has been difficult, largely due to rules of procedure in the us and abroad. However, these problems can be overcome. Arrangements to share confidential information are underway, and the U.S. has enacted a new statute that will permit greater freedom to do so. We are less sanguine, however, about the prospect of a global resolution of market access issues. The complexity of different laws in different countries creates a practical difficulty to cooperate in this area. To conclude, we are pleased with the cooperation that has occurred to date. We hope to see more international cooperation in the future, and we continue to be sceptical about what we can do with respect to market access. • PROF. MATSUSHITA—My remarks are based on the assumption that it is desirable to have some kind of agreement on competition policy within the W.T.O. framework. Globalisation of economies and the reduction of trade barriers following trade negotiations have brought about the situation that domestic restrictive practices increasingly affect international trade. Accordingly, market access has become an important issue. It appears that a competition code is needed within the W.T.O. On the other hand, the W.T.O. agreements are based on a very delicate balance between the powers of the member governments and those of international organisations. When the member governments are concerned about their sovereignty, this balance must be safeguarded. Violation of a W.T.O. competition code would be subject to the W.T.O.'s dispute settlement procedures, and economic sanctions could be imposed. Thus, we should proceed gradually and incrementally. One option is bilateral agreements, which are easier to conclude than multilateral agreements. Japan has not entered any bilateral agreements to date. We should also consider how the existing W.T.O. agreements might be used, as many of them contain competition provisions, including the agreements on trade in services, safeguard, T.R.I.M.S., and anti-dumping. Second, since we are in an early stage of the development of an agreement on competition policy at the W.T.O., I advocate an agreement whose scope is limited to restrictive business practices that directly affect international trade. It may cover, for instance, export cartels, import cartels and boycotts of foreign products. It is appropriate that a W.T.O. competition code cover these areas because such practices can nullify or impair the benefits of the trade concessions achieved through the international negotiations. Thereafter, we could proceed to a more comprehensive international code, such as that proposed by the Munich group.
142
Panel Discussion
Finally, supplemental agreements that rely upon private enforcement should be considered, such as an agreement concerning injunctive relief. The Japanese anti-monopoly law does not contain an injunctive relief provision. However, such a provision could guarantee access by foreign parties to the judicial process when they have market access problems. Other possibilities are international agreements providing for recognition of damage judgments, or recognising the effects of antitrust laws of foreign countries. The latter could provide that a contract that contravenes a foreign country's antitrust law would be null and void. • PROF. MAVROIDIS—The paper which I co-authored with Henrik Horn and Marc Baccheta is a straightforward application of the theory of regulation and delegation to the question whether there is a need to intervene and, if so, whether this should occur on the international plane. In essence, we consider whether negative spillovers from domestic competition policies mandate a multilateral agreement on competition. My colleagues suggest that negative externalities can occur from enforcement of domestic policies. They classify negative externalities in two categories: cases where domestic authorities do not take account of foreign producers' welfare; and cases where domestic authorities do not take account of foreign consumers' welfare. However, the magnitude of these problems is not known. Rather, these problems are only assumed to exist in theory. Nonetheless, the extent to which the existing legal regime can address those externalities is analysed. We find that existing trade and competition law instruments can largely address the existing theoretical problem. For instance, Art. XI of the G.A.T.T. can be used to address export cartels, as the semiconductors case at the end of the 1980s demonstrated. On the other hand, persuasive evidence has been gathered demonstrating that bilateral cooperation agreements in the competition field function well. We conclude that no problem exists with the existing regime that mandates a new approach. Actual use of the system is, however, limited. Only one case has been referred to the W.T.O. in 50 years: the Kodak-Fuji case. Thus, if use of the existing instruments is an indication, externalities do not present a serious problem. However, this could be deceiving. Perhaps the present system is not adequate. It is difficult to imagine how a case like McDonnell Douglas could go to a W.T.O. panel, a body consisting of three ad hoc adjudicating members, aided by a member of the legal affairs division, who are not permanent. Consistency cannot be guaranteed. Legitimacy of W.T.O. panel decisions is needed, and could be created by rectifying institutional aspects, in order to provide consistency, internal coherence, and long-term application of the same law, indepen-
3 - Competition Policy Objectives in a Multilateral Competiton Code
143
dently of the litigators. Such legitimacy currently does not exist at the W.T.O., largely because of the ad hoc nature of panels. However, new international substantive rules are not needed. Finally, our paper analyses the parallelism between the domestic level and the international level. At the domestic level, the list of per se prohibitions is shrinking, and the rule of reason is being over-extended. The same is true at the international level. At the W.T.O., two substantive rules are relevant: Art. Ill, which can be used against national champions, and Art. XI, which can be used against export cartels. Bilateral competition agreements do not contain any substantive rules, but alert the parties to each other's interests. Domestic competition authorities should begin to understand that their national interest is different from what it used to be. It includes not only consumers' and producers' welfare, but also foreign nations' welfare. This can be rectified only through existing procedural requirements that have not been used to date. • MR. SOUTY—The E.U. provides a useful model of how to internationalise competition rules, both from an institutional and a substantive perspective. Important differences in legal and economic cultures are present. There is convergence in the E.U. with respect to three elements: independence of enforcers, checks and balances, and advisory bodies, or 'toothless watchdogs'. Second, limited convergence on substance has occurred within the E.U. Since National enforcement objectives are not identical in the Member States, either among themselves or to the Commission's. Objectives have also varied within each Member State over time. Treatment even of the single anticompetitive practice about which we seem to agree, namely cartels, differs, both in terms of how cartels are defined and how they are treated under the law. As to the future of the W.T.O., I conclude with three brief remarks. First, history has taught us that international competition rules in Europe did not start with detailed rules. Second an international system should integrate the principles of institutional autonomy, independence, checks and balances, and advice from expert, toothless watchdogs. Third, an international system of competition rules should ensure that new national competition authorities and systems do not unnecessarily impair business operations, but impose market efficiency and, chiefly, transparency.
144
Panel Discussion
Observations and Comments • MR. GOLDMAN—A joint working party of the International Chamber of Commerce on trade and competition policy, which I chair and which has been ongoing for one and one half years, is dealing with the very issues that we are discussing. It is difficult for trade experts and competition experts to speak the same language on issues such as anti-dumping and antitrust. We have had such debates in Canada, as well as debates on the right of private action and private relief for competition law offences. Moreover, competition experts from different countries do not always speak the same language. A joint working group within N.A.F.T.A. on trade and competition policy has existed for about five years, which has made little progress. Accordingly, attempting to resolve issues of trade and competition policy, and as to whether there should be an international forum with some form of dispute resolution, is extremely difficult. We must not move too quickly; we ought to learn to walk together before we attempt to run, because otherwise we may end up with recommendations that constitute the lowest common denominator, and which may be ineffectual. Two years is a good timeframe to initiate the discussion, but these issues are going to take many, many years to resolve. The participants in this discussion should include government and the business community. Moreover, we ought to build on bilateral experience in the interim. • MR. WARNER—Market access is a very important and complex issue. Earlier discussions here focused on the use of economics, and the difficulties with economic analysis. Most market access issues involve vertical restraints. Economics is used to analyse the effects of vertical restraints. In working through the consequences of a vertical restraint with respect to market access, we must recognise the lack of consensus on economics, as well as issues of market definition. Query whether it is reasonable to rely on national competition authorities to handle a complaint from a foreign aggrieved party, and to expect that authority to ignore the domestic effects of its actions and instead pursue the world interest. It seems like something that academics might ponder, but unlikely to occur in the real world. Apparently, Kodak did not expect that the Japanese competiton authorities would resolve their concerns. The problem of legitimacy may have to be resolved internationally and institutionally. • PROF. HOVENKAMP—Any discussion of the extent of multilateral agreements must begin with the premise that such agreements are appropriate only where the economic consequences of the practice are unambiguous, and the
3 - Competition Policy Objectives in a Multilateral Competiton Code
145
principal effect of the practice about which the agreement is made is to export monopoly elsewhere. As long as nations pay a significant portion of the economic or social cost of a practice, they will have an inducement not to engage in it. Second, and more importantly, in cases where practices are not fully understood, differential policies across the world act as a very important laboratory in which we can test their consequences. Full multilateral agreement, such as an agreement to condemn vertical non-price restraints, or vertical maximum pricefixing,would eliminate the possibility of observing the impact of various rules. Until we have a very high degree of certainty that vertical maximum price fixing is economically harmful, such multilateral agreement would do much more harm than good. Accordingly, such agreements are only appropriate where the practices they cover are unambiguously harmful and the principle effect is to export monopoly elsewhere. • MR. FAULL—The experience of the U.S. Webb-Pomerene Act has not been an obstacle to U.S./E.U. cooperation in the competition field. In fact, the E.U. has jurisdiction over Webb-Pomerene cartels that operate in our territory. The Americans do not appear to object to that, and we could not object if the U.S. prosecuted firms for the harmful effects there of activities that we had exempted under Art. 85(3). Regarding the O.E.C.D. and W.T.O., I believe it is premature to discuss an O.E.C.D. agenda. Discussions are ongoing at the O.E.C.D. I agree that two years is a short period of time for the W.T.O. agenda. During this period, we are meant to explore but not to begin negotiating. There have been some surprising misunderstandings in the business community with regard to what we are doing. One should not expect business to like all aspects of cooperation between competition agencies. We just have to live with that, as they must as well. We must make great efforts to explain and consult, but there comes a time when competition agencies must take responsibility to enforce competition law, and business must comply with the laws. Finally, if the competition community does not work together, we risk seeing the trade community hijack what we do and hold dear, because they will not wait. With state-imposed barriers to trade coming down, the competition problems have become much more obvious. If we sit around for the next ten years arguing about the finer points of vertical restraints, the trade community is going to say this is a market access problem and that they will handle it. I do not believe this would be a satisfactory outcome. We must deal with vertical agreements that foreclose market access in a significant way. We can characterise them as having horizontal effects where appropriate. If we do not deal
146
Panel Discussion
with them in cooperation with each other, our trade friends will, and using less sophisticated economic analysis in competition policy terms. • MR. KHEMANI—Trade policy issues are separate from antitrust issues. In the context of domestic application of antitrust policy, market access issues should be treated as entry issues. That is, when a firm, or one or a group of exporters, finds it difficult to enter a market, domestic antitrust policy should be applied as it is in any other situation where a firm is attempting to enter a market and is facing barriers. A domestic competition agency would have to consider the sources of those barriers, and whether a substantial lessening of competition has occurred. Competition authorities receive many complaints, from both domestic and foreign firms, of barriers to entry into particular markets. We have developed a framework for deciding whether to act. In addressing this issue, I do not agree with others here that global economic welfare must be considered. No national authority currently has the power to do so. Finally, cartels may exist without government supporting mechanisms. Cartels, by nature, tend to be secretive, and not necessarily supported or facilitated by governments. When a market access problem exists where there is a transborder cartel, such as occurred in the Canada-U.S. Fax Paper case, it can be handled through the principles of cooperation and comity. However, when a foreign firm complains about a domestic cartel, an investigation may not take place because the foreign firm lacks standing. This is a procedural issue that can be easily addressed. • MR. OTTEN—At the W.T.O., a new working group has a mandate to study the interaction between trade and competition policy over an initial period of two years. We must identify issues that merit further study. At the end of 1998, the W.T.O. General Council should take a decision on what further work is needed. We are asking our members to identify the issues that they would like to see studied in the Group. The Members may submit written papers to this end. The European Commission has submitted a paper indicating that its thinking is further developed than those of other Members. Until now, other Members have been fairly discreet about the agenda they would like to see pursued in this Group. Thus there is no question of precipitous action in the W.T.O. This group will address areas where trade problems result from anticompetitive practices and how competition law is applied. Members are likely also to raise issues concerning existing trade instruments, and how they fit with the approach taken from a competition policy perspective. A range of provisions already in the W.T.O. agreements address enterprise practices which may distort competition, including in the areas of telecoms,
3 - Competition Policy Objectives in a Multilateral Competiton Code
147
state trading, monopolies and exclusive service suppliers. A number of issues that arise in accession working parties concern the implications for access to the markets of countries seeking accession to the W.T.O. of the extent to which there are functioning competitive domestic markets in the countries concerned. • MR. KOBAYASHI—I wish to comment on how well the W.T.O. is equipped to deal with competition issues. I am involved in the Kodak case. It is difficult to discuss this case because it is ongoing, although both sides have published most of their submissions. One point is that this case is not between Kodak and Fuji, but Kodak and the Japanese government, and it is not a competition case, but a trade case. Kodak's main argument is that it is not the private practices that have caused the problem. Rather, the problem is that the primary wholesalers in Japanese film distribution are virtually monopolised by Fuji. Further, Kodak contends that measures taken by the Japanese government have brought this situation about. These measures are now being questioned in the W.T.O. as a trade issue. In particular, the measures taken by the Japanese Ministry in charge of industry with regard to the distribution system are at issue. The J.F.T.C. is involved, not because it is responsible for the distribution system as such, but because it administers regulations on premiums and representations. It is alleged that this regulation blocks entry of a particular product from abroad. So again, this is a trade issue, not a competition issue. However, one aspect of the case relates to market structure: the extent to which the distribution system of Japanese film products is closed for foreign manufacturers. Kodak has presented a large volume of 'evidence' on this point. This volume of paper heavily burdens those who must decide the case, which is extremely fact-intensive. • PROF. NEVEN—International cooperation is needed only when there is a conflict between the various constituencies or authorities. When a conflict is present, one jurisdiction must take account of the cost or benefit that accrues to the other jurisdiction. Even if both jurisdictions agree on rules, it is necessary to ensure that these rules are actually implemented, where implementation is left to the discretion of the authorities. Some argue that an institution and sanctions are needed to ensure implementation. I believe that the problem of competition agencies in this context is a little bit like the problem of collusion among firms: all would be better off if all competition agencies take into account the interests of the others. Perhaps this is not completely unrealistic, to the extent that competition agencies have repeated interactions, that they have cases that require them to take account of the interests of each other. For instance, the half-
148
Panel Discussion
defunct E.E.A. agreement had common rules, as well as a rule to allocate jurisdiction. This required great trust among the authorities, because each was to implement the rules in the same way as the other when it had jurisdiction. • MR. SCHAUB—In the Kodak-Fuji case, I believe the trade action that has been taken was not the first preference of the U.S. For a long time, they had sought to get unilateral sanctions under Section 301, and it was only because this did not work out that they finally chose the second best approach. This matter could have been handled under the competition rules, but this approach was not pursued for fear that the J.F.T.C. would not take it seriously. This illustrates that no serious and effective international cooperation will occur as long as the competition authorities are not perceived as credible operators. They may be considered more or less credible in a given case depending upon whether an obvious national interest is involved. I would be interested to hear from Mr. Kobayashi about how the treatment of the Kodak-Fuji case was viewed from the inside of the J.F.T.C. Has the J.F.T.C.'s investigation led to anything? Is Kodak's perception correct that involving the J.F.T.C. would not lead to useful results? • MR. KOBAYASHI—Currently, three types of issues are before the W.T.O. on the Kodak matter. One is the question related to G.A.T.T. 1994, which I have already described. The second is a service-related issue, involving an alleged violation of the G.A.T.S. agreement, and this relates to the Japanese large-scale stores law. Kodak claims that this law hinders imports of films into Japan by restricting the growth of the numbers of large-scale shops, which the Japanese government disputes. The third is the competition issue, which concerns private practice in the Japanese market. The U.S. has proposed discussion in the W.T.O., under the G.A.T.T. decision of 1960, which relates to restrictive trade practices. However, this is outside the dispute settlement mechanism. This G.A.T.T. decision requires consultation to solve the problems related to competition policy issues. Japan accepted the proposal of consultation based on the 1960 decision under specified conditions. However, the other party did not accept one of these conditions. As a result, these proceedings have stalled, and it is not clear when this discussion on the 1960 decision will occur. As to the seriousness of the J.F.T.C, of course we inside the J.F.T.C. believe in it. Kodak itself came to the J.F.T.C, and provided it with some information. However, Kodak has provided no further information. On its own initiative, the J.F.T.C. is conducting an economic survey (as distinguished from a compulsory investigation). We are gathering information that is provided voluntarily, and have made very good progress.
3 - Competition Policy Objectives in a Multilateral Competiton Code
149
• MR. ABBOT—In a situation where a competition authority cannot investigate, for whatever reason, grounds may exist for a trade policy dispute under the W.T.O. rules, but no grounds for action on the basis of competition law. While the W.T.O. trade rules offer a possible solution, their use requires the W.T.O. to make a judgement on anticompetitive behaviour based upon trade provisions. This would perhaps lead to confusion, as may have occurred in the case on autoparts, had it not settled. It is only the first case in 50 years. Nevertheless, it is an important example of what could occur as markets become more globalised. Second, I am not suggesting that a choice must be made for either a network of bilateral agreements or a multilateral agreement. Rather, both are possible. Currently, we are exploring the possibilities for a two-year period in the W.T.O.; in the meantime, using and expanding bilaterals, especially regarding positive comity, presents no problem. Bilaterals should resolve some of the problems, but they may not resolve all of them. In the future, it may be necessary to compel a competition authority to make an investigation. The question would then be who would take action. The government of the country involved may decide to do something. Moreover, a basis for W.T.O. action may be established on the basis of an objective report rather than in the manner being described, i.e. a trade case with allegations flying to and fro with no real analysis involved. Therefore, I believe that both bilateral and multilateral efforts should be pursued, and the bilateral efforts are likely to move more quickly than the multilateral efforts. We are interested in measures affecting trade, and areas where governments have commitments and obligations but private enterprises do not. In contrast, negative spillovers involve the result of a decision taken by a competition authority, which could then have a negative effect. • Ms. ROBINSON—The large quantity of documents referred to by Mr. Kobayashi typically includes the most sensitive business plans and the strategic long-term plans of a company. This is the reason the business community is outspoken regarding positive comity. Moreover, these cases typically involve an entire industry. As to remedies, we must guard against being overly regulatory in designing relief. I have been involved with enforcement of the A.T.&T. consent decree for many years, which required me to act as a regulator for the telecommunications industry. The consent decree entered in the BTIMCI merger case has also been criticised for being so inherently regulatory, and having the potential to create problems in the long run. Even the relief in a fairly routine merger case must not be permitted to become overly regulatory. Relief must be designed to
150
Panel Discussion
allow the efficiency enhancing parts of the merger to go through, and remedy the part of the merger that should be remedied. Frequently, that involves an investigation, as well as long-term monitoring. • MR. FORRESTER—I would like to make a pessimistic point of a rather technical nature. Over the past years, it has been suggested that anti-dumping measures and competition measures are mirror images, complementary mechanisms, and that the one should take the place of the other. As dumping has been defined in the United States and in the European Union, there is very little correlation between what a businessperson or an economist would regard as dumping and what the statutes as upheld by the courts regard as dumping. Anti-dumping measures are, therefore, not normally a means of restoring fair trade (although sometimes they can be). Rather, they are a protective mechanism, even a protectionist mechanism. It would be very optimistic to assume that the elimination of anti-dumping measures would easily follow from the widespread institutionalisation of competition measures. If I were the government of a developing country and had the choice between elimination of developed world anti-dumping measures against my country in exchange for implementing competition rules in my country, I would regard that as a wonderful bargain. Correspondingly, I would imagine that wealthy business would probably regard the opportunity of doing business 'fairly' in that country as inadequate compensation for surrendering the benefits of anti-dumping measures. • PROF. JENNY—It is not clear whether competition policy is relevant for developing countries. We have acknowledged that competition policies should be tailored to local needs. We have assumed that Singapore, Hong Kong, Malaysia, and Indonesia need a competition policy, and that they are aware of this need. In reality, however, they do not know that they need one, and it is questionable whether they actually do. Second, if vertical restrictions are the main trade problem, then it would not be satisfactory to deal with everything except that. Mr. Faull has argued that we must deal with this issue because otherwise the trade community is going to hijack the policy. But is it correct to assume we are talking among ourselves as people involved in competition policy enforcement in developed countries that have a competition policy, or are we really trying to address the general issue of what could be done with respect to interaction between trade and competition? • MR. KHEMANI—I am a carpetbagger travelling around the world selling competition policy. Passing a competition law does not guarantee competition;
3 - Competition Policy Objectives in a Multilateral Competiton Code
151
and it is possible to have competition and competitiveness without having a competition law. The performance of Singapore, Taiwan, Japan and other countries that were viewed as not having a sufficient degree of competition have been highly competitive in the international arena. Most countries have little choice, due to internationalisation and globalisation of markets: they must address the issues of competitiveness. They find that becoming increasingly open, they can inject competition in their economy because export market performance is the best way to foster economic development and growth. It is possible to restrict competition in domestic markets, but it is impossible to restrict competition in international markets. Competition in international markets functions as a transmission mechanism for injecting competition in domestic markets. Moreover, governments do not have thefiscalspace to continue to operate state enterprises or provide subsidies as internationalisation occurs. Thus, they privatise and deregulate, and then focus on how to prevent private monopolies from taking over where government monopolies left off. They then begin to consider competition law. That is the normal sequence. There is no lobby for competition, because everyone recommends competition for someone else, but not for himself. Opponents to competition law tend to be rent-seekers. In many countries, such as Indonesia and other developing countries, the opponents are a small coterie of individuals. With increasing international communications and internationalisation of markets, the general population has begun to question why it does not have a share of the pie. That becomes a stimulus for competition. • PROF. MAVROIDIS—It is possible to have export cartels without government intervention, but the question is what is the size of the problem. It is questionable whether priority should be given to common competition rules rather than to finishing the unfinished agenda of the W.T.O. In the case of reiterated prisoners' dilemma, the incentive is to cooperate, and this is precisely why bilateral cooperation agreements function well. • PROF. FOX—To conclude this session, I make the following observations. First, it does not appear that the pursuit of different values in enacting and enforcing competition laws by different countries creates a serious obstacle to internationalisation. Therefore, the idea that competition laws must be harmonised before an international agenda can be realised appears to have withered away following this discussion. Other problems loomed much larger, such as whether any imperative exists for internationalisation of antitrust, and what the problems are in doing so. These problems do not, however, include a lack of common goals among the countries with antitrust rules.
152
Panel Discussion
Some participants believe that there is no need for internationalisation because they believe nations can take care of most problems themselves, and any new system would be costly. In addition, a new system might end up being perverse by handicapping efficient firms and therefore more costly. Some cite substantial differences between the goals of trade policy and competition policy and worry that a mixture of the policies will dilute antitrust. Several positive agendas were articulated. Prof. Bourgeois recommended a positive agenda for those inclined to adopt an international policy. To this end, he suggested that to advance internationalisation, those areas where there is a serious trade impact must be addressed. Where governments are prohibited from certain trade-restraining conduct, trade/competition agreements should require private parties also to refrain from the conduct. Moreover, where national interpretations lead to conflicts, an international rule of law should be formulated that would resolve the conflict. The participants held differing views as to whether a minimal agenda should be established on a broad scale or limited to the trade/competition interface and market access. If the latter, nations could be required to have and enforce laws prohibiting anticompetitive blockage of market access. Establishment of minimal principles on all kinds of restraints would be a very ambitious agenda, and would raise a series of problems. Reference was made to the Kodak-Fuji case, and the fact that its present posture separates the trade and competition issues. The separation of these disciplines, with trade people talking only to trade people, and competition people only to competition people, exacerbates the problems. Mr. Faull expressed the concern that if antitrust people do not address the problems of globalised competition, trade people might hijack this area. I take that to mean that competition policy will be hijacked by trade people who want fair trade as opposed to liberal trade. This is true, and it may be the impulse that leads even sceptics to address essential issues. Global issues arise, but tend to be viewed as national. No one is charged with responsibility to address the whole problem. National enforcement alone is not sufficient because nations tend to have blinkered vision. Internationalisation requires governments not to take nationalistic positions. This goes beyond the export cartel problem, although having export exemptions is one way to view the problem. Another is counting the costs on consumers beyond national borders. In the more remote future, disciplines could be imposed against anti-competitive dumping laws. I believe that disciplines on governments, as well as rules for settlement of conflicts, are two of the most important aspects of addressing internationalisation. Finally, Mr. Khemani suggested that developing countries have begun to
3 - Competition Policy Objectives in a Multilateral Competitor! Code
153
understand that they need to have open markets in order to integrate with the world in some way. Further, they understand that a competition system will help bring this about. In the United States, for many years the idea of open markets was a guiding force of antitrust. In the last ten or fifteen years, a more technical, economic paradigm was adopted. The value of open markets is no longer considered as a good in itself. Open markets have been a very important part of E.C. law, the E.E.A. agreement and the Europe Agreements. Open markets and market access are part of the same picture. The trade people perceive that we need to have an open market paradigm as a counterpart to the trade agreements that constantly rachet back government restraints. Nations perceive that openness of markets is an important, dynamic antitrust value. I believe that the concept of market access and openness of markets might help lead the way in to the first, and perhaps only, substantive international rule: that there should be no unreasonable private as well as public restraints barring market access. There is a linkage between liberal trade and competition. The trade people are saying it very clearly; the antitrust people are far behind. I believe that the discussion has moved us in the direction of making it a more solid point of reference for globalisation.
I Eleanor M. Fox Rapporteur of Session Two Walter Derenberg Professor of Trade Regulation New York University School of Law New York, New York, USA
Introduction Different goals, cultures, and institutions lie behind the competition laws of the various nations. For example, U.S. antitrust today is driven largely by the goal of maximising U.S. consumer welfare. U.S. law is relatively non-interventionist, and even monopoly firms are encouraged to engage in hard competition without regard to its impact on small firms. Canada seeks to maximise total Canadian welfare, and anticompetitive mergers may be justified not only by productive efficiencies, but also by their potential to increase exports ('competitiveness')- Even so, Canada applies some fairness principles; dominant firms are constrained not to take action that will unnecessarily foreclose smaller actors. European Community competition law is driven by a complex mix of integrating markets, preserving the economic opportunity of small and middle-sized businesses, aiding consumers and fostering competitiveness. Many newly reindustrialised countries use competition laws to keep markets open in order to safeguard the new economic liberalism and democracy, to foster freedom of enterprise and to protect small actors from the exploitations of still-dominant enterprises that are the legacy of the command and control era. The task of Session Two was to reflect on these differences and to consider their significance vis-a-vis efforts to internationalise competition policy. There was a remarkable consensus of the participants. All speakers agreed that differences in competition law goals are not, as such, an impediment to internationalisation, although not all were in favour of efforts to internationalise. Several participants stated that there is no need to press for further harmonisation of national laws as a precondition to building a foundation for an international understanding, and that insistence on harmonisation first would be likely to impede progress at building an international system. The participants in Session Two considered whether and why an international regime might be desirable or necessary, and if it is, what form that regime
158
Working Paper I
might take. While there was some divergence among the panel members, there was again remarkable agreement, and near consensus that: • nations should adopt competition laws and be obliged to enforce them against anticompetitive restraints of international trade; • the G.A.T.T./W.T.O. principles of transparency, national treatment and non-discrimination, and against beggar-thy-neighbour restraints, should be included in competition law obligations; • nations should continue to develop bilateral cooperation, including cooperation in discovery and other comity measures; and • if work begins on substantive rules against certain private or hybrid restraints, one should cautiously build the foundation, possibly focusing on: effects resulting from conflicting national regimes; core principles on which there is already international consensus; and the private equivalents of G.A.T.T.-illegal government restraints (e.g. quantitative restrictions on imports and exports and strategic nationalistic behaviour). On the other hand, there were many points of tension among the panelists, to which I shall turn.
The Participants' Papers As originally conceived, Session Two concerned 'Competition Policy Objectives in the Context of a Multilateral Code'. Stated differently, how could an international code be wise or attainable given that nations march to different drummers in devising and applying their competition laws? That question cannot be answered without reference to a concept of an international agreement, and, perhaps for that reason, most panelists quickly turned to sketching their views of what an international agreement might contain. The pursuit of divergent goals by nations was, for many, a minor theme. Prof. Bourgeois wisely observed that nations with different objectives often support common rules for different reasons. Accordingly, in this section, I attempt to synthesise the participants' perspectives on what, if anything, an international agreement should contain. The most fully developed proposal is presented by Mr. Khemani and Mr. Schone in their paper. I shall begin with a review of this paper, then work through several other papers that make discrete first-stage proposals and several that elaborate on W.T.O. institutions and what they offer as a framework for solving the problem of private restraints. Finally, I shall address the two most sceptical papers—those of Prof. Mavroidis and Mr. Klein.
4 - Competition Policy Objectives
159
The Khemani/Schone paper argues that the problem of market access restriction by private firm behaviour is a gap left largely unaddressed by the G.A.T.T.AV.T.O. system, and that where insufficient relief is available through current arrangements, the gap should befilledby an international agreement. The paper cites data, case examples and studies to show that trade liberalisation does not cure the problem; indeed, it presents facts illustrating that liberalisation may itself inspire further public and private efforts to protect domestic markets by new means. The paper makes a number of proposals, and offers case-specific issues for consideration by the W.T.O. Working Group on Trade and Competition. For instance, it suggests that: nations should be required to have and enforce antitrust law, and to do so on a non-discriminatory basis, especially against market access restraints; the competition agency should be independent; excluded exporters should have a right to sue in the excluding nation; to address uneven playing field issues due to differences in antitrust policy, the paper favours convergence on principles and minimum standards, backed by cooperative links among competition authorities; strategic antitrust policy, such as export cartels, should be prohibited, and minimum standards for R&D cooperation should be adopted; procedural minimum standards should govern merger control; antidumping laws should be abolished; positive and negative comity should be implemented; and W.T.O. dispute resolution should be available. The essay of Mr. Souty begins in a somewhat different place but ends with a compatible—if less detailed—suggestion. Mr. Souty presents a comparative view of competition policy and its evolution in a context of political economy, noting that some countries, such as the U.S., stress hard competition, and others, such as the U.K. and Japan, stress fair trading. Institutionally, he notes the successes of the qualities of independence, collegiality and expertise of and within antitrust agencies, and describes how the competition authorities in Germany, the U.K. and France are supplemented by a 'toothless watchdog' commission, which provides both expertise and advisement. Secondly, Mr. Souty observes how the European Union has been successful in promoting convergence of competition policy, even among nations with strongly different traditions. He draws from that experience the lesson that different traditions and objectives are not an obstacle to internationalisation. Thirdly, regarding cartels, Mr. Souty observes that all competition policies in the world have been basically designed to fight against cartels. This observation leads to a proposal: 'a limited prohibition which allows national enforcement under the review of a collective and autonomous body, advised by an independent toothless watchdog with a strong expertise in the field of law and economics, and subject to review by an appeal authority'.
160
Working Paper I
While Mr. Souty seems implicitly to anticipate the evolution of a much more fully developed international system, Prof. Immenga does so expressly. Convinced that internationalisation of competition law is inevitable, Prof. Immenga looks in the direction of a plurilateral system, based on internationally agreed rules, which leaves the enforcement of these rules to national agencies and courts while assuring that the application of the rules by national authorities is under international control. His suggestion goes farther than those put forward by the other panelists in contemplating a complete set of antitrust rules. Prof. Immenga believes it to be more important and essential than adopting specific rules to recognise the forces that are impelling us towards internationalisation, and to find the commonalities of systems that buttress the common cause in promoting 'the general virtues of maintaining competition'. Areas of convergence, he suggests, should be identified, and—because conflicts will be most easily overcome in these areas—they should become the core of the common policy. He identifies consensus wrongs, such as naked cartels. Prof. Immenga identifies the following categories of divergences in competition policies: differences within competition policies (e.g. different socioeconomic conceptions of competition policy and different economic theories); divergences which result from valuing certain external goals more highly than competition (e.g., jobs, the environment); diverging enforcement policies (margin of discretion, control by political bodies such as ministries or independent agencies or courts); and culture and tradition (e.g. degree of liberalism versus statism). These differences may make agreement on specific rules difficult, especially where national interests are at stake and where political interests— such as jobs—may trump competition interests in particular nations. Accordingly, agreement on wrongs such as naked cartels will be easier to reach than agreement on mergers, which are often linked to industrial policy. Nonetheless, the points of common interest 'should encourage a process to develop an international code and to overcome existing national divergences'. The paper by Prof. Matsushita is premised on the proposition that world competition is increasingly overshadowing domestic competition so that, in the future, a clear line between domestic and international trade policies will no longer be distinguishable. In this first and current stage, like Mr. Khemani and Prof. Immenga, Prof. Matsushita sees the 'need to control private restrictive practices that offset the effect of liberalisation achieved through trade negotiations and agreements'. However, in contemplating the later stage—a more nearly integrated world—he argues that a more ambitious approach will be required, although nations should not be requested to give up more sovereignty than necessary to deal with the global problems.
4 - Competition Policy Objectives
161
Prof. Matsushita believes that the W.T.O. is the right forum for integrating trade and competition issues. He explains that the W.T.O. is a forum for liberalisation which incorporates a new focus on the rule of law, in view of the enforcement and appeal mechanisms adopted in the Uruguay Round. He outlines existing W.T.O. agreements that contain competition policy aspects, prohibiting or authorising nations to prohibit public or private anticompetitive restraints. These agreements include: Agreement on Technical Barriers to Trade (T.B.T.), Agreement on Trade-Related Investment Measures (T.R.I.M.s), Agreement on (i.e. limiting) Antidumping Subsidies and Countervailing Duties, the Safeguard Agreement (prohibiting agreements on voluntary export restraints), General Agreement on Trade in Services (G.A.T.S.), and Agreement on Trade-Related Aspects of Intellectual Property Rights (T.R.I.P.s). He encourages policymakers to build upon those agreements. Prof. Matsushita describes the diverse goals and values of nations perhaps more poignantly than others. He identifies as the common thread 'preservation of the market mechanism'. Behind this common thread are nations' diverging needs and preferences, e.g. for efficiency, pluralism (to counteract suspicion regarding concentrated power), equity in transactions (to counteract undue use of bargaining power to impose harsh terms), market integration (E.U.—which might provide some lessons for an integrating world), and the goals of transition economies and less developed or developing countries like China, whose needs include a mixture of establishing a modern civil law system and protecting market actors from unfair competition. Prof. Mitsushita believes that eventually it will be necessary to enter a W.T.O. agreement wide enough to accommodate nations' diverse objectives. He contemplates openly hearing the voices of developing countries. For the present, his proposal is modest: to deal with core problems (e.g. cartels and boycotts) that directly impact international trade. Mr. Abbott provides a view from the trade community. While national competition authorities' bilateral cooperation on pure competition issues is welcome, he said, '[i]t is not what some in the trade community envisioned when we called for progress on international rules relating to trade and competition policy'. Mr. Abbott focuses on cases in which market access is impaired by private behaviour. He observes that many W.T.O. countries do not have antitrust laws, and others that do might not fairly investigate and enforce their laws to enhance foreign access. Countries are experimenting bilaterally with positive comity, but this has yet to be tested. Although market access is an international trade and competition problem, no legal base for multilateral surveillance has been established.
162
Working Paper I
Mr. Abbott recommends multilateral agreement obliging nations to adopt competition rules with a proper structure for enforcement, and obliging competition authorities to investigate market access complaints and to make their investigative reports publicly available. Failure to do so would be an actionable breach within the W.T.O. dispute settlement system. Prof. Bourgeois, who is expert in both the trade and competition disciplines, observes that unfair trade practice law causes the greatest tension with competition law. He suggests that the two bodies of law be kept separate and, for the moment, that we simply live with the inconsistencies. In time, we can face the need to adjust unfair trade laws, such as antidumping law, to competition goals. Second, Prof. Bourgeois contrasts the perspective of the trade lawyer with that of the competition lawyer and suggests lessons that we can learn from one another. He suggests that competition lawyers 'may learn that the absence of international competition rules also stifles competition'. Third, he provides the insight that an international competition agreement does not presume consensus on objectives. Experience shows that actors in international negotiations reach agreement without agreeing on objectives. Fourth, states may reach a common formulation when the nuisance value of many diverse rules is high enough. The nuisance value includes: adverse effects on a nation's own regulatory system and on its market actors which result from multitudinous laws with different standards, and the anticompetitive effects of trade laws adopted in the name of the unlevel playing field. Applying these lessons, Prof. Bourgeois suggests that the most likely target of an international competition agreement would be conduct having an impact on international trade and its most likely purpose would be to prevent anticompetitive conduct that restricts international trade. Most likely subjects for rules would be forms of conduct with respect to which there are already some international rules regulating the conduct of governments. For instance, boycotts of imports and exports, abuse of dominance by exclusionary, discriminatory and exploitative practices, rules against bid rigging, and barrier-creating standard-setting can all be seen as extensions or counterparts of W.T.O. agreements, such as those on Government Procurement and Technical Barriers to Trade. Also a good candidate for rules would be 'forms of conduct whose disparate regulation by states is most likely to result either in excessive costs and delays for multi-jurisdictional transactions, or even in allowing one state to block such transactions [as mergers and alliances] which have been cleared by other states . . . .' Unlike the first five author/panellists, who believe that there is an international problem that would naturally be solved by international measures and who focus on how best to address this problem, Mr. Klein and Prof. Mavroidis
4 - Competition Policy Objectives
163
question these premises. Mr. Klein's paper identifies three areas of concern in the global economy: transnational mergers, which may be subject to review by numerous national authorities; international cartels; and market access blockage. He contends that multi-nation review does not hinder mergers. Moreover, the parties can decrease transaction costs by allowing the various reviewing authorities to share confidential information. Mr. Klein identified cartels and market access restraints as areas in which arrangements between governments may significantly enhance enforcement. In both areas, the U.S. Department of Justice has sought, and in some cases obtained, the cooperation of antitrust authorities abroad. Without cooperation and without being able to gather evidence abroad, enforcement may be stymied. This concern led the U.S. Congress to pass a statute giving the Justice Department authority to negotiate bilateral antitrust cooperation agreements and to exchange confidential information on a reciprocal basis with trusted foreign counterparts. Mr. Klein also reported the U.S. initiative in the O.E.C.D. contemplating bilateral agreements against hard core offences. Finally, he detailed the refinement of positive comity and recommended agreement to fit the following scenario: Country A, believing that its companies are closed out of country B by private restraints, may make a preliminary determination that there are reasonable grounds for investigation. If it does so, it may refer the matter to country B's authorities, who would then have the duty to investigate and report back to country A, and to consult with country A as to: the nature of the investigation, B'sfindingsand any remedy B's authorities are considering. Mr. Klein expressed fear that W.T.O. initiatives on trade and competition could undermine the ongoing cooperative efforts among competition authorities. He argued that: agreement on sound competition rules will be hard to reach, because different national objectives will result in forced trade-offs; the process might result in a lowest common denominator, legitimating weak and ineffective rules; W.T.O. dispute resolution would be infeasible, inappropriate or—if it is limited to review of nations' obligations to adopt and enforce antitrust rules—ineffective, because competition laws go beyond core W.T.O. concerns, and competition law enforcement is fact-intensive and involves confidential information. Profs. Bacchetta, Horn and Mavroidis challenged the 'internationalists' by methodically addressing the focused question: do negative spillovers from nationally pursued competition policies provide a case for multilateral competition rules? In effect, they ask when negative spillovers produce inefficiencies from an international point of view, and whether there are any significant such inefficiencies (acts or conduct that decrease aggregate world wealth) that
164
Working Paper I
cannot be addressed in the existing legal framework. They conclude that since conduct with spillovers (e.g. import cartels that foreclose non-nationals) does not necessarily decrease world wealth, it does not provide an economic rationale for a multilateral competition agreement. The Bacchetta/Horn/Mavroidis paper investigates the extent to which economic incentives would induce nations to pursue competition policies that impair world resource allocation. It finds no economic reason to believe that incentives run in this direction (finding, among other things, no economic reason to believe that liberalisation will give nations added incentives to pursue beggar-thy-neighbour policies). The paper then considers the extent to which current law, especially trade law, is or can be used to resolve the problems. It finds this extent large, actually or potentially, after detailing and discussing all available G.A.T.T.AV.T.O. instruments and obligations, bilateral cooperation agreements, and the availability of extraterritorial applications of national law. The authors have applied economic models which expressly exclude certain fact-based realities. For example, they assume for their model that nations pursue the goal of national aggregate wealth, though they recognise in their text that nations take protectionist measures. The authors conclude that no international agreement should be pursued unless it can be shown to increase aggregate world wealth. The paper has a contrapuntal relationship with Session One, which shows that competition policy was adopted in various nations for a variety of policy reasons, including concerns with the distribution of wealth and economic opportunity, openness, fairness and legitimacy. Indeed, it has been observed that most nations would not adopt competition laws if their goal was to improve the allocative efficiency of the nation, because anyone with this burden of proof could not meet it.' Accordingly, some may believe that the objective of internationalising competition law is not to achieve more aggregate world wealth, but to mediate systems dissonance, eliminating narrow nationalism and opening the world to economic opportunity on the merits, and to gain a world vision of world transactions. Therefore, it would be interesting to ask the authors to flip their question and ask: assume a system such as proposed by Khemani. What if any are the costs to world welfare?
1
See Dominick T. Armentano, Antitrust and Monopoly—Anatomy of a Policy Failure (1982); John S. McGee, Why not 'Deregulation' for Antitrust?, in Industrial Concentration and the Market System, 53 (E. Fox and J. Halverson, (eds.) 1979).
4 - Competition Policy Objectives
165
Conclusion The papers are uniquely harmonious and especially noteworthy for the set of liberal principles, outlined in the introduction, to which they basically subscribe. Nonetheless, there are many differences, the most obvious of which is that between the majority of participants (who accept the internationalisation of competition law and work 'in the vineyards' to try to guide its development along fruitful paths), and the minority (who do not accept internationalisation of the law and work to stem the tide). Most in the majority seem to have the positive view that we are living in a world of global markets, and that there is a need for architects to fit the law to the problems. The majority also includes, however, those persuaded by the remark of Mr. Faull that if the competition community does not work together to consider what form of competition agreement makes sense, its worse fears will come true—the trade community will do so.2 Even among the 'positivists', there are many different proposed routes. Some envision, ultimately, a complete set of world competition rules; others would limit world rules to a few hard core consensus categories; yet others would limit world rules to anticompetitive restraints (and maybe core categories) that directly restrain international trade. Some would identify consensus wrongs as the convenient starting point for a fuller world system, pragmatically starting where conflict is least; others would identify consensus wrongs as the place to start and end. Some envision competition policy broadly and use the opportunity to seek to eliminate inconsistencies between competition as the rule of trade and protectionist national laws, such as antidumping; others would work more narrowly inside competition proper. Some would accept the wide range of values underlying policies that nations may call competition policy (including protecting the weak against the strong); others construct their proposals so as to avoid fairness issues and only to build upon efficient competition rules. Some see the international exercise as levelling the playingfield;others see the exercise as a way to promote world efficiency; still others see the project as simply a way to link systems and put a lid on nationalistic behaviour. All 'positivists' contemplate some dispute resolution but none work out the details of a dispute resolution system how the issues for dispute resolution would be framed, when an issue would be ripe for dispute resolution, what if 2
It is said that the trade community is more likely than the antitrust community to embrace protectionist ideas and bargaining methodologies rather than rule of law in the interests of competition and consumers.
166
WorkingWaper I
any, fact-finding authority a dispute resolution panel would have, and how the panel would receive and assess the evidence. This remains an important discussion item, for the feasibility of internationalisation with dispute resolution cannot be fully addressed without positing a dispute settlement system that will work. Regarding enforcement, most 'positivists' had in mind public and private enforcement, beginning at national level under national law that would incorporate agreed international substantive principles accompanied by duties of non-discrimination and national treatment. The panelists of Session Two have significantly advanced the debate on internationalisation of competition law. Engagement at the points of difference remain for another day.
II Roderick E. Abbott Head of Delegation of the European Commission Geneva, Switzerland
Introduction This paper sets forth the views of what the trade policy community considers to be the important elements of the interface between trade and competition policy. In particular, it will focus on the kinds of international cooperation that could be achieved, and how the two communities could reinforce each other as to the ultimate objectives that they largely share. The papers in Session One indicate that the legitimate objectives of competition policy are generally considered to be complementary to the goals of trade liberalisation. In practice, however, these objectives are being pursued by the two policy communities in a largely separate fashion. For example, only two of those papers explicitly mention the competition/trade interface, or the need for international cooperation or multilateral agreements. Others refer in passing to the trend towards globalisation of world markets, or trade liberalisation as factors affecting competition objectives. This is quite natural, since Session One addressed the legitimate objectives of competition policy. However, it gives the impression that the analysis of policy objectives is essentially focused on economic efficiency in the context of the development of national economies, and that application of these concepts in the context of the global economy is not yet seen as of similar importance. The papers for Session Two recognise the global economy, but take considerably different approaches regarding the purpose that a multilateral competition code could serve and the purpose of international cooperation. This paper shall consider whether mere complementarity is enough, or whether a more proactive stance must be taken to ensure that competition policy objectives can help achieve a greater degree of market access in a world where globalisation is reducing the importance of national boundaries.
A. International Cooperation Major concerns exist in the competition policy community relating to conflicts over which competition authority has jurisdiction over certain activities (especially over cartels and mergers leading to dominant market positions). In
168
Working Paper II
particular cases, more than one authority may have a specific interest, which raises the possibility that these authorities share information and exchange views. In some cases, more than one authority may even assert decisionmaking authority. This constitutes a typical international problem for competition policy-makers. For instance, the competition authorities of both the U.S. and the E.U. have investigated the current merger proposal of Boeing aircraft company with the civil aircraft interests of McDonnell Douglas. The European Commission investigated this proposed transaction and its possible competition effects on the European market for sales of aircraft and of parts and maintenance services. Similarly, several years ago, a merger was proposed between two regional aircraft companies, A.T.R and Alenia, and the Canadian company, DeHavilland. The European Commission prohibited that proposed merger, based upon the negative effects it would be likely to have on competition in the European market. These concerns have led to the establishment of bilateral agreements between governments, which establish the procedures for improved cooperation between their competition authorities. This is welcome, but it is not what some in the trade community envisioned when we called for progress on international rules relating to trade and competition policy. In the trade community, we were concerned that, in certain cases, anti-competitive practices in the market of another country led to difficulties for exportingfirms.Such a case would clearly fall outside the jurisdiction of the competition authority of the exporting firm, but no mechanism was available to have effective action taken in the other country to address such problems. For instance, in recent years, there have been well-publicised difficulties between the United States and Japan over access to the Japanese market for car sales and for the supply and sale of car parts. Some of these difficulties related to the Japanese system of distribution via auto dealerships, as well as the way in which maintenance and car part sales were organised within the system. These problems were partially covered by competition rules; they also constituted a trade issue, to the extent that the U.S. felt that it was denied opportunities to develop its export sales in Japan. Another case, currently before a W.T.O. panel, concerns the conflict between Kodak of the U.S. and Fuji of Japan regarding sales of films in each others' markets. This case also involves questions of distribution systems and of dominant market positions in both countries. In such cases, it was usually possible for an exporting firm to lodge a complaint with the competition authority of the country concerned, but there was no guarantee that it would be investigated or, if an investigation was made to
4 - Competition Policy Objectives
169
guarantee that a substantive report would be published. Further, competition authorities often took the position that, although exporters experienced difficulties, no clear violation of their competition law had occurred and thus that they could take no enforcement action. At present, international law offers no mechanism for multilateral surveillance of procedural or substantive decisions of national competition authorities. Moreover, although the principle of positive comity is gaining ground in some bilateral agreements, it is still to be tested in situations such as those described above. This left a situation where grounds could arise for a trade policy dispute within the W.T.O. rules, but no case for action on the basis of competition law. Although the W.T.O. could make a judgement on competition law violations on the basis of trade provisions, this could lead to confusion over the respective roles of the two policies. It seemed that it would be more effective if rules could be established which would oblige a competition authority to investigate such a complaint, and to make a report publicly available. This would allow any W.T.O. complaint to be based on an objective factual report from the competition authority concerned, rather than unsubstantiated allegations and counter-allegations. Further, the failure to fulfill such commitments could itself be actionable as a breach of a government obligation within the W.T.O. dispute settlement system. B. Multilateral Code Thinking on these issues has evolved since they were initially set forth. Although a certain degree of homogeneity has developed in O.E.C.D. countries regarding the existence and application of competition laws, no such general statement could be made with respect to the W.T.O. member countries. On the contrary, many W.T.O. members have no competition law at present, although that number is decreasing; still less any form of independent enforcement authority. Thus, primary focus at the W.T.O. has been on establishing recognition by all governments of the need for such a law, and commitment to introduce it and a proper structure for its enforcement. Only following a more general, commonly shared set of principles and legal provisions would it make sense to seek international rules in any form. Annex At the Ministerial meeting of the World Trade Organisation held in Singapore in December 1996, the members decided to establish a Working Group to
170
Working Paper II
study the interaction between trade and competition policy. In doing so, the W.T.O. has broken new ground. Previously, competition law and its enforcement had been considered domestic policy. Following this decision, however, it is now broadly recognised that the world economy is globalised, and often liberal and integrated. Competition policy, if it is to remain effective, must keep pace with these developments. This Annex addresses the international dimension of competition policy from a W.T.O. perspective, since binding global rules or principals may be developed in the future in this forum. The W.T.O. Working Group faces many challenges: • Competition policy of W.T.O. members is at widely different levels of development: some have had a competition policy for decades, while others have only recently drafted legislation, and still others have not yet taken any legislative action; • The objectives and emphasis of competition policy varies from country to country; • Competition enforcement structures vary from country to country, some relying predominantly on administrative action for enforcement, others relying on both private action (which may be encouraged, inter alia, through provisions for treble damages, etc.) and administrative action, implemented through judicial decisions of civil courts. The Working Group may consider the following questions: • Assuming that it would not be feasible to create a supranational competition authority with its own powers of investigation and enforcement in the near future, can we agree that any international rules which may be developed should be intergovernmental in nature, as are all W.T.O. rules? • Could a feasible first step towards an international framework of competition rules be taken through an agreement among all W.T.O. Members to adopt competition policy and enforcement structures domestically? • With respect to which categories of anticompetitive practices is it likely that international consensus on a set of principles could be mustered? • An effective international code should ultimately contain a number of binding elements, in order to ensure that the provisions which support market access objectives (i.e. where anticompetitive practices are restricting exports or investments in foreign markets) can function properly. How can enforcement of agreed principles best be guaranteed? Should this be through intergovernmental administrative cooperation? If one competition agency asks another to take investigative, and if necessary, enforcement action, to what procedural and substantive obligations should the latter be subject?
4 - Competition Policy Objectives
171
• A requesting agency may refrain from taking action itself, and instead seek to ensure that a requested agency is effectively addressing the alleged anticompetitive practices through the procurement of information from that agency and the use of the 'positive comity' instrument when its important interests are affected. How can information exchange and positive comity instruments be developed further? What might be the implications of the development of interagency cooperation on the prosecutorial discretion of competition agencies? • In view of the increasing scope of W.T.O. rules (goods, services, in the future possibly foreign investment), should an international code allow one country to ask another to take enforcement action if the important interests of the former are affected regarding both its exports and its investments?
Ill Jacques H.J. Bourgeois Partner, Akin, Gump, Strauss, Hauer & Feld Brussels, Belgium Professor, College of Europe Bruges, Belgium
A. Scope of Competition Rules Countries developing their first competition policy generally seek to foster competition, ensure fair competition and protect consumers. Countries that already have competition rules in place may also seek to realise these objectives, but some do so in separate laws. For instance, a country may have competition rules, including a prohibition of restrictive business practices and of abuse of market power, as well as laws prohibiting unfair trade practices, designed to protect both competitors and consumers. The rules designed to achieve these various objectives, whether included in a single law or enacted in separate laws, may conflict. For instance, on 14 July 1991, the Belgian Parliament passed an (amended) Act on Trade Practices and Information and Protection of the Consumer1; less than one month later, on 5 August 1991, it passed the Act on the Protection of Economic Competition,2 which is modelled on E.C. competition rules. However, neither Act sets forth the relationship between the two Acts. In particular, they do not address the question whether a business practice consistent with the latter Act may still be held contrary to the former Act. The Competition Council, which is the special tribunal established to enforce the Act on the Protection of Economic Competition, prohibited a merger on grounds that had less to do with this Act than with the Act on Trade Practices.3 Accordingly, when reflecting on competition policy objectives in a multilateral context, countries should first agree on the definition of the subject matter, clarifying whether objectives pursued by legislation on unfair trade practices are to be considered competition policy objectives. I believe that these objectives should not be included and should be kept separate. This does not exclude the possibility that if and when international competition rules are established, 1
[1991] Moniteur Beige {MB), 18712. [1991] Moniteur Beige 22493; ECLR (Supplement). 3 Conseil de la Concurrence—Decision of 18 May 1994, no 94, C/C 14, Parfumerie Douglas/Compartilux [1994] MB 18412 (12 July). 2
174
Working Paper III
they will have an impact on, and perhaps require a modification of, unfair trade practices rules, such as antidumping laws.
B. The Need for International Competition Rules The need for international competition rules has been considered in the literature from two perspectives: from the perspective of competition policy and from the perspective of trade policy. The trade policy community generally supports the idea4; however, the competition policy community is more sceptical.5 The issue should be dealt with simultaneously by both the trade policy and the competition policy communities. In doing so, the trade policy community may learn that regulating competition at the international level is more complex than regulating government procurement or the trade-related aspects of intellectual property. The competition policy community may learn that the absence of international competition rules also stifles competition. The evidence which places the two communities at odds with each other with respect to the need for international competition rules is, thus far, not persuasive. For instance, the contention that trade policy measures have a stifling effect on competition is supported only by anecdotal evidence.6 Likewise, with respect to the competition problem resulting from the coexistence in an increasingly globalising world of different competition systems, the evidence appears to be anecdotal.
C. The Need for a Consensus on Objectives Any effort to build consensus on policy objectives before negotiating on international rules may be counter-productive, and could even have the effect of making any progress on rules impossible. Experience shows that actors in 4
E.g. Bernard Hoeckman and Petros Mavroidis, Competition Policy and the G.A.T.T., 17 World Econ. 121 (1994) (the authors see four holes in the W.T.O. as far as competition policies are concerned). 5 See various contributions in O.E.C.D. Documents, New Dimensions of Market Access in a Globalizing World Economy (1995). The O.E.C.D. Committee on Competition Law and Policy is opting for 'convergence' rather than for a set of international rules. See O.E.C.D. Interim Report on Conveyence of Competition Policies (June 1994). 6 See some examples analysed by Jacques Bourgeois and Paul Demaret, in European Policies On Competition, Trade And Industry. Conflict And Complementarities 65 (Pierre Buyges et al. (eds.) 1995).
4 - Competition Policy Objectives
175
international negotiations may be able to agree on rules, even if they each have different reasons for accepting common rules. Thus, they may not agree on the objectives that underlie the agreed rules. Paradoxically, a successful negotiation on rules may require that policymakers devise rules that pursue different objectives. For instance, the objectives pursued in the Tokyo and in the Uruguay Round negotiations on subsidies differ from the objectives pursued by the E.C. rules and the U.S. rules. The E.C. rules are based on the 'injury-only' theory, under which subsidies play a role in correcting market distortions and may thereby pursue policy objectives other than maximisation of economic efficiency. Under this theory, subsidies should only be prohibited and counteracted where they cause injury to the industry of another country. In contrast, the U.S. rules are based on the 'antidistortion' theory, under which subsidies are viewed as inconsistent with the law of comparative advantage, which maximises production and distribution efficiencies. These contrasting objectives are reflected in different regulatory approaches: under the 'injury-only' theory, subsidies are permitted unless their injurious effects are established; under the 'anti-distortion' theory, subsidies are prohibited unless it is established that they have no such effect. Notwithstanding these differences, both sides were able to reach a comprehensive agreement regulating subsidies and countervailing measures in the Uruguay Round.
D. Feasibility of a Commonly Accepted Formulation of Competition Law and Principles As states become less able effectively to control anticompetitive conduct in global markets, and as differences in their laws and principles have increasingly adverse effects on their interests, they will probably become more willing to depart from their own competition law and principles in order to reach a common formulation. Accordingly, the nuisance value of the absence of a common formulation could be an important factor in consensus-building. In assessing such nuisance value, one should also have regard to the competition stifling effects of trade-policy measures whose purported justification lies in the absence of a common formulation of competition law and principles. This is particularly the case for antidumping laws. Governments and domestic industries often justify the existence of such laws by noting that, in the absence of international competition rules, exporters can sell abroad at low prices while protecting high prices in their home markets by means of competition restrictions. It is noteworthy that the European Community was willing to renounce enforcement of its antidumping rules within the European Economic Area
176
Working Paper III
when a comprehensive set of competition rules and a mechanism for enforcement were established there. It is doubtful, however, whether certain states would be prepared to abandon their antidumping rules if and when a comprehensive set of international competition rules and a mechanism for their enforcement were established.
E. Differences in Competition Policy Objectives as an Obstacle to Achieving International Agreement The extent to which differences, in competition policy objectives impede the achievement of agreement on a commonly accepted formulation of law and principles with respect to a series of typical forms of conduct would depend on a number of factors, including: • Whether the formulation would apply generally or only to conduct likely to have an impact on international trade—the more likely object of the exercise would be conduct having an impact on international trade, and its more likely purpose would be to prevent anti-competitive conduct that restricts international trade. This would require devising some brightline tests to define when anticompetitive conduct is deemed to have an impact on international trade, and is therefore covered by international rules. • The extent to which such international competition rules or principles would be detailed—experience at both national and international level would caution against detailed substantive rules or principles. Establishing a detailed catalogue of competition restraints at the international level would probably be impossible, and futile in view of the dynamics of the market-place. The less detailed the substantive rules, the more necessary an international dispute settlement machinery would be. The acceptability of having international panels of arbitrators interpret a generally worded international antitrust code has been questioned.7 Interestingly, arguments which have been made against this are similar to arguments made in the Uruguay Round negotiations that led to the W.T.O. Dispute Settlement Understanding.8 One such argument was that review of complex decisions of national authorities would constitute 'second 7
E.g. Daniel Tarullo, Wrong Lessonfrom Boeing, Financial Times, 13 Aug. 1997, at 12. See Steven P. Croley & John H. Sackson, W.T.O. Dispute Panel Defence to National Government Decisions: The Misplaced Analogy to the U.S. Chevron Standardof-Review Doctrine, in International Trade Law and the G.A.T.T.-W.T.O. Dispute Settlement System 187, 189 (Ernst-Ulrich Petersmann (ed.) 1997). 8
4 — Competition Policy Objectives
177
guessing' their conclusions. However, the review by dispute settlement bodies need not be de novo, and certain rules could be agreed upon denning the scope and the standard of review by such bodies. • whether such formulation would leave room for national policy concerns where it sets detailed rules—it may be possible to avoid dealing with the differences expressed in 'efficiency versus fairness' by seeking to distinguish between competition policy proper and fair-trade policy, and by focusing on competition policy proper, while continuing to live with possible inconsistencies between fair trade policy and competition policy. For the purpose of selecting the most appropriate area for common formulation, the following distinctions should be made: • A first distinction should be made between forms of conduct already subject to some international rules (although such rules regulate the conduct of governments) versus those for which there are no international rules. Several forms of conduct belong to the first category: bid-rigging could be addressed as an extension of the existing W.T.O. plurilateral Agreement on Government Procurement; boycotts by enterprises could be viewed as seeking the same result as import or export restrictions imposed by governments, as could abuse of dominance (including exclusive and other dealing that tends to foreclose), refusals to deal, high price strategies and discriminatory exclusions that have the same results as governmental restrictions on international trade. A similar situation exists with respect to standard setting by private industry where it would lead to the same sort of obstacles to international trade as those which the W.T.O. Agreement on Technical Barriers to Trade addresses. • A second distinction should be made between forms of conduct whose disparate regulation by states is most likely to result in either excessive costs and delays for multi-jurisdictional transactions, or even in allowing one state to block such transactions which have been cleared by other states, and other forms of conduct. International joint ventures, alliances and mergers are most likely to fall into the first category. The current uncertainty in a number of jurisdictions with respect to when joint ventures will be treated as mergers has a chilling effect. F. Differences in Competition Policy Objectives as an Obstacle to Implementation of an International Agreement If the scope of a common formulation were limited to the forms of conduct identified in Section E above, it could be successfully implemented, provided
178
Working Paper III
the parties also agreed to certain ancillary commitments. For instance, national competition authorities could be required to engage in mutual consultation for alliances, joint ventures and mergers that were deemed to have an impact on international trade. A 'one-stop-shop' principle could be established, which implies an agreement on allocation rules. In certain cases, it appears that competition authorities have cleared mergers and merger-like operations that should have been prohibited in order to further national industrial policy concerns. Accordingly, states might be reluctant to accept a 'one-stop-shop' principle, thereby relinquishing control over some of such transactions. This would be short-sighted, as this 'one-stop-shop' principle would apply precisely where there are competing claims of jurisdiction. Even if one state were to clear such a transaction, another state would be likely to prohibit it, with the consequence that the parties abandon it. Moreover, a carefully drafted 'one-stop-shop' rule would mean that a state which relinquished jurisdiction in some cases would acquire exclusive jurisdiction in others. G. Differences in States' Commitment t o Competition as an Obstacle to an Agreement Imposing Disciplines on Governments Regarding state aid, W.T.O. Members have accepted a number of disciplines by entering the Agreement on Subsidies and Countervailing Measures: some subsidies are prohibited (Art. 3), others are 'actionable', i.e. may not be used by W.T.O. Members when they cause adverse effects to the interests of other W.T.O. Members (Art. 5); still others are 'non-actionable' (Art. 8). Moreover, this Agreement imposes on W.T.O. Members the duty to notify 'any subsidy as denned in paragraph 1 of Art. 1, which is specific...' identifies the information that notifications must contain (Art. 25), and introduces surveillance by a W.T.O. Committee (Art. 26). Thus, a consensus on objectives is not always a prerequisite for a consensus on disciplines. Regarding the facilitation by a state of conduct by enterprises that runs a foul of consensus rules, the more similar such conduct is to measures prohibited by international trade rules if taken by the State, the more likely an agreement on disciplines applicable to states would be.
IV Ulrich Immenga Professor of Law, Director, Institute of International Economic Law, University of Gottingen Gottingen, Germany
Introduction Differences among nations with respect to the objectives of competition policy will probably eliminate the possibility of any quick and easy understanding among nations. Identification of categories of competition policies will aid evaluation of these differences and their impact on the potential for harmonisation or development of a multilateral competition code. These categories aid in identification of the importance and degree of obstacles which must be considered in order to proceed on an international path of competition policy. Four such categories are suggested here: • Differences in competition policy in a strict sense; • Differences resulting from political influences on competition policy; • Different forms of national enforcement of competition policy; • Differences resulting from viewing competition policy in specific national contexts with different cultures and traditions. This paper will discuss these categories of differences, and present arguments in support of internationalisation of competition policy. Thereafter, international competition policy objectives will be assessed in the perspective of divergences in national competition policies.
A. Differences in Competition Policy in a Strict Sense Three goals of competition policy should be distinguished. First is the decentralisation of economic power, which implies the protection of individual freedom and individual rights. Monopolies and cartels are departures from such individualism. In 1890, the Sherman Act was enacted in the U.S. to prohibit these kinds of restraints of competition. Second, and closely related to the first, is preservation of economic freedom of competitors, and not of competition. Coercion and discrimination are targets of this policy. Third, and economically
180
Working Paper IV
the most important, is assurance of consumer welfare, including both allocative and productive efficiencies. These objectives are valid not only for a domestic competition regime, but also for an international competition regime. In principle, they are in accordance with the free trade goal of international trade policy. However, a crossborder competition policy should focus on market access to permit penetration of national economies. This focus is evident in the European Community's competition policy to aid in establishment of the internal market. National perceptions differ as the result of the diverging economic theories that underlie national policies. The most evident clash is that between the Chicago School's focus on efficiency as the prevailing goal versus the focus on non-economic values. These divergences still exist. For instance, the United States just revised its Merger Guidelines, placing greater emphasis on productive efficiency. Moreover, consumer protection, the market position of small and middle-sized firms or the interest of economic sectors (e.g. with respect to distribution) are considerations that a national policy might rely upon as the basis for modification of strict competitive policies. Moreover, economic perceptions of the role of competition in regulated industries (e.g. transport) determine when national laws shall provide exemptions. Views also differ with regard to the effects on competition of vertical restraints, which is also reflected in national competition policies. The main players in the competition field are specialists, particularly economists, pressure groups or ministries. In international discussions, differing views should be easier to overcome than in the domestic realm, since specific national interests are not at stake. An agreement on consensus wrongs, such as naked cartels, should be possible. Accordingly, it seems that some degree of common understanding can be achieved with regard to internationalisation of competition policies. However, in some areas, such as mergers, policy will be heavily influenced by criteria that are determined according to policy backgrounds.
B. Differences Resulting from Political Influences on Competition Policy National antitrust authorities are inclined to allow mergers that might create 'national champions', considered to be ahead in international competition and to keep workers in their jobs. Moreover, technical progress or the promotion of environmental interests may be a reason to permit collaboration between competing firms. Arguments related to health policy or energy supply may also
4 - Competition Policy Objectives
181
have an impact on the assessment of restraints of competition. Such influences do not create conflicts within the field of competition policies; rather, they are exterior impacts. Policy-makers who are not competent in the field of competition policy thus exert their influence to temper or modify objectives of competition policy. Among the various nations, there are differences in the specific kinds of public interest considerations that are regarded to be of a higher importance than maintaining competition. In principle, such public policy concerns may influence all categories of restraints. For instance, horizontal agreements may improve rationalisation, or bring about progress in research and development. Export cartels may be used to foster national exports. Vertical restraints might be used to close national markets, which governments might view as a positive result. With regard to dominant firms, whether a misuse has occurred might be determined by applying non-competition criteria, or may be a vehicle for price control. The latter occurred in Germany in the 1970s, when the reach of misuse control was being debated. Merger policy is closely related to industrial policy, which is by its nature still national. It would be difficult to achieve agreement in these fields, where numerous differences exist. Any progress would require an understanding that competition policy is to be predominant over other policies, or that there will be limits to the impact that other policies can have on competition policy. However, the latter would not be possible, since the influence of public policy on competition policy reflects the importance to each nation of its competition policy as compared with its other public policies, both economic and non-economic. Therefore, agreement on the benefits of competition as an instrument to advance national and international welfare should be sought. This might seem Utopian, but such an evolution has occurred since 1958 in the European Community. At that time, no one could have imagined the high regard that the Member States would come to place on the benefits of competition. In the current timeframe, the Central and East European countries are following the E.C. model. Any expectation of such an understanding on a global level must be seen within the context of the arguments put forward in support of internationalisation of competition policies.
C. Different Forms of National Enforcement of Competition Policy The role of competition law enforcement is often neglected, but it is crucial. Even a perfect antitrust law would be a tiger without teeth if it were not applied effectively. Not surprisingly, national enforcement policies differ greatly, as
182
Working Paper IV
they reflect the respect for competition within a given economic order. In general, enforcement largely corresponds with the respective substantive rules. However, the role of enforcement must be further evaluated. Substantive rules expressly provide for the discretion of the relevant authority, particularly in the field of competition law. More importantly, the authorities enjoy a large degree of interpretational discretion when applying the complex economic terms of antitrust rules. What is the relevant market, when is afirmdominating, what is the meaning of substantially lessening competition? Significantly, the European Court of First Instance refrains from resolving such issues when it reviews Commission decisions. The Court considers only whether the Commission applied relevant criteria, acted within a given margin of discretion, and its decision does not contain evident errors. Therefore, a predominant aspect of competition policy is who decides and enforces competition laws. Moreover, experience has shown that agencies may be influenced by governments to restrict their activities in specificfieldswhich are close to trade policy or other government policies. The Japanese Fair Trade Commission is said to be under such pressure. In principle, a nation's enforcement policy reflects its basic understanding of competition. Enforcement activities may be executed in many ways. At one end of the spectrum, competition policy may fall completely under the rule of politics; at the other end, competition policy may be seen as a social and economic value comparable to other principal values, such as a stable monetary system, and characterised as a rule of law. Enforcement policy depends less on rules than on institutions. Therefore, the institutional aspects of competition policy must be included in the discussion of the impact of differences on the internationalisation of competition policy. In practice, three types of institutions may be distinguished: • • •
governmental institutions; agencies or commissions; courts.
Enforcement of competition policy by governmental institutions implies that it will be more or less strictly under the control of politics. The enforcement structure in the Netherlands and Luxembourg comes close to this form. However, such control need not be comprehensive. For instance, with regard to merger policy, France reserves enforcement to ministries. Most countries, however, have set up agencies or commissions that are institutionally separated from government and administration. This is designed to ensure the independence of decisions and to bring together the necessary pro-
4 - Competition Policy Objectives
183
fessional experience in law and economics. Great Britain and Germany have set up such independent institutions. Independent institutions were also established in other European countries that implemented their competition laws in accordance with E.C. rules. A key question with respect to such institutions is the degree of their independence, which depends on several factors, such as: who nominates the personnel; what is the duration of their appointment; whether they are appointed to a full time or a part time post; and what is the distribution of competences within the agency. The most objective procedure might be one that relies on courts to decide competition cases. This establishes the highest degree of separation from political influence. With such an enforcement structure, it is important that the competition laws are precise and clear, and that the competence to initiate court procedures is not excessively discretionary. For instance, the U.S. system relies heavily on courts to make decisions in competition cases. However, the role of the F.T.C. in this regard should not be underestimated. What are the consequences for internationalisation of competition policy which follow from these considerable divergences? For the moment, an autonomous international body which rules on the basis of the laws of a multilateral system should be excluded, since national enforcement policies cannot be disregarded. Therefore, a plurilateral system based on internationally agreed rules, which leaves the enforcement of those rules to national agencies and courts, would be an appropriate path to pursue. However, application of the rules by national authorities must be under international control. The Draft International Antitrust Code, developed by the so-called Munich Group, presents a proposal with detailed provisions for further discussion. Art. 19, Section 2, of the Draft Code offers an appropriate compromise between national and international interests in enforcement of international competition rules. It provides for the creation of an International Antitrust Authority, which can bring actions against national antitrust authorities in individual cases before national law courts whenever a national antitrust authority refuses to take appropriate measures against individual restraints of competition. The international authority would also have the right to sue private persons and undertakings before national courts, and to seek injunctions against restraint on competition, and the right to appeal within the national court system, even when the authority is not a party to the case, under the same conditions as parties.
184
Working Paper IV
D. Differences Resulting from Viewing Competition Policy in Specific National Contexts with Different Cultures and Traditions Differences in national cultures and traditions are important, since laws generally cannot be explained and evaluated simply by their wording, but instead must be seen within their social context. Laws vary with culture, although the ideas of justice and legal solutions are often similar. Laws may differ according to liberal or corporatist traditions, religious ideas, or social values. Competition policy and laws vary with culture, in the same way as other laws do, and cannot be regarded simply as a mechanical instrument to achieve specific results, especially if non-economic objectives are an essential aspect of competition policy. A competition system corresponds not only with free markets, but also with individualism, freedom of expression, free flow of information and non-intervention by governments in the market process. This is a liberal view of state and man, driven by self-interest as the supreme motivating force. There are, however, differences in the degree to which liberalism functions as the guiding principle. A long-standing tradition of state intervention in the economy generally implies that regulatory, interventionist models will be presented, especially with respect to systems that view individuals as only one aspect of a greater web of society. Some societies are based on a common understanding and tradition to adapt to life in groups. Behaviour is determined by social restraint and individual modesty, and laws and their enforcement reflect this culture. Other societies are characterised by an increasing corporatist trend. The power of associations, unions or other pressure groups has a tendency to create pressure to modify competition laws to meet their needs. Insofar as competition is closely related to culture and tradition, the possibility to harmonise competition policies is doubtful, as the roots of divergences might be deep. Conclusion The categories of divergences discussed above were intended to facilitate understanding of how to overcome national divergences in order to internationalise competition policies. In addition, a certain relation between the categories has emerged. Policies which rely on competition as an instrument to achieve specific goals, such as efficiency, consumer protection, or protection of small and medium-sized firms, will come close to systems which subordinate competition to other public-policy objectives, whose enforcement policy will
4 - Competition Policy Objectives
185
generally not be strictly competition-oriented. Whether competition-oriented regimes are less influenced by social values based on liberalism and a restricted role of government in the economy remains as a question for further study. The categories may shed light on what are the obstacles to an international antitrust code. The variety of national competition regimes must be assessed in light of the objectives of international competition policies. Two points should be considered. First, arguments for international policies must be considered in the context of the national divergences that exist. However, consideration of an international code does not require an initial analysis of national differences. A general understanding may be achieved concerning the need for internationalisation of competition policies. Such arguments go beyond the virtues of maintaining competition. On the basis of such understanding, discussion of the substance of a code will be possible. Only at this point do national divergences become relevant. Secondly, areas of convergence between national and international policies should be identified which would allow conflicts to be overcome more easily. As to the first point, a major argument favouring internationalisation follows from the success of the Uruguay Round. State barriers to cross-border trade were set aside to a considerable degree. This process will continue, as confirmed by the recent W.T.O. agreement to liberalise the telecommunications sector. This policy would be undermined by private firms, which would like to re-establish these barriers by mutual agreements or through the use of economic power. Internationalisation, therefore, seems inevitable. This argument, by its nature, goes beyond national divergences. Furthermore, the linkage of trade and competition policy, as for instance in thefieldof market access and anti-dumping, is a key issue in the present discussion to establish an international regime for maintaining competition. Another point of discussion relates to the need to provide a 'level playing field' for players in international markets. These issues are of common interest to nations, which should encourage a process to develop an international code and to overcome existing national divergences. The second category, the areas of conflict or concurrence between national and international policies, should be the subject of further discussion. For instance, policies that favour efficiency and show little concern for economic power will generally accept the idea of providing firms with a 'level playing field' through an international regime. Such converging interests might be found in other fields as well. Discussion of the categories of the divergences of national policies supports the need for further research in this field.
186
Working Paper IV
References DONALD BLACK, The Behaviour of Law (1976), 63.
EUROPEAN COMMISSION, Competition Policy in the New Trade Order: Strengthening International Cooperation and Rules, Report of the Group of Experts (1995). WOLFGANG FIKENTSCHER, WOLFGANG and ULRICH IMMENGA (eds.) Draft International
Antitrust Code (1995). ELEANOR FOX and A. ORDOVER JANUSZ, 'The Harmonisation of Competition and Trade
Law—The Case for Modest Linkages of Law and Limits to Parochial State Action', 19 World Competition 5 (1995). HELMUT GRONER and ANDREAS KNORR, Internationalisierung der Wettbewerbpolitik,
Festschrift fur Ernst-Joachim Mestmacker (1996). JOHN O. HALEY, 'Competition and Trade Policy: Antitrust Enforcement: Do Differences Matter!', in John O. Haley and Iyori (eds.), Antitrust: A New International Trade Remedy? (1995), 303. FRIEDRICH LUDWIG HAUSMANN, Staatliche Kartellrechtsdurchsetzung im Internationalen
Vergleich, Dissertation (1997). ULRICH IMMENGA, Rechtsregeln fur eine internationale Wettbewerbsordnung, Festschrift fur Ernst-Joachim Mestmacker (1996), 593. ALEXIS JACQUEMIN, Antitrust Economics—New Challenges for Competition Policy (1994), 27. ERNST-ULRICH PETERSMANN, 'Proposals for Negotiating International Competition Rules in the G.A. T. T.-W. TO. World Trade and Legal System', AuBenwirtschaft, Heft II/III (1994), 231. SONYA
MARGARET
WILLIMSKY,
Competition L. Rev. 54.
'The Concept(s)
of Competition',
[1997] Eur.
R. Shyam Khemani Private Sector Development Dept. The World Bank Washington, D.C., usA.
Rainer Schone1 Neue Ziircher Zeitung Zurich, Switzerland
A. Introduction The World Trade Organisation (W.T.O.) and its predecessor, the General Agreement on Tariffs and Trade (G. A.T.T.), traditionally have focused on governmental distortions of international competition that arise from restrictions on imports, such as tariffs, quotas, discriminatory practices, antidumping and state subsidies. The objective of the G.A.T.T./W.T.O. is to enhance global economic welfare through a predictable and liberal international trade order, which allows specialisation and fosters trade and investmentflowsacross national borders according to the comparative advantage of nations. The various rounds of multilateral negotiations during the postwar period have gradually reduced barriers to international transactions, thus enhancing market access and preventing unnecessary output restrictions to the detriment of consumers. With the resultant increased transactions between firms located in different countries, competition among firms involved in international commerce has increased. However, with growing competitive pressures, firms may also have increased incentives to engage in practices that restrict competition, particularly since it is more difficult to obtain protection through government actions. Given the W.T.O.'s general objective to facilitate removal of restrictions to international market access, both anticompetitive measures adopted by governments and by private agents logically fall in its mandate. Following the constant reduction of government imposed barriers in international trade, 1
The authors are at The World Bank and Neue Ziircher Zeitung respectively. The views expressed in this paper are their own and do not necessarily reflect those of their institutional affiliation. Valuable comments and suggestions on a earlier draft were provided by Bernard Hoekman, and by the participants at this Workshop.
188
Working Paper V
international cooperation must now be extended to antitrust policy to address trade barriers erected by private restrictive business practices. The G.A.T.T.AV.T.O. currently does not contain either specific rules or general principles applicable to anticompetitive behaviour by private firms. Given the W.T.O.'s mandate to improve market access in international trade, there is a case to extend its discipline to restrictive business practices, if remedies currently available are insufficient to secure access to national markets. In some cases, national antitrust laws might well be able to deal with barriers to entry which prevent market access and adversely affect international trade. However, in other cases, no relief or insufficient relief may be available under current law(s). As a result, if existing national competition rules prove insufficient to deal with restrictions of market access by private firms, a case arises for developing new international rules within the W.T.O. regarding anticompetitive conduct by private agents. The aim of this paper is to develop a proposal regarding the manner in which the W.T.O. could address restrictive business practices in international trade.2 Recently, a number of proposals have been put forward regarding the appropriate terms of reference, governing principles, mandate and scope of the work which the W.T.O. can/should embark upon in this area.3 The approach adopted in this paper is somewhat different, in that it is case-oriented. We identify cases where anticompetitive behaviour by private firms restricts market access in international commerce, thus resulting in a welfare-reducing restraint of output. We subsequently discuss current remedies and their inadequacy in order to suggest possible improvement at the national level and within the W.T.O. We believe there is a need to address international restrictive business practices only where market access is restricted and there is insufficient relief available through existing legal-institutional arrangements. Market access is a basic tenet that both trade and competition policy experts would embrace. To trade economists, enhanced market access further improves theflowof trade and investment. To competition policy economists, enhanced market access increases actual and/or potential competition and improves consumer welfare. Securing and/or enhancing market access, there2
At the conclusion of the Uruguay Round in Marrakesh in 1994, several countries expressed their desire that the future work agenda of the W.T.O. should include antitrust issues. As discussed below, the Singapore meetings of the W.T.O. held in Dec. 1996 resolved that a working group should be established to study the interface between trade and competition. 3 A useful summary prepared by the Global Forum on Competition and Trade Policy can be found in Current Proposals for an International Code, 24 Int'l Bus. Law 448 (1996).
4 - Competition Policy Objectives
189
fore, represents an important point of confluence between trade and competition policies. Measures which generally foster 'global contestability of markets' will be consistent with the objectives of both trade and competition policies.4 Our main concern, therefore, relates to making trade liberalisation and antitrust remedies complementary, where anticompetitive practices stem from private restrictions of market access. Trade policy instruments, such as tariffs, safeguards, antidumping, countervailing duties, and voluntary export restraints, erect barriers that do not foster competition by restricting market access. Trade-policy instruments usually protect domestic competitors from foreign competition, whereas antitrust policy protects competition. Consequently, there is a need to augment trade policy instruments with antitrust policy where private firm behaviour restricts market access—a goal that is consistent with the mandate of the W.T.O. of liberalising international trade and fostering open markets. As a result, we suggest that a distinction should be made between restrictions to market access by governments on the one hand, and by private firms on the other hand. Market access restrictions should be addressed by both the W.T.O. and the countries affected. Restrictions caused by governmental policies are trade policy issues, and require a solution between governments at the international level. This clearly falls under the present mandate of the W.T.O. By contrast, private restrictions pertain to interfirm rivalry, and should (at least first) be addressed by national antitrust laws. It is not desirable that governments represent the competitive interests of private firms in disputes on anticompetitive conduct between private firms. If a government gets involved in enhancing the competitive opportunities of particular domesticfirmsvis-a-vis foreign rivals, this is an act of strategic trade policy, which may distort international competition. Therefore, direct government involvement should be limited to trade relations issues, where the conduct of foreign governments creates barriers to market access. Governments should be concerned with establishing a generally supportive environment for firms to compete on their respective merits, which includes providing the means for private firms to resolve conflicts as to inter-firm rivalry under antitrust laws. In this connection, we believe that all countries, whether industrialised, developing or in transition from communism or central planning to a market oriented economic system, must have effective competition laws and policy in place. 4
See Robert Z. Lawrence, Towards Globally Contestable Markets, paper prepared for the O.E.C.D. trade committee (Paris, February 1995). Lawrence argues that it is not necessary to harmonise policies in all areas such as competition policy, government regulation, technology policy, government procurement, corporate governance, standard setting and tax policies. National diversity should be respected but transparent, minimum rules and operational procedures aimed at market access are desirable.
190
Working Paper V
The issue of anticompetitive practices in international commerce is not new.5 There have been various efforts by the international community to subject antitrust problems to global rules,6 but none has proven to be an overall success. The first attempt to make restrictive business practices subject to international discipline was the 1948 Havana Charter, which aimed to establish the International Trade Organisation (I.T.O.).7 The Havana Charter included, inter alia, (1) an obligation for members to take appropriate measures against restrictive business practices, i. e. to adopt national antitrust laws; (2) a list of prohibited restrictive business practices, such as price-fixing, exclusionary and discriminatory practices; and (3) a non-binding international dispute settlement system applicable to antitrust enforcement against the prohibited R.B.P.s, including extensive investigative powers of the I.T.O. to demand information.8 The Havana Charter was, however, not approved by national governments. 5
Eleanor M. Fox, Competition Law and the Agenda for the W. T. O.: Forging the Links of Competition and Trade, 4 Pac. Rim & Pol. J. 1, 4 (1995). Fox states that 'antitrust has periodically surfaced on the world agenda. In the 1940s, the concern was driven by the horror of world cartels that were used by fascist governments to reinforce their power. In the 1960s and 1970s, the concern was triggered by the growth of multinational corporations and their perceived social, political and economic power to shut off opportunity for outsiders and to repress less developed nations. The 1980s witnessed an unleashing of world competition along with freer trade, dissipating much of the market power that large corporations had accumulated in the 1960s. The years 1989 and 1990 saw attempts to construct free enterprise economies from the ashes of communism. Freer competition and trade continues to break down economic and political power and to create incentives for new forms of nationalism and protectionism, on the one hand, and for new avenues for progressive, integral world transactions, on the other hand.' We could add that in the 1990s, one of the most obvious concerns is about perceived anticompetitive behaviour by import-competing companies, especially in Japan, which restricts market access to foreign exporters resulting in complaints on 'unfair' trade conditions. 6 See, e.g., John H. Jackson, Alternative Approaches for Implementing Competition Rules in International Economic Relations, 49 AuBenwirtschaft 177, 191-3 (1994) hereinafter 'Jackson 1994'; John H. Jackson, Statement on Competition and Trade Policy before the US Senate Committee on the Judiciary, 26 J. World Trade 111 (1992) (hereinafter 'Jackson 1992'); Ernst-Ulrich Petersmann, Proposals for Negotiating International Competition Rules in the G.A.T.T.-W.T.O. World Trade and Legal System, 49 AuCenwirtschaft 231, 238-9 (1994) (hereinafter 'Petersmann 1994c'); ErnstUlrich Petersmann, Competition Policy Aspects of the Uruguay Round (1994) (hereinafter 'Petersmann 1994b') 3-4; Fox, supra note 5. 7 United Nations Conference on Trade and Employment, Havana Charter for an International Trade Organization, Final Act and Related Documents (1948). 8 Id., Chapter V; Ernst-Ulrich Petersmann, Competition Elements in International Instruments (1994) (hereinafter 'Petersmann 1994a'); Petersmann 1994c, supra note 6, at 238-9; Petersmann 1994b, supra note 6, at 3; Jackson 1994, supra note 6, at 191-3; Jackson 1992, supra note 6, at 113-14; Fox, supra note 5.
4 - Competition Policy Objectives
191
After the failure of the draft I.T.O., subsequent discussions and attempts were held both within and outside the newly created General Agreement on Trade and Tariffs (G.A.T.T.), but in the end, these did not lead to the inclusion of restrictive business practices within the G.A.T.T. The most important achievements in addressing restrictive business practices in international trade were the 1976 O.E.C.D. Guidelines for Multinational Enterprises, several other O.E.C.D. guidelines on cooperation in antitrust matters, and the 1980 U.N.C.T.A.D. Set of Principles on Restrictive Business Practices. All of these codes, however, were non-binding and virtually non-enforceable. The G.A.T.T. Uruguay Round did not directly address anticompetitive business practices by private firms that restrict international trade.9 Nevertheless, some Uruguay Round agreements contain provisions regarding private trade restrictions. The Agreement on Trade-Related Intellectual Property Rights (T.R.I.P.s) recognises that intellectual property rights (I.P.R.s) may be abused in anticompetitive ways, but leaves members a choice whether to adopt national laws to control anticompetitive abuses.10 The Agreement on TradeRelated Investment Measures (T.R.I.M.s) provides that a review within five years shall determine 'whether [the agreement] should be complemented with provisions on investment policy and competition policy'.11 The General Agreement on Trade in Services (G.A.T.S.) contains a requirement that monopoly service suppliers that have been granted exclusive rights by their government do not abuse their monopoly position outside the scope of their monopoly privilege, but makes no provisions regarding anticompetitive
9
Some developing countries tried to bring restrictive business practices by multinational firms on the agenda of the Uruguay Round multilateral trade negotiations, but no consensus could be reached on the inclusion of those issues. At the opening of the G.A.T.T. Uruguay Round international trade negotiations, the Chairman of the 1986 Ministerial Conference at Punta del Este noted that the issue of restrictive business practices has been raised by some countries, but that no consensus on including those issues was reached. G.A.T.T. Activities 1986; Petersmann 1994a, supra note 8, at 9; Richard Blackhurst, Competition Policies: National Versus Multilateral Jurisdiction, 49 AuBenwirtschaft 223, 227 (1994). 10 Agreement on Trade-Related Aspects of Intellectual Property Rights, 15 Apr. 1994, Arts. 8:2 and 39^40; Petersmann 1994c, supra note 6, at 247-56; Petersmann 1994b, supra note 6, at 8-14. 11 Agreement on Trade-Related Aspects of Investment Measures, 15 Apr. 1994, Art. 9; Patrick Low and Arvind Subramanian, T.R.I.M.s in the Uruguay Round: An Unfinished Business? (1995), 1-10 (unpublished memeo, on file at the World Bank, International Economics Department); Petersmann 1994c, supra note 6, at 247-56; Petersmann 1994b, supra note 6, at 8-14.
192
Working Paper V
behaviour by other private service suppliers, whose occurrence the G.A.T.S. merely recognises.12 Recently, the issue of the interface between trade and competition policy has received renewed impetus from the W.T.O. At the conclusion, of the Singapore Ministerial Meetings (13 December 1996), a resolution was passed to 'establish a working group 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 W.T.O. framework'.13 However, beyond these global approaches, there are already several regional trade agreements that have been concluded between U.N.C.T.A.D. and O.E.C.D. member countries, which include antitrust rules and/or provisions for the coordination and application of national antitrust policies. The E.U., the A.N.D.E.A.N. Pact countries, Australia-New Zealand Closer Economic Relations Trade Agreement (A.N.Z.C.E.R.T.A.), and the North American Free Trade Agreement (N.A.F.T.A.) are among such examples. In addition, there are several bilateral agreements, such as the Canada/U.S. agreement, and the E.C./U.S. agreement. While we do not review these regional or bilateral agreements in detail, we will mention these models during the ensuing discussion wherever they are relevant.14 These models contain useful features that should be considered within the W.T.O. The organisation of this paper is as follows. Section B will argue that international trade liberalisation in and of itself is not sufficient to discipline anticompetitive behaviour by private firms, and therefore cannot replace effective antitrust enforcement. As we are focusing on restrictions of market access 12
General Agreement on Trade in Services, Part II, Arts. VIII and IX, in Rusults of the Uruguay Round of Multilateral Trade Negotiations: The Legal Texts (Geneva, 1994). Art. VIII and IX; Petersmann 1994c, supra note 6, at 247-56; Petersmann, 1994b, supra note 6, at 8-14. Moreover, the Uruguay Round Agreement on Preshipment Inspection provides rules for private preshipment inspection firms; the Agreement on Technical Barriers to Trade includes provisions to ensure that private bodies do not adopt or apply technical regulations and standards that are more trade-restrictive than necessary. Petersmann 1994c, supra note 6, at 247-56; Petersmann, 1994b, supra note 6, at 8-14. 13 Singapore Ministerial Declaration WT/MIN(96)/DEC. WTO/(1996). Para. 20. 14 See, e.g., Petersmann 1994a, supra note 8; Bernard M. Hoekman and Petros C. Mavroidis, Competition, Competition Policy and the Gatt (1994) (Discussion Paper, Centre for European Policy Studies, London); Richard Harmsen and Michael Leidy, Regional Trading Arrangements, in R. Shyam Khemani (ed.) International Trade Policies: The Uruguay Round and Beyond (1994) (Background Papers, vol. II, I.M.F., Washington, D.C.).
4 - Competition Policy Objectives
193
through anticompetitive conduct with impact on international trade, the three main sections that follow deal with the three areas where such restrictions of market access occur, namely (1) market access restriction by import-competing firms in their home market that exclude foreign rivals (Section C), (2) restriction of market access by exporters in the importing country (Section D), and global non-market-specific restriction of market access due to spillovers between national antitrust policies (Section E). In addition, in all three areas, governments may tolerate and promote market access restrictions by domestic firms to enhance their competitive position vis-a-vis foreign companies for strategic trade policy reasons, which we will subsequently discuss as 'strategic antitrust policy' (Section F). In a final section, we will address the argument that differences in national antitrust approaches create an uneven playing field for competition between companies located in different jurisdictions (Section G). Competition policy may include a variety of policies that directly affect the behaviour of enterprises and the structure of industry. An appropriate competition policy, as we define it, includes both (1) policies that enhance competition in local and national markets, such as trade liberalisation, relaxation of foreign investment and ownership requirements, and economic deregulation; and (2) competition law, designed to prevent anticompetitive business practices by firms (antitrust policy) and unnecessary government intervention in the marketplace. In this study, we focus on antitrust policy, which includes the regulation of cartel agreements, monopolisation, abuses of a dominant position and mergers.
B. Why Trade Liberalisation Cannot Replace Antitrust Enforcement15 In many economies where high levels of industrial concentration exist, anticompetitive business practices are less feasible if domestic markets are exposed to international competition. In the absence of barriers to trade, domestic monopolists or oligopolists lose their ability to exercise market power irrespective of actual imports' share of the domestic market, in view of the threat of potential competition.16 15 This section is in part derived from R. Shyam Khemani and Mark Dutz, 'The Instruments of Competition Policy and Their Relevance for Economic Development', in Frischtak (ed.), Regulatory Policies and Reform: A Comparative Perspective, draft on file at the World Bank (December 1995). 16 See H.C. Eastman and S. Stykolt, A Model for the Study of Protected Oligopolies, 70 Econ. J. 336 (1960); Jagdish Bhagwati, On the Equivalence of Tariffs and Quotas, in R. Baldwin (ed.) Trade, Growth and the Balance of Payments (1965).
194
Working Paper V
The view that imports limit market power is supported by studies that find differing degrees of convergence between domestic and international prices in the face of trade liberalisation, and a negative relationship between price and cost or profit margins and imports.17 Several recent empirical studies, however, suggest that effects of trade liberalisation may be less significant than previously thought, raising questions about the true effect of trade liberalisation on competition.18 The procompetitive effects of tariff reductions may be diluted if the import supply is not very elastic. This occurs when increased demand for imports can be met only at significantly higher prices, or when imports are comparatively insensitive to changes in domestic prices. It is possible to construct various economic models where imports meet only a small part of domestic demand and are provided by a fringe of firms, which domestic oligopolists have already taken into account in their pricing strategies.19 In addition, in an environment of floating exchange rates, if domestic firms fail to rationalise high cost operations and improve productivity, the domestic currency is likely to depreciate, offering new protection from import competition. In Mexico, for example, recent currency devaluations have more than offset the tariff reductions negotiated under N.A.F.T.A. Furthermore, trade policy consists of more than tariff policy. Quotas, voluntary export restraints, antidumping and countervailing duties are among the instruments that governments can wield to limit import competition. While a true 'free trade' policy would require that all such measures are removed, in reality this never happens. Indeed, as import tariffs are liberalised, the pressures to invoke other measures only increase. Over the past few years, as more countries have liberalised trade policies, a simultaneous movement has occurred to put in place systems of protection against dumping and subsidies. By the end of the 1980s, at least 50 countries had become new signatories or observers of the G.A.T.T. antidumping code, and in the period from 1980 until
17
See, e.g., Frederich Scherer and D. Ross, Industrial Market Structure and Economic Performance 424-426 (1990). 18 See S. Globerman, Trade Liberalization and Competitive Behavior: A Note Assessing the Evidence and the Public Policy Implications, 9 J. Pol. Analysis & Mgmt. 80 (1990); E.E. Learner, Cross-Section Estimates of the Effects of Trade Barriers, in R.C. Feenstra (ed.), Empirical Methods for International Trade (1988); A. Fishlow, The Latin American State, 4 J. Econ. Perspectives 61 (1990). 19 See M. Perry and R. Porter, Oligopoly and the Incentive for Horizontal Merger, 75 Am. Econ. Rev. 219 (1985); T.W. Ross, Movements Towards Free Trade and Domestic Market Performance with Imperfect Competition, 21 Can. J. Econ. 507 (1988).
4 - Competition Policy Objectives
195
1992, more than 2,000 antidumping and countervailing duty actions were taken.20 Even if trade barriers such as antidumping are eliminated, other factors can impede the procompetitive effect of trade liberalisation. First, an increasing share of economic activity in developing as well as industrial countries relates to non-tradable goods and services. These include high weight-to-value products with high transport costs (such as cement and steel), perishables (such as food), and legal, financial, and other services.21 Secondly, in the absence of effective competition, domestic firms can raise prices up to the international price plus transport costs and still keep out imports. Thirdly, differences in income, tastes and culture and in product safety, consumer protection and technical standards may also separate markets. Fourthly, inter-firm contractual arrangements and vertical integration may prevent the development of new sources of inputs or new distribution channels. This problem has been cited by many American firms as limiting their ability to gain access to markets in Japan, and forms part of the 'framework' and 'structural impediments initiatives' discussions between the U.S. and Japanese governments. Fifthly, international cartels may divide up markets through price-fixing or geographic 20
See Antidumping: How it Works and W h o Gets Hurt (J. Michael Finger ed., 1993); Patrick Low and Arvind Subramanian, Trade Protection in Agriculture: A Special Casel (1993) (mimeo on file with the World Bank, International Economics Department, Washington, D.C.); Bernard M . Hoekman and Petros C. Mavroidis, Dumping, Antidumping, and Antitrust, 30 J. World Trade 27 (1996). It is often suggested that enacting an antidumping law does not generally serve the best interests of a country. Such a law can thwart the benefits of trade liberalisation; it tends to be used to protect (inefficient) competitors rather than competition. Consumers pay higher prices while having reduced product choice and quality. Provisions in competition laws dealing with predation and price discrimination offer a reasonable substitute to remedy dumping practices by firms. Two main motivations appear to be responsible for the trend toward the use of antidumping. First, as governments lower tariffs and nontariff barriers, they come under increasing pressure from adversely affected import-competing domestic producers. Antidumping duties are a convenient substitute mechanism for providing protection to firms that claim injury. Secondly, with increased trade liberalisation, many countries joined the G.A.T.T. for the first time in the 1980s. Following the example of such economies as the U.S. and the E.U., these countries adopted the full arsenal of provisions permitted by the Antidumping Code to ensure that they respond to similar protective trade practices by other governments and that domestic firms have a set of protective remedies similar to those of their foreign counterparts. 21 T h e G.A.T.s would help in this regard by dismantling 'within border' regulatory barriers. Almost a quarter of world trade is service-related. See H . G . B r o a d m a n , G.A. T. T: The Uruguay Round Accord on International Trade and Investment Services, 17 World Econ. 281(1994).
196
Working Paper V
market-sharing agreements. Importers and foreign firms have, in fact, often been charged with precisely this offence in several jurisdictions. In concentrated industries such as pharmaceuticals, petro-chemicals, and telecommunications equipment, where the total number of firms worldwide is small, such arrangements are particularly common. Some empirical examples/cases serve to illustrate these points.22 In Colombia, the leading brewer allegedly has geographic market sharing agreements with existing and potential competitors in neighbouring countries. It also owns and controls the sole bottle manufacturing plant, and has exclusive dealing clauses with the great majority of distributors. Recently, the same dominant Colombian brewer acquired all the breweries in Ecuador. A leading U.S. biscuit manufacturer found it difficult to enter the Colombian market due to exclusive distribution clauses between the dominant domestic manufacturer and major retailers. It has instead entered into licensing and joint marketing arrangements with the dominant firm because of the barriers these restrictive distribution agreements posed to both importation and local manufacture. In many product areas, such as fertilisers, pharmaceuticals and plate glass, domestic and international price differentials persist despite trade liberalisation, due to manufacturing and marketing linkages between domestic and foreign firms. In Ecuador, government enterprises and private sector counterparts are alleged to engage in tacit price and market sharing agreements in cement and steel. In the case of many other industrial products, the government continues to regulate the market through price controls. The industry associations for domestic oil and pharmaceuticals have successfully obtained a government mandate to limit entry, coordinate and increase prices. While importation of automobiles has been liberalised, distribution remains the exclusive area of government-owned or appointed dealers. In the European Union, complaints of discriminatory practices have covered products such as computers, aircraft, high-speed trains, power generation and telecommunication equipment. In Indonesia, importation, production and/or domestic marketing of products as diverse as automobiles, oil and gas, rattan, steel, citrus fruit and wheat remain with state designated monopolies. In Japan, antitrust complaints have related to products such as automobiles and parts, film, soda ash, plate glass, and cellular telephones. For these and other such reasons, liberalised trade cannot effectively substitute for competition (antitrust) law. The two policy areas should be viewed as complementary. Chapter 15 of N.A.F.T.A. explicitly recognised the view that 22
Based on author's advisory work in several of these countries.
4 - Competition Policy Objectives
197
competition law can foster trade and investment. To enhance business and investor confidence, Mexico decided to modernise its competition law prior to N.A.F.T.A. In a world where multinational companies have grown accustomed to operating under competition laws, the absence of such a law or a poorly designed one in a country can act as a barrier to trade and foreign investment. A pending matter in Brazil is illustrative of this point. Colgate Palmolive recently acquired a 'dominant' Brazilian health care products manufacturer, Kolynos. Through exclusive dealing and vertical arrangements, Kolynos is reputed to control 80 percent of the toothpaste market. Another U.S. multinational enterprise, Proctor and Gamble (P&G), has argued that market penetration is difficult due to anticompetitive business practices, including various interfirm contractual arrangements in a highly concentrated wholesale-retail distribution network, and weak competition law enforcement by Brazilian competition authorities. It is reconsidering its medium- to longterm investments, including closing its subsidiary operations in Brazil. While the current dispute between Kodak and Fuji films in Japan also relates to alleged non-enforcement of competition law and differential treatment of U.S. v. Japanese firms, the Brazilian case illustrates how two firms of the same nationality may pre-empt each other in international markets. Of course, the merits of each case, as well as the nature of the competitive harm, have yet to be established. As local enterprises begin to operate more in international markets, a national law that converges or is in harmony with the national laws of competitors makes it easier for them to adapt. It also spares foreign firms any additional hurdles in their business activities.23 In this connection, a number of bilateral and multilateral approaches to the enforcement of competition law have been forged by different countries. Canada and the United States have signed a memorandum of understanding (M.O.U.) and a mutual legal assistance treaty (M.L.A.T.). M.O.U.s have been drafted, though not yet finalised, between Canada and the European Union and the United States and the European Union. Twenty-four member nations are signatories to the 23
E.g., the merger between Gillette and Wilkinson, two multinational razor blade manufacturers, led to a simultaneous review of the transaction by authorities in 14 jurisdictions. The focus and nature of the reviews varied extensively across the 14 jurisdictions. While in the end the deal was allowed to proceed with minor alterations prompted by different authorities, the delays, procedures, and compliance costs the firms experienced were far from trivial. See Richard Whish and Diane Wood, O.E.C.D. Merger Process Convergence Project, D.A.F.F.E./C.L.P./W.P.3(93)6 (1993). The Whish/Wood study reviews 8 other case examples involving such firms as Westinghouse Electric-A.B.B., Matsushita-M.C.A., Renault-Volvo, and Fiat-Ford.
198
Working Paper V
O.E.C.D. Council Recommendations Concerning Cooperation on Restrictive Business practices (1986). The U.N.C.T.A.D. Set of Multilateral^ Agreed Equitable Principles and Rules for the Control of Restrictive Business Practices (1980) is another multilateral agreement. These arrangements are designed to facilitate cooperation and information sharing. While these measures do not involve complete harmonisation or the development of common principles for administering and enforcing competition law, they contribute towards reducing business transaction costs that may arise from frictions between different legal/economic systems. Other measures may also be worth exploring. For example, in most jurisdictions, competition law enforcement rests within the discretionary authority of an appointed body. Permitting private actions and granting standing to foreign firms affected by anticompetitive arrangements may provide channels for relief.24 C. Restriction of Market Access by Import-Competing Firms Following our basic distinction between trade policy issues and inter-firm rivalry issues, this section will discuss how restrictions of market access by import-competing firms should be dealt with, and what should be done at the international level to bring sufficient relief where currently no adequate remedy is available. The restrictions we will examine in this section, which have the effect of impeding market access of foreign exporters, are mainly due to vertical restraints, such as vertical integration, exclusive dealing, exclusive distributorship and resale price maintenance. Restriction of access to a foreign market may be the result of different factors, including governmental barriers to trade and private exclusion of foreign competitors, as well as a mix of private and governmental action. An adequate solution to problems pertaining to such barriers to trade depends on identifying the sources of restricted market access. This requires examining whether limited market access is due to anticompetitive behaviour of private firms on the one hand, or governmental measures which support or tolerate such restrictions, on the other hand. 24
In the Kodak-Fuji dispute, it has been pointed out that although the Japanese Fair Trade Act allows private actions, the permission of the Japanese Fair Trade Commission (J.F.T.C.) is needed before such an action can be initiated. In the 50 years since this provision was adopted, only seven such cases have been permitted, none of which was successful. It has been argued that the J.F.T.C, in effect, has the complete power to extinguish private rights of action. In the E.U., there is no provision for private actions. In other jurisdictions, the nature and scope of private actions varies from non-existent to widely (perhaps overly) used, such as in the U.S.
4 - Competition Policy Objectives
199
Where access to a foreign market is restricted, we suggest enforcement of national antitrust law in the importing country, where market access barriers are due to private firm conduct (interfirm rivalry issue); and an international solution among governments, where market access barriers are due to government conduct (trade policy issue). Where it is alleged that private anticompetitive behaviour by importcompeting firms impedes market access to exporting firms, the first step must be tofilean antitrust complaint or law suit in the importing country. However, several circumstances may obstruct the pursuit of this avenue. We can identify five factors that limit the possibility of antitrust relief in the importing country: (1) legal standing of foreign companies; (2) legal standing of private parties to file an antitrust case; (3) discriminatory non-enforcement of antitrust law; (4) discriminatory enforcement of antitrust law; and (5) differences in substantive antitrust standards. Rather than generating trade disputes between governments, we suggest opening the way for effective antitrust enforcement in the importing country. Exporters that are denied market access to foreign markets must have the means of obtaining an antitrust remedy against restrictive business practices in the importing country. The assistance of the exporters' home government in a particular case will not always be necessary. Where possible, focus must be on the creation of ways and instruments that enable competitors to resolve disputes pertaining to interfirm rivalry under antitrust laws. An international agreement within the W.T.O. should address impediments to effective antitrust enforcement in the importing country, thus avoiding future governmental intervention in specific cases. The sections that follow will discuss the types of problems that may arise with antitrust enforcement in the importing country, and suggest how international rules could be established to remove these impediments.
1. Legal Standing of Foreign Companies Exporters may not have legal standing in the importing country. This prevents them fromfilingan antitrust complaint with foreign antitrust agencies or a lawsuit before an independent court. In such cases, no antitrust remedy will be available for foreign exporters in the importing country. At times, the lack of legal standing may result because an exporter does not have a subsidiary in the importing country. In such a case, establishing a subsidiary in the importing country could grant the exporting company legal standing and access to antitrust remedies. However, the possibility of setting
200
Working Paper V
up a subsidiary depends on domestic regulations, such as those pertaining to registration and foreign direct investment, which may obstruct the pursuit of this avenue. As a result, there may be a case for granting the same rights of legal standing as domestic competitors have to foreign firms that are affected by anticompetitive behaviour of import-competing industries, even without domestic subsidiaries.25 An international agreement on legal standing of foreign firms would give exporters equal access to domestic antitrust remedies in the importing country.
2. Legal Standing of Private Firms to File an Antitrust Case A further reason that an exporter may lack legal standing in an importing country is the absence of right of any private party to request initiation of an antitrust investigation. Some national antitrust enforcement procedures rely completely on administrative enforcement, not containing any rights for private parties, either domestic or foreign, to demand an investigation and enforcement action. This is the case with Japan, for example, where private parties have neither the right to compel the Japanese Fair Trade Commission (J.F.T.C.) to initiate an investigation, nor tofilea private lawsuit alleging anticompetitive behaviour before an independent court. Rights of legal standing of private parties under the Japanese Antimonopoly Act (A.M.A.) are limited to (1) reporting to the J.F.T.C. about anticompetitive practices, with no right to compel the agency to initiate an investigation, and (2) suing another party for damages caused by anticompetitive behaviour, once an administrative decision has been rendered determining that the practices are in violation of antitrust law.26 There are no remedies available to exporters if antitrust authorities have discretion as to whether to initiate an antitrust investigation, and they decide not to prosecute a case. Private parties should have the possibility both to demand an administrative investigation and tofilea private lawsuit. The latter is especially important for foreign firms. Exporters might sometimes be hesitant to file a complaint with administrative antitrust authorities, because they fear that those agencies are not independent enough to render impartial judgement between domestic and 25 Ideally, hurdles to the establishment of subsidiaries should be reduced, including excessive registration requirements as well as foreign direct investment controls. 26 A n t i m o n o p o l y a n d Fair trade Maintenance Act, Japan, Arts. 14(1) 25(1947). Takashi B. Y a m a m o t o , Japan, in James J. Garrett (ed.), World Antitrust L a w a n d Practice (1995), 33-45.
4 - Competition Policy Objectives
201
foreign firms. Independent courts have a better reputation of being able to apply the law in a nondiscriminatory manner. The following measures would render effective antitrust enforcement at the request of a private party and sustain exporters' trust in impartial enforcement in the importing country, thus decreasing the pressure for trade policy measures against the importing country, such as those according to 'section 301' in the U.S.27: (1) domestic, as well as foreign, firms affected by anticompetitive conduct should be granted legal standing to request the initiation of an antitrust investigation; (2) private parties should have a choice either to complain to administrative antitrust agencies or tofilea case before an independent court28; (3) countries should take measures to guarantee that administrative antitrust authorities are politically independent.29
27
T h e E . U . G r o u p of Experts suggests an international rule of 'positive c o m i t y ' , which includes a legal right of antitrust authorities of one country to request the initiation of an administrative antitrust investigation in another country. E u r o p e a n Commission, Competition Policy in the N e w T r a d e Order: Strengthening I n t e r n a t i o n a l Cooperation (Office for Official Publications of the European Communities, 1995) (hereinafter 'Van Miert Report'). Nevertheless, we believe it is indispensable t h a t adversely affected private parties that are denied market access have legal s t a n d i n g directly to request the initiation of a case without a need to lobby their home a n t i t r u s t authorities, in order t o act via positive comity obligations. Only where h a r m of restrictive business practices affects consumers, as opposed to competitors, in another c o u n try, m a y there be a need for intervention by antitrust authorities of the affected c o u n t r y to request enforcement of antitrust laws in the country where anticompetitive b e h a v i o u r originates. Positive comity has the following disadvantages in comparison to legal standing of foreign exporters: (1) affected firms must lobby their antitrust authorities t o make use of their positive comity rights a n d request antitrust enforcement in a foreign country; (2) antitrust authorities in the affected country have discretion w h e t h e r t o request antitrust enforcement abroad; (3) antitrust authorities in the requested c o u n t r y have discretion whether to initiate a case; and (4) antitrust authorities in the r e q u e s t e d country may not have an incentive t o enforce their antitrust law if enforcement benefits foreign firms only by restricting domestic firms. 28 Petersmann stated that '[a Plurilateral Agreement on Competition a n d T r a d e ] should . . . guarantee private access to domestic competition authorities a n d c o u r t s ' : Statement of Ernst-Ulrich Petersmann in the Van Miert Report, supra note 27, at 3 0 . I n contrast, the M a x Planck Draft International Antitrust C o d e relies exclusively o n administrative antitrust enforcement. It suggests granting legal standing to affected p a r ties to request an administrative antitrust investigation while ensuring that n a t i o n a l administrative antitrust agencies are politically independent (Draft I n t e r n a t i o n a l Antitrust Code, Art. 17, Sees. 1 (b) & 3 (b) (hereinafte ' D I A C ' ) ; Josef Drexl et al., Draft International Antitrust Code, 64 Antitrust & T r a d e Reg. Rep. 1629 (1993). 29 See D . I . A . C , supra n o t e 28, Art. 17 Sec. 1 (b); Drexl et al., supra n o t e 28.
202
Working Paper V
3. Discriminatory Non-Enforcement of Antitrust Law In a country where private parties have no rights to demand initiation of an antitrust investigation, the decision whether a case will be prosecuted depends exclusively on the discretion of administrative antitrust enforcement agencies. There is no legal discrimination against foreign companies, since both foreign and domestic firms have no legal standing to initiate an antitrust proceeding. However, there may be discrimination if antitrust authorities refuse to open an investigation, which is aimed at protecting domesticfirmsagainst increased market access by foreigners firms. Non-enforcement of antitrust law can also be discriminatory if enforcement is lax in cases where foreign firms would benefit from enforcement action, while being rigorous when domestic firms benefit. The discriminatory denial to initiate an antitrust investigation could be the subject of an appeal before the W.T.O. for violation of the national treatment requirement. As established in Art. 111:4 of the G.A.T.T., national treatment requires equal treatment of domestic and foreign products under internal domestic regulations once an imported product has entered the market.30 An affected country can file a complaint under Art. XXIII: 1 (a) of the G.A.T.T. with respect to any violation of national treatment. If antitrust authorities refuse to initiate an antitrust investigation following a complaint by a foreign firm, in obvious discrimination against foreign goods, this could provide grounds for a W.T.O. appeal.31 A violation complaint will initiate a regular 30
G.A.T.T. Art. 111:4 states:
'The products of the territory of any contracting party imported into the territory of any other contracting party shall be accorded treatment no less favourable than that accorded to like products of national origin in respect of all laws, regulation and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use.' 31
Alternatively, a country may seek to file a non-violation complaint as set out in Art. XXIII: l(b) of the G.A.T.T. if none of the explicit G.A.T.T. rules is violated, but where the country considers its negotiated rights of market access nullified or impaired by not expressly prohibited governmental measures. A non-violation complaint protects exporting countries against governmental measures in importing countries that 'nullify or impair' their 'expectations on the competitive relationship between imported and domestic products' based on mutual market opening concessions. As a consequence, only unforeseen governmental measures can impair reasonable expectations, which are taken subsequently to the agreement o n mutual market opening concessions. A non-violation complaint is therefore possible only where a country turns to a discriminatory antitrust policy only after having committed itself to a certain level of openness. Thus, lax antitrust enforcement that has always been lax is not eligible for a non-violation complaint. Three conditions must be met in order to file a non-violation
4 - Competition Policy Objectives
203
W.T.O. dispute settlement mechanism, including consultations, a binding ruling by an independent panel, and the possibility of appealing the decision to the W.T.O. appellate body.32 To date, there has not been a test case before a G.A.T.T.AV.T.O. panel on discriminatory enforcement or non-enforcement of antitrust law against foreign goods. Thus, the scope of violation or non-violation complaints regarding antitrust enforcement practice remains unclear. As previously indicated, disputes on antitrust and restrictive business practices are generally not within the scope of the G.A.T.T. Historically, antitrust issues are excluded from the G.A.T.T. discipline, for the G.A.T.T. came into force only after provisions had been deleted regarding antitrust issues.33 Whether the non-discrimination rule applies to national antitrust enforcement must, therefore, be clarified. We suggest reaching explicit international agreement on the inclusion within the W.T.O. of a clear and enforceable rule that antitrust enforcement or nonenforcement must not discriminate against foreign goods, and on extending the G.A.T.T.AV.T.O. dispute settlement mechanism to such issues. This would establish the possibility of using W.T.O. dispute settlement in cases of discriminatory non-enforcement of antitrust laws against foreign products, thus ensuring non-discrimination. Although an improved W.T.O. dispute settlement regarding non-discrimination in national antitrust enforcement may appear a desirable solution, it cannot replace prior application of domestic antitrust law. In most cases, discrimination is not likely to be obvious without initiation of an investigation in the importing country. Consequently, discrimination or non-discrimination will be difficult to detect. This reinforces the above recommendation that complaint, as established by the 1990 G.A.T.T. panel report on oilseeds: (1) there must be a negotiated tariff concession, (2) a governmental measure, otherwise G.A.T.T. consistent, must alter the competitive relationship between domestic and imported goods established through the tariff binding, and (3) the measure could not have been reasonably anticipated at the time the tariff concessions were negotiated. {Oilseeds Panel Report, B.I.S.D. 37/86-132); see also Petersmann 1995, supra note 28, at 26; Hoekman and Mavroidis, supra note 14, at 23. Discriminatory non-enforcement of antitrust law meets these three requirements only if it became discriminatory after a country had committed itself to certain market opening concessions. 32 For more detail on antitrust issues under the G.A.T.T., see Hoekman and Mavroidis, supra note 14; Petersmann 1994a, supra note 8, at 8-9. 33 The 1948 Havana Charter, which was to establish the International Trade Organization, was not approved by the U.S. Congress, because it contained rules on antitrust policy. Subsequently, its agenda was reduced to that of the G.A.T.T., which excluded antitrust. Havana Charter for an International Trade Organization, U.N. Dec. E/Conf 2/78 [1948].
204
Working Paper V
access to antitrust enforcement in the importing country should be improved by granting legal standing to private parties to request the initiation of a case before administrative authorities and courts, and by ensuring political independence of administrative agencies and courts.34 This measure fosters accountability and transparency in the application of antitrust law. 4. Discriminatory Antitrust Enforcement Against Foreigners Enforcement of antitrust law in the importing country may discriminate against foreign goods, tolerating restrictive business practices that are aimed at foreign competitors while prohibiting such behaviour among domestic rivals. Exporters injured by discriminatory enforcement of antitrust law may seek to appeal to the W.T.O. for violation of national treatment as set out in Art. Ill of the G.A.T.T. Discriminatory enforcement of antitrust law against foreign goods in a documented case is likely to provide a sufficient basis for a violation complaint with the W.T.O. As with discriminatory non-enforcement of antitrust law, there might be doubts whether the non-discrimination provision of the G.A.T.T. could be applied to discriminatory antitrust enforcement. We therefore suggest that the international community make it clear that antitrust enforcement by national authorities comes under the G.A.T.T. AV.T.O. requirement of non-discrimination. In addition, as discussed above, improvements with respect to legal standing, private action suits, and political independence of administrative authorities will help prevent discriminatory enforcement of antitrust law. 5. Lax Foreign Antitrust Standards or Enforcement Dissatisfaction may arise because antitrust enforcement in the importing country is lax, in a non-discriminatory way, or because antitrust standards are less 34
In comparison, the Max Planck Draft International Antitrust Code suggests an 'International Procedural Initiative' to guarantee both enforcement of national law and compliance of national enforcement with international law. An independent International Antitrust Authority is vested with powers, on the one hand, to challenge national antitrust authorities before national courts, sue private parties before national courts, and appeal against antitrust decisions before national courts for violation of national antitrust law, and on the other hand, sue a country before an International Antitrust Panel for violation of the International Antitrust Code. D.I.A.C., supra note 28, Art. 19, Sec. 2 (a) through (e); Drexl et al., supra note 28; Wolfgang Fikentscher, Competition Rules for Private Agents in the G.A.T.T./W.T.O. System, 49 AuBenwirtschaft 281, 292 (1994).
4 — Competition Policy Objectives
205
strict than in the home country of the exporter. A different antitrust approach may, in fact, restrict market access of foreign exporters if the importing country tolerates practices that would be held anticompetitive in the exporting country. Filing an antitrust case will not bring relief. There is no antitrust remedy available for exporters in the importing country. Generally, the sovereignty of nations suggests that every country should have a deliberate choice of internal regulation of commerce, including antitrust law and the rigour of its enforcement. If this situation is not satisfactory for the exporting country, it may be fruitful to discuss whether to harmonise or agree on minimum standards,35 or mutually to recognise differences of national antitrust laws. The five situations discussed above point to circumstances where enforcement of national antitrust law in the importing country is not possible or does not bring sufficient relief to exporters affected by restrictive business practices by import-competingfirms/industries.Allfivecases illustrate the need for governments to introduce or improve antitrust instruments, and to guarantee nondiscriminatory treatment in order to open the way for initiating antitrust enforcement procedures in the importing country. The effectiveness of antitrust remedies in the importing country should be improved before trade policy interventions are made in specific cases. Box 3.1: The Auto Case—The Japanese Supply and Distribution System for Cars and Car Parts In 1995, U.S. car and car parts manufacturers filed a '301 complaint' to get the U.S. government involved in their struggle to pry open the Japanese auto market, which was allegedly closed due to restrictive business practices of Japanese companies. Section 301 of the Trade Act of 1974 allows the U.S. government to impose retaliatory measures against imports when a foreign country violates U.S. rights or pursues 'unfair' or 'discriminatory' policies against U.S. exports. After negotiations under the threat of punitive tariffs of 100 per cent on selected Japanese cars, the U.S. and Japan reached agreement on improved access to the Japanese distribution system for cars and on increases in Japanese purchases of U.S. car parts.
35
See Joel I. Klein, A Note of Caution With Respect to a W.T.O. Agenda on Competition Policy, Address presented to the Royal Institute of International Affairs, 18 Nov. 1996. Klein argues for a cautious, de minimis approach first.
206
Working Paper V
The problems of U.S. firms with entering the Japanese market were allegedly related to vertical integration and cooperation in Japan, especially the 'keiretsu' system. Keiretsus are networks between auto manufacturers, part makers, and repair shops including cross-ownership, long-term business relationships and exclusive dealing. According to U.S. complainants, this prevented (1) U.S. cars from being exported to Japanese car dealers, and (2) U.S. car parts from being exported to car makers and repair shops in Japan. The U.S. insisted that Japan commit itself to enforceable quantitative import expansion targets. This strategy was motivated, on the one hand, by the striking imbalance in car trade between the U.S. and Japan—more than half of the U.S. trade deficit with Japan stemmed from trade in cars and car parts—and, on the other hand, by little success with earlier attempts to open up the Japanese market and increase U.S. exports to Japan. Whether the final agreement will result in a breakthrough is open, considering the different interpretations of the agreement and that commitments may not extend much further than what has already been planned. The U.S. chose not to follow an approach referring to violation of antitrust standards by Japanese car manufacturers, and it invoked neither the Japanese Antimonopoly Law nor U.S. antitrust law. Indeed, it is not clear whether any antitrust rules have been violated. An analysis suggests that the Japanese keiretsu system and perceived anticompetitive practices in the auto sector are most likely no violation of U.S., E.U., or Japanese antitrust laws. Strong competition exists among the different keiretsus of the bigfivecar producers in Japan, although they have a combined market share of 92 per cent. Furthermore, examples of successful non-U.S. car exporters to Japan suggest that the Japanese market is not really closed. According to Jagdish Bhagwati, the Japanese car market is open, but it 'is expensive to enter and the sales prospects are unexciting.' Several European car producers achieved higher market shares than U.S. companies by establishing their own dealerships, such as Mercedes Benz and B.M.W., or joint distribution with Japanese car producers, such as Volkswagen/Toyota.
Sources: Sheryl WuDunn, A Deal on Auto Trade: The Market; but is Japan Indeed Protectionist1} The New York Times, Late Edition—Final, 30 June 1995, Sec. D Financial Desk, p. 5; David E. Sanger, A Deal on Auto Trade: The Agreement; U.S. Settles Trade Dispute, Averting Billions in Tariffs on Japanese Luxury Autos, The New York Times, Late Edition—Final, 29 June, 1995, Sec. A National Desk, p. 1; David E. Sanger, Session Extended by U. S. and Japan in Impasse on Cars, The New York Times, Late Edition—Final, 24 June 1995, Sec. 1, Financial Desk, p. 1; Jagdish Bhagwati, On the Equivalence of Tariffs and Quotas, in Trade, Growth and the Balance of Payments (ed.) (1995).
4 - Competition Policy Objectives
207
Box 3.2: The Kodak Case—Fuji's Distribution System of Photographic Film and Paper in Japan
In May 1995, Kodak filed a '301 case' with the U.S. government complaining that Fuji of Japan was restricting Kodak's access to the Japanese market for photographic film and paper. The U.S. government initiated an investigation, but until now, no concrete threat of trade sanctions against Japan have been launched. The allegations by Kodak focus on Fuji's exclusive distribution agreements with retail stores in Japan, resale price maintenance, and Fuji's practices to enforce those contracts. It is claimed that since Fuji has a dominant position in Japan with a market share of 75 per cent, Fuji precludes Kodak from selling via the regular retail system in Japan. This allows Fuji to maintain high prices in Japan and to preclude rivals from the distribution channel, while dumping its products in foreign markets through crosssubsidisation. As a result, Kodak's market share in Japan could not exceed 7 per cent, although in most other foreign markets, its share is around 40 per cent. Other foreign competitors, such as Agfa-Gevaert of Germany, have similar problems, but are trying to enter the market through nonconventional distribution channels, such as supermarkets, convenience stores, or discount camera chains. Kodak's complaint emphasises that Fuji's business practices would even violate Japanese antitrust law, but they have been tolerated by the Japanese government. In fact, the J.F.T.C. found that Fuji has not violated the antimonopoly laws. Fuji has denied Kodak's allegations and has counterclaimed that Kodak used restrictive business practices in the U.S. market, such as exclusive agreements and tying, to foreclose foreign competitors, thereby preventing Fuji from attaining a market share above 10 % in the U.S. In December 1997 a interim decision by the W.T.O. rejected Kodak's claims against Fuji and the Government of Japan. Sources: Dewey Ballantine, Privatizing Protection: Japanese Market Barriers in Consumer Photographic Film and Consumer Photographic Paper (1995); Acceptance of imported cars among Japanese, Financial Times, 5 June 1995, p. 4; World Trade News: W.T.O. completes legal framework—the body providing teeth to the complaints procedures, Financial Times, 29 Nov. 1995, p. 5; Michiyo Nakamoto, Japan urged to bolster cartel rules, 25 Nov. 1995, p. 3; Guy De Jonquieres, Kodak steps up assault on Fuji, Financial Times, 6 Nov. 1995, p. 6; Michiyo Nakamoto, World Trade News: Kodak wins deal to sell cut-price film in Japan, Financial Times, 24 Aug. 1995, p. 4; Michiyo Nakamoto, World Trade News: Kodak and Fuji trade anti-competitive allegations,
Illegal Rule of reason approach; no administrative prosecution
Illegal
Illegal
No violation
No violation No violation, as long as there is competition between different keiretsu systems
Not an antitrust issue Generally, technical standards are the deliberate choice of a nation. Discrimination against foreigners is prohibited under the G.A.T.T. (Art. Ill)
Resale price maintenance
Price-fixing
Long-term supplier-customer relationships (keiretsu)
Cross-ownership, vertical integration, (keiretsu)
Governmental regulations on car inspections
No violation
No violation
Illegal
Illegal
Legal, as a means to enforce Illegal if practised by a dominant legal forms of exclusive contracts firm
Illegal to enforce resale price maintenance, monopolisation, or as a collective refusal to deal
Distributor boycotts, and other retaliatory practices Rule of reason; no administrative prosecution
Illegal, if practised by a dominant firm; Block exemptions; Block exemption for car dealership (however recently limited in scope)
Rule of reason approach: illegal, if practised by a dominant firm, restricting supply, or no alternative distribution is available; legal, if efficiency gains prevail; no administrative prosecution
Illegal, if imposed by a dominant firm —auto manufacturers are considered dominant —Fuji's market share of 75 per cent would constitute a dominant position
Exclusive distributorship and exclusive dealing
EU
us
Japan
Practices
Box 3.3: Alleged Anticompetitive Practices Under Japanese, U.S., and E. U. Antitrust Laws
4 - Competition Policy Objectives
209
Sources to Box 3.2 (continued): Financial Times, 1 Aug. 1995, p. 4; Emiko Terazono, World Trade News: Fuji to rebut Kodak charges, Financial Times, 27 Jul. 1995, p. 6; Nancy Dunne, World Trade News: Fuji challenges Kodak case on film market, Financial Times, 14 Jul. 1995, p. 4; Guy De Jonquieres, World Trade News: Kodak sparks U.S. challenge on access, Financial Times, 4 Jul. 1995, p. 4; Michiyo Nakamoto, World Trade News: Kodak claims exposes Fuji's grip on market—Sales of film in Japan, Financial Times, 1 Jun. 1995, p. 6; Nancy Dunne, U.S. steps up pressure on Japan: Kodak in trade action against Fuji as Kantor accuses carmakers, Financial Times, 19 May 1995, p. 18.
Box 3.4: The Pilkington Case—Abuse of I.P.R.S to Exclude Competitors in Foreign Markets In 1994, the U.S. D.O.J. filed an antitrust action against Pilkington of the U.K. and its U.S. subsidiary for abusing intellectual property rights (I.P.R.s) with the intent and effect to preclude US firms from entering foreign markets. Pilkington is the world's largest float glass producer and possesses patents on the production process for flat glass. Beginning in 1962, Pilkington entered into licence agreements with its competitors, which basically restricted the territories of its licensed competitors and the use of the patents, thus establishing an international cartel. The allegation by the D.O.J. included that Pilkington continued to enforce the licence agreements even after the patent rights had expired in 1982, in order to prevent its rivals from entering foreign markets and building float glass plants there. The U.S. D.O.J. initiated the action against Pilkington because Pilkington's behaviour negatively affected U.S. exports by restricting market access of U.S. exporters to foreign markets. The case ended in a consent decree with the D.O.J., in which Pilkington committed to terminate the anticompetitive practices. Source: U.S. v. Pilkington et ai, 59 Fed. Rg. f U , 339 at 41, 432 (1994). Sources to Box 3.3 (opposite): Lawrence J. White, Competition Policy in the United States: An Overview, 9 Oxford Rev. Econ. Pol. 133 (1993); Joel Davidow, Application of U.S. Antitrust Laws to Keiretsu Practices, 18 World Competition 5 (1994); Takashi B. Yamamoto, Japan, in James J. Garrett (ed.), World Antitrust Law and Practice (1995); Philip D. Bartz, U.S. Antitrust Law and Practice, Relations Between Suppliers and Customers; Pricing Products and Services, in World Antitrust Law and Practice James J. Garrett (ed.), (1995); Thomas C. Vinje and Kathleen D. Paisley, Competition Law and Practice in the European Union: Nonprice Restraints on the Distribution of Products and Services; Pricing Products and Services; Monopolization, in James J. Garrett (ed.), World Antitrust Law and Practice (1995); Mitsuo Matsushita, The Role of Competition Law and Policy in Reducing Trade Barriers in Japan, 14 World Econ. 181 (1992).
210
Working Paper V
Box 3.5: The Semiconductor Case—Anticompetitive Market Structure in Japan and Dumping in the U. S. 1985 301 petition by U.S. semiconductor manufacturers, alleging an anticompetitive market structure in the Japanese semiconductor market, including broad antitrust exemptions for joint research of semiconductor manufacturers and non-enforcement of antitrust law with regard to implicit agreement among Japanese producers and users to buy semiconductors from each other rather than from U.S. producers. Several antidumping cases by the U.S. against Japanese producers of semiconductors 1986 Settlement between the U.S. and Japan: Japanese government commits itself to encourage semiconductor users to increase purchases of U.S. semiconductors, and to discourage semiconductor producers from dumping in the U.S. U.S. imposes punitive tariffs on televisions, power hand tools and data 1987 processing machines, because U.S. is not satisfied with the implementation of the agreement in Japan Subsequent price increases of Japanese exports to the U.S. leads to partial suspension of duties New agreement between the U.S. and Japan addresses market access and dumping, establishing a goal of a 20 percent market share of U.S. semiconductors in Japan by 1992.
Sources: J. Michael Finger and K.C. Fung, Can Competition Policy Control "301", 49 AuBenwirtschaft 379, 386-9 (1994); Davidow, supra, Box 3.3 sources, at 47.
D. Anticompetitive Behaviour by Exporters in the Import Market Restrictive business practices in import markets, as opposed to global anticompetitive strategies, apply to internationally segmented markets. In an integrated global market, no market-specific strategies can theoretically be implemented, given the possibility of parallel trade and re-imports, which would offset such strategies. Abuses of dominant positions in global markets will be discussed below, in the section covering spillovers of competition policies. This Section addresses other forms of anticompetitive behaviour by exporting firms, including dumping.
4 - Competition Policy Objectives
211
1. International Cartels and Market-sharing Agreements International cartels and market sharing agreements between firms in two or more countries are generally recognised as akin to horizontal pricefixingand other collusive agreements within a single country. In both cases, competition is limited, output is restricted, and/or markets are allocated for the private benefit offirms.As in the case of domestic cartels, international cartels must be vigorously prevented, and heavily penalised. Prevention of international cartels is the one area where consensus and cooperation are likely to emerge among trading nations. Such cartels have an adverse impact on competition in industrialised, as well as developing country markets.36 Since international cartels extend to two or more jurisdictions, cooperative agreements in the administration and enforcement of competition laws by different national agencies must be facilitated. As discussed above, several bilateral and limited multilateral agreements exist relating to information sharing, mutual legal assistance, etc. Such inter-governmental agreements provide useful models for consideration within the W.T.O., and as in the case of regional trade agreements, need not be superseded by the principles which the W.T.O. may adopt. However, an obvious anomaly would exist if the treatment of international cartels is different from that of export cartels. The latter is discussed in a separate section below, and generally involves agreements between firms in the same country which, in many jurisdictions, are granted antitrust exemptions. The W.T.O. should make it a priority to take measures to abolish export cartels on the same grounds as it or national governments contest international and horizontal cartels. 2. Predatory Dumping Predatory dumping by an exporting firm in an importing market has emerged as a major concern in several countries, especially during the post-war period. Consequently, many countries have introduced antidumping statutes, which have become a major instrument of trade protection against low-priced imports in industrialised countries, and increasingly in developing countries as well.37 36
See Klein, supra note 35, for description of U . S . / C a n a d a case examples. The main users of antidumping are still the E.U., the U.S., Canada, Australia, and more recently Mexico. However, since the 1980s, many more countries have adopted or reactivated antidumping laws, such as Argentina, Bolivia, Brazil, Chile, China, Colombia, Egypt, India, Indonesia, Israel, Jamaica, Malaysia, Morocco, Peru, the Philippines, South Africa, Thailand, Trinidad and Tobago, and Venezuela. See Antidumping (J. Michael Finer ed. 1993). 37
212
Working Paper V
Predatory dumping, or its equivalent predatory pricing, is an antitrust policy concern, for it may lead to monopolisation. An exporter may attempt to drive its competitors in the importing country from the market in order to gain monopoly power and thereafter increase prices.38 In domestic trade, cases of predatory pricing are few, whereas in international trade, antidumping cases are frequent. No case, however, has ever been reported where a predatory pricing strategy has resulted in monopoly. Currently, antidumping measures eliminate the possibility that low-priced imports will harm competition in the importing country. Whenever antidumping agencies detect that imports are sold below their home market price or below cost, and domestic import-competing industries lose profits or market shares, the importing country may impose antidumping duties on the dumped imports to prevent further injury. However, antidumping is not restricted to predatory pricing. Antitrust concern about monopolisation is only one objective of antidumping policy. Other goals include creating a 'level playingfield'between domestic and foreign firms, protecting 'strategic' domestic industries, and easing adjustment costs in industries injured by increasing imports. As a result, the substantive standards under antidumping law vary from those under antitrust law to prevent predatory pricing. In fact, while antitrust law is limited to protecting the competitive process from monopolisation, antidumping law protects specific domestic competitors from injury.39 This difference between antidumping and antitrust policy became very obvious in the Matsushita case in the U.S.,40 where parallel antidumping and antitrust proceedings came to different conclusions. While antidumping agencies called for protective measures, antitrust authorities could not detect any danger to competition (see below, Box 4.1). 38
T h e O . E . C . D . a n d U . N . C . T . A . D . have recognised that predatory pricing m a y constitute a d a n g e r for international competition by putting predatory behaviour on non-binding lists of restrictive business practices from which firms shall refrain. T h e O . E . C . D . Guidelines for Multinational Enterprises of 1976 provide that firms shall not abuse their d o m i n a n t m a r k e t position through, for example, ' p r e d a t o r y behaviour t o w a r d s c o m p e t i t o r s ' . T h e U . N . C . T . A . D . Restrictive Business Practices C o d e of 1980 states t h a t firms shall n o t engage in 'below cost-pricing to eliminate competitors'. 39 T h e r e is an a b u n d a n t literature on the differences between a n t i d u m p i n g and antitrust. E.g., R i c h a r d Dale, Anti-dumping Law in a Liberal T r a d e Order (1980); Patrick A. Messerlin, The E.C. Antidumping Regulations: A First Economic Appraisal, 125 Weltwirtschaftliches Archiv 563 (1989); Ernst-Ulrich Petersmann, Need for Reforming Antidumping Rules for the G.A.T.T.- W. T. O. World Trade and Legal System, 27 J. W o r l d T r a d e 35 (1990); Davidow, supra, sources of Box 3.3, at 4 1 - 7 ; Finger (ed.), supra n o t e 20. 40 M a t s u s h i t a Electric Industrial C o . v. Zenith R a d i o C o r p . , 475 U.S. 574 (1986).
4 - Competition Policy Objectives
213
Antidumping measures have several negative consequences: (1) they produce losses of consumer welfare due to unnecessary protection41; (2) they place domestic downstream user industries at a competitive disadvantage vis-a-vis foreign competitors that are allowed to buy inputs at a pre-duty price; (3) they discriminate against foreign exporters, who must respect higher standards under antidumping law than domestic competitors do under antitrust laws; (4) by excluding foreign exporters from the market and artificially increasing import prices, they reduce competition in the import market. Given these negative effects, antidumping measures should be repealed and replaced with antitrust standards. We recommend replacement of antidumping laws with enforcement of national antitrust laws against dumped imports. Antitrust laws usually apply to any anticompetitive conduct that has a negative impact on competition in the domestic market, irrespective of whether afirmis located within that nation or abroad.42 Difficulties in establishing personal jurisdiction may arise if a dumping exporter has no subsidiaries in the importing country. Thus, we recommend that an international agreement should be entered which (1) explicitly allows application of national antitrust standards to foreign companies, according to the effects doctrine,43 if dumping is based on price discrimination,44 and produces negative effects on competition, (2) provides for cooperation between national antitrust agencies regarding the collection of evidence abroad and 41
F o r the U.S., the I.T.C. calculated a welfare loss of $1.59 billion dollars d u e to existing antidumping a n d countervailing duty orders in 1991. 42 In the E.U. a n d A.N.Z.C.E.R.T.A., t w o regional economic integration agreements, antidumping has been superseded by antitrust standards. T h e E . U . replaced antidumping measures between its members with a c o m m o n supranational antitrust policy, including c o m m o n harmonised rules a n d a central enforcement agency, while the A.N.Z.C.E.R.T. A. allows enforcement of national antitrust laws according t o the effects doctrine as a substitute for antidumping, but only after the approximation of antitrust laws. The C a n a d a - U n i t e d States Free T r a d e Agreement (C.U.S.F.T.A.) recognised the link between antidumping a n d antitrust, a n d recommended the establishment of a special task force to study the possibility of replacing antidumping with antitrust—which was, however, not accomplished. C a n a d a - U . S . Free T r a d e Agreement, Statutes of C a n a d a , Ch. 65 (2 Jan. 1988). 43 Recognition of the effects doctrine is one of the basic principles of the M a x Planck Draft International Antitrust Code. D . I . A . C , supra note 28, Art. 3, Sec. 2; Drexl et al., supra note 28. It has also been recognised in the 1990 A . N . Z . C . E . R . T . A . 44 In more general terms, this includes that markets are segmented a n d the dumping strategy is market-specific. If the market is global a n d the anticompetitive behaviour is not market-specific in the importing country, antitrust control by one country is likely to have spillover effects in other countries. This situation will be discussed in the section on spillover effects of antitrust enforcement.
214
Working Paper V
(3) provides for cooperation of enforcement action (discussed in further detail in the following section). Enforcement of antitrust law in the importing country against dumped imports is not likely to produce any adverse effects on competition in the exporting country if markets are segmented. Therefore, control of dumping in international trade does not necessarily require harmonisation of national predatory pricing rules. Existing national antitrust laws applied to foreign dumpers are sufficient to prevent monopolisation in the import market, with the above -mentioned procedural amendments regarding cooperation on the collection of evidence and enforcement action abroad. In addition, segmentation of national markets should be reduced in order to limit the occurrence of predatory pricing strategies in international trade. Market segmentation has two effects on dumping strategies. First, a segmented home market of the exporters allows them to make supracompetitive profits at home and to cross-subsidise below cost sales in foreign markets. International market segmentation distinguishes dumping in international trade from domestic dumping, as it allows cross-subsidisation between markets. This is not possible in an integrated market, where all competitors must compete under the same competitive conditions. Barriers to entry in export markets thus increase the possibility of predatory pricing strategies in international trade. Secondly, a segmented import market increases the profitability of predatory pricing strategies. Where barriers to entry in the import market prevent newcomers from entering the market, predatory pricing may be a profitable strategy to drive existing competitors from the market and attain a monopoly.45 Although a segmented export market may give exporters a competitive advantage over rivals in the import market, exporters will engage in predatory pricing only if it is profitable, i.e., if they have a chance to raise their prices later. The focus of antitrust investigation into low-price strategies, there45
Most national antitrust policies toward predatory pricing therefore include examination of either barriers to entry or the profitability of predatory pricing. An O.E.C.D. study on predatory pricing suggests an antitrust approach that, in a first stage, examines market structure and entry conditions to determine whether the market could possibly be monopolised by predatory pricing. If actual and potential competition are strong, the investigation should end. '[C]ompetition authorities would look first to the market in question and determine whether it is susceptible to successful predation . . . If it appeared unlikely due to market structure and entry conditions that the alleged predator would be able to exercise market power in the post-predation period, the inquiry should end, as there would be no harm to competition even if some competitors suffered during the price-cutting episode'. O.E.C.D., Predatory Pricing (1989) 82. Such an antitrust approach has been adopted toward domestic dumping by Canada.
4 - Competition Policy Objectives
215
fore, must be on the importing market, regardless of what competitive conditions prevail in the exporters' home market.46 As a result, an international strategy to prevent predatory pricing strategies should replace antidumping with antitrust standards. It should also reduce segmentation of national markets, in order to lessen both the possibility of predatory dumping through cross-subsidisation between markets, and the profitability of a predatory pricing strategy due to the lack of potential competition in the import market.47 The reduction of market segmentation involves: internal deregulation in every country to ease market access; international negotiations on the reduction of governmental barriers to trade; and effective antitrust enforcement at the national level, supplemented with international cooperation (as discussed above) to prevent private agents—including the dumping firms themselves—from restricting market access to their home market through anticompetitive behaviour. Meanwhile, countries negatively affected by dumping practices should apply national antitrust laws against predatory pricing by foreign firms. We suggest also that a firm injured by dumped imports should be able to file an antitrust case in the home market of the dumpers, if restrictive business practices abroad give rise to dumping. This would help create a 'level playing field', thus erasing unfair advantages due to the lack of competition in the home market of dumping firms. We recognise that some problems may arise with this approach, which have been discussed (along with possible remedies) in the previous section on restriction of market access by import-competing firms.48 46
This was correctly assessed by the U . S . Supreme Court in the 1986 Matsushita case, where the Court did not consider the U.S. defendants' arguments that the alleged Japanese dumpers could m a k e supracompetitive profits at home while d u m p i n g in the U S , because even these high profits could not as such furnish a motive for predatory pricing: Matsushita Electric Industrial C o . v. Zenith Radio Corp., supra note 40, 475 U.S. at 593; Daniel J. Gifford, Predatory Pricing Analysis in the Supreme Court, 39 Antitrust Bull. 431, 460-464 (1994). 47 The O.E.C.D. study on predatory pricing concludes similarly that 'efforts to improve the conditions for entry a n d expansion in a given market, including the removal of barriers to international competition, should help combat the threat of effective predation': O.E.C.D., supra note 45, at 82. 48 W e d o n o t suggest that the c o u n t r y affected by d u m p e d imports either enforce its own antitrust standards extraterritorially against anticompetitive practices in the home country of the dumping firms, or use trade sanctions to pry open the foreign market. Harris First, An Antitrust Remedy for International Price Predation: Lessons from Zenith v. Matsushita, 4 Pac. Rim L. & Pol. J. 211,241 (1995) suggested extraterritorial enforcement of U.S. antitrust standards in cases where dumping is possible because of R.P.B.s in the home market: '[t]aking the view that the existence of a closed home market
216
Working Paper V
Box 4.1: The Matsushita Case—Dumping of Japanese Televisions in the U.S. Although television was invented in the U.S., American manufacturers subsequently faced strong competition from Japanese producers. This increasingly threatening situation led U.S. firms to seek relief from low-priced imports under antitrust laws, as well as under trade protection laws. The series of cases brought against Japanese manufacturers started with an antidumping complaint (1968), followed by two private actions for damages under antitrust law (1971 & 1974), a complaint by workers in the U.S. T.V. industry (1971), a '337' complaint against unfair trade practices (1976), and finally a '201' petition for escape clause protection (1976). The allegations focused on dumping of Japanese televisions in the U.S., and conspiracy among Japanese producers, which kept prices high in Japan while U.S. prices were below cost, in an attempt to injure U.S. competitors and finally drive them from the market. As a result, the U.S. decided to impose antidumping duties on Japanese T.V.s in 1971 (which, however, were never collected) and entered an Orderly Market Agreement with Japanese producers in 1977, which limited exports to the U.S. The antitrust case was finally decided against U.S. complainants by the Supreme Court in 1986. In the entire period from 1963, when exports to the U.S. began, through the early 1990s, the structure of the television industry changed fundamentally. Today, many former U.S.firmsare controlled by foreign companies, and foreign manufacturers shifted assembly to the U.S., with imports still remaining around 50 per cent. However, the major exporting country is no longer Japan, but Mexico, Taiwan, Singapore, Malaysia, Korea and others. Industry concentration has not increased; Japanese producers could not monopolise the market. Comparing the antitrust case with the trade cases, the most surprising outcome is that violation was found under trade statutes but not under antitrust law, which points to substantial differences between the two approaches. Obviously, protection under trade laws goes much further that protecting competition and market access, which is the explicit goal of antitrust policy. The Supreme Court, as well as the Department of Justice,
explains the willingness to sell at predatory prices and insures against retaliation, a remedy might be to order the foreign defendants to end their exclusionary efforts in their domestic markets.' A specific information provision is included in all existing bilateral cooperation agreements, whereas cooperation in enforcement action is provided for only in the E.C./U.S. agreement and the A.N.Z.C.E.R.T.A. Petersmann 1994a, supra note 8; Hachigan, supra note 49, at 132-41.
4 - Competition Policy Objectives
217
were convinced this was not a case of unfair competition or monopolisation, but a case of a competitive advantage due to inventiveness and technical expertise of Japanese manufacturers, which benefits the American consumer. Sources: Matsushita Electric Industrial Co. v. Zenith Radio Corp., supra note 40, 475 U.S. 574; First, supra note 48; Kenneth G. Elzinga, 'Collusive Predation', in John E. Kwoka and Lawrence J. White (eds.), The Antitrust Revolution: The Role of Economics (1994); Gifford, supra note 46.
1. Collection of Evidence and Enforcement Action Abroad An importing country might face difficulties when restrictive business practices are conducted by a foreign exporter that does not have a subsidiary in the importing country. Problems may arise as to (1) the collection of information about the alleged anticompetitive behaviour by afirmlocated abroad,49 and (2) the enforcement of a decision against a foreign firm. Currently, some jurisdictions have concluded agreements on cooperation between national antitrust authorities, such as the 1991 agreement between the U.S. and the E.C. Most of those cooperative agreements in the area of antitrust contain some provisions on the exchange of information between national antitrust authorities.50 The E.C./U.S. agreement, for instance, provides for the 49
See, e.g., Nina Hachigian, Essential Mutual Assistance in International Antitrust Enforcement, 29 Int'l L. 117 (1995). 50 T h e following bilateral antitrust cooperation agreements have been concluded so far: Germany/U.S., 1976 (Agreement Between the United States of America a n d the G o v e r n m e n t of the Federal Republic of G e r m a n y Relating t o Mutual Cooperation Regarding Restrictive Business Practices, T.I.A.S. 8291, 27 U.S.T. 1956 (entered into force 11 Sept. 1976); Australia/U.S., 1982 (Agreement between the Government of the United States of America a n d the Government of Australia Relating to Cooperation o n Antitrust Matters, 29 June 1981, reproduced in 21 Int'l Legal Materials 702 (1982); Canada7U.S., 1984 ( M e m o r a n d u m of Understanding as to Notification, Consultation and Cooperation with Respect to the Application of National Antitrust Laws, 23 Int'l Legal Materials 275 (1984); France/Germany, 1984 (Agreement between the G o v e r n m e n t of the Federal Republic of G e r m a n y and the Government of the French Republic concerning Cooperation on Restrictive Business Practices, Paris, 28 M a y 1984, 26 I.L.M. 531 (1987)); E.C./U.S., 1991 (Agreement Between the Government of the United States a n d the Commission of the European Communities Regarding the Application of their Competition Laws, [1995] OJ L47 (27 April); A.N.Z.C.E.R.T.A., 1990. Australia/New Zealand (Cooperation a n d Coordination agreement between the Australian T r a d e Practices Commission a n d the N e w Zealand Commerce Commission, 26 July 1994).
218
Working Paper V
exchange of information between antitrust authorities in three respects: first, on current antitrust enforcement policy; second, on anticompetitive conduct that is 'relevant to, or may warrant, enforcement activity by the other Party's competition authorities'; thirdly, the antitrust authorities of one jurisdiction may request information in the possession of their counterpart agency, which may be necessary for the requesting party's antitrust enforcement activities. The agreement specifically excludes the exchange of confidential information, which is protected under national confidentiality laws, unless the company involved agrees to the exchange of such information.51 The provisions of the E.C./U.S. antitrust cooperation agreement have limited scope. They do not give antitrust authorities any investigative powers to collect evidence abroad, where they need to examine anticompetitive conduct of foreign firms. Also, the right to request foreign antitrust authorities for information is limited to data that are already within their possession, and which are not confidential. The agreement does not provide for the possibility of assistance in the collection of evidence. Generally, even upon specific request, antitrust authorities do not gather domestic information needed by foreign antitrust authorities. The U.S. and Canada entered a Mutual Legal Assistance Treaty (M.L.A.T.) in 1990, which contains a stronger information exchange provision. It provides for the exchange and collection of data by the other party's authorities upon request, including the provision of documents and records, the execution of searches and the taking of oral testimony.52 However, this agreement is limited to criminal proceedings53; it does not apply to civil antitrust cases,54 which 51
Neither party is required to provide information 'if disclosure of that information . . . is prohibited by the law of the Party possessing the information, or would be incompatible with important interests of the Party possessing the information': E.C./U.S. cooperation agreement, supra note 50, Art. VIII. T h e exchange of confidential information was allowed in the 1994 Microsoft case upon agreement of Microsoft. 52 T h e first case under the U.S./Canada M.L.A.T. was a joint criminal antitrust prosecution of U.S. and Canadian authorities against price fixing in N o r t h America with regard to thermal fax paper by Kanzaki of the U.S. and Mitsubishi from Japan, which resulted in fines against both companies. U.S. and Canadian prosecutors attack cartel behaviour by fax paper distributors, 67 Antitrust & T r a d e Reg. Rep. 108-109 (1994). 53 In the U.S., criminal violations of the Sherman Act include price fixing a n d conspiracy. In C a n a d a , criminal violations of the Competition Act include conspiracy a n d bid rigging. 54 See, e.g., Calvin S. Goldman and Joel T. Kissack, Cooperative Antitrust Enforcement Efforts Between C a n a d a a n d the United States: Investigations, Information Sharing a n d Confidentiality in Criminal Proceedings, Discussion Paper prepared for the A n n u a l Meeting of the American Bar Association Section of Antitrust Law, 6 August 1995; Diane P. W o o d , The Internationalisation of Antitrust Law:
4 - Competition Policy Objectives
219
is the approach generally taken with respect to predatory pricing and merger control. Regarding enforcement of decisions in another jurisdiction, only the E.C./U.S. agreement and the A.N.Z.C.E.R.T.A. provide for cooperation. The E.C./U.S. agreement contains a provision on interagency assistance in enforcement activities. U.S. authorities shall assist E.U. authorities in the enforcement of a decision against a company in the territory of the former, and vice versa. However, the agreement does not limit the sovereignty of any party. Therefore, activities to enforce antitrust decisions of one party in the territory of the other party are restricted to cases where both parties have a common interest in doing so.55 In contrast, the A.N.Z.C.E.R.T.A. provides for mutual recognition and enforcement of court decisions.56 Options for the Future, Address before the DePaul Law Review Symposium, 3 Feb. 1995 (U.S. D.O.J., Washington, D.C.), 6-8; Georges Addy, Competition Act Amendments (Canadian Competition Bureau 1995) (hereinafter 'Addy 1995a'), 11-14; Georges Addy, International Harmonization and Enforcement Cooperation: The Canadian Experience, in
Wang Chih-Kang et al. (eds.), International Harmonization of Competition Laws (1995) (hereinafter 'Addy 1995b'), 400-4; Hachigan, supra note 49. The 1994 NA.F.T.A. encourages member states to cooperate on antitrust enforcement including mutual legal assistance, notification, consultation and exchange of information relating to enforcement. The North American Free Trade Agreement between Canada, the United States and Mexico (N.A.F.T.A.), Ch. 15 (Competition Policy, Monopolies, and State Enterprises) (17 Dec, 1992). In 1994, the U.S. adopted the International Antitrust Enforcement Assistance Act (I.A.E.A.A.), which aims to facilitate cooperation between U.S. antitrust enforcement agencies and their foreign counterparts. The I.A.E.A.A. authorizes the Department of Justice and the Federal Trade Commission to conclude Mutual Assistance Agreements (M.A.A.s) with foreign antitrust authorities, and to disclose information to those foreign agencies which have entered into M.A.A.s with the U.S. This includes the collection of information by U.S. agencies for foreign antitrust authorities, as well as the disclosure of information already in the possession of U.S. authorities to foreign antitrust agencies. Canada has a similar law, the Mutual Legal Assistance in Criminal Matters Act, which on the one hand implements the M.L.A.T. with the U.S., and, on the other hand, aims to conclude M.L.A.T.s with other foreign governments, although restricted to criminal matters. 55
Art. IV of the E.C./U.S. cooperation agreement, supra note 50, states: '(1) T h e competition authorities of each Party will render assistance t o the competition authorities of the other Party in their enforcement activities, t o the extent compatible with the assisting Party's laws a n d important interests, a n d within its reasonably available resources. (2) In cases where both Parties have a n interest in pursuing enforcement activities with regard t o related situations, they m a y agree that it is in their mutual interest t o coordinate their enforcement activities . . . ' 56 Petersmann 1994a, supra note 8, at 17.
220
Working Paper V
We recommend the conclusion of a multilateral agreement within the W.T.O. which provides for cooperation among jurisdictions as to the collection of evidence and enforcement action. This includes the following services of antitrust agencies in one country upon request by authorities in another country in civil and criminal antitrust proceedings: disclosure of information in the possession of antitrust authorities, even confidential—complemented with adequate safeguards57; collection of data through foreign authorities and their compulsory investigative powers58; and mutual recognition and enforcement of antitrust decisions rendered by administrative agencies or a court in one country through authorities and/or courts in another country. Such far-reaching obligations to provide information to foreign authorities and to enforce decisions rendered in a foreign country may entail a loss of sovereignty, which can be counterbalanced only through both reciprocal rights and effective rules on negative and positive comity.
2. Multinationals in Developing Countries Many developing countries are concerned about restrictive business practices by powerful multinational enterprises (M.N.E.s) and the abuse of their market power at the expense of developing countries. For instance, such monopolistic and trade-restricting practices of M.N.E.s may consist of requirements for subsidiaries in developing countries to buy imports from the parent company instead of local suppliers, the prohibition of subsidiaries from exporting products, or abusive licensing of technology. Traditionally, developing countries have been using regulations of foreign direct investment to control anticompetitive business behaviour of multi57
T h e V a n M i e r t R e p o r t emphasises the need to eliminate 'current obstacles relating to confidentiality rules applicable to exchanges of information' in the bilateral E.C./U.S. cooperation agreement: V a n Miert R e p o r t , supra note 27, at 14. Similarly, Georges Addy, former D i r e c t o r of the C a n a d i a n Bureau of Competition Policy, recognised 'the need t o c o m m u n i c a t e confidential information selectively... to foreign competition law agencies' a n d t h a t this requires 'a well-established framework for the treatment of confidential information a n d with adequate safeguards for the protection of those whose interests a r e affected': A d d y 1995a, supra note 54, at 11-14. 58 Georges A d d y m a d e a similar suggestion for bilateral cooperation agreements: '[m]utual assistance could include the following: authorizing the Director to use the compulsory p o w e r s available u n d e r the Act to obtain information for the enforcement
of another country's competition laws; and authorizing the Director to provide a foreign authority, upon request or at his own initiative, with information in his possession that may be relevant to the enforcement of the Competition Act or foreign competition law': id.
4 — Competition Policy Objectives
221
nationals, including regulation of imports, exports, manufacturing, foreign exchange, repatriation of returns, local equity and the transfer of technology. Because of trade distortions caused by excessive investment requirements, the Uruguay Round agreement on Trade-Related Investment Measures (T.R.I.M.s) has declared some measures relating to imports and exports to be incompatible with the G. A.T.T. rules of non-discrimination and the prohibition of quantitative restrictions. However, many restrictions of trade and investment in developing countries will still be allowed under the T.R.I.M.s agreement.59 In parallel, developing countries have supported attempts to establish disciplines for restrictive business practices by powerful multinational enterprises on the international level. As a result, the international community, as gathered within the U.N.C.T.A.D., adopted 'The Set of Multilaterally Agreed Principles and Rules for the Control of Restrictive Business Practices' 60 in 1980, and 'The Code of Conduct on Transfer of Technology' in 1981, which, however, are not legally binding. The U.N.C.T.A.D. R.B.P. code emphasises the danger that anticompetitive conduct in international trade may restrict market access and competition, entailing negative effects on trade, especially with respect to developing countries.61 The code establishes detailed substantive antitrust rules for the conduct of private enterprises, including cartel agreements, the abuse of monopoly positions and mergers; and recommends cooperation between governments with regard to antitrust enforcement, exchange of information and multilateral consultations.62 Furthermore, some 59
A list of trade related investment measures, along with a discussion of the results of the U r u g u a y R o u n d T.R.I.M.s agreement, is provided by L o w a n d Subramanian, supra note 11, at 1-6. T h e following are inconsistent with the T.R.I.M.s agreement: local content rules, trade balancing requirements, foreign exchange balancing requirements, a n d domestic sales requirements. T h e following regulations will be not inconsistent with the agreement: manufacturing requirements or limitations, export performance requirements, product mandating requirements, exchange restrictions, technology transfer requirements, licensing requirements, remittance restrictions, a n d local equity requirements. 60 U . N . C . T . A . D . document T.D./R.P.B./CONF/10 (1980). 61 U . N . C . T . A . D . , C o d e of Conduct on Transfer of Technology, Int'l Investment Instruments: A C o m p e n d i u m , 1 Multilateral Instruments 181 ( U . N . 1996). T h e code condemns 'acts or behaviour of enterprises which, through an abuse or acquisition a n d abuse of a d o m i n a n t position of market power, limit access to markets or otherwise unduly restrain competition, having or being likely to have adverse effects on international trade, particularly that of developing countries, a n d on the economic development of these countries, or which through formal, informal, written or unwritten agreements or arrangements a m o n g enterprises, have the same impact'. U . N . C . T . A . D . (1980), supra note 60. 62 Petersmann 1994a, supra note 8, at 7-8, 17-18, 24.
222
Working Paper V
developing countries have unsuccessfully attempted to get the issue of restrictive business practices on the agenda of the Uruguay Round international trade negotiations.63 International antitrust rules could remedy market access restrictions caused by anticompetitive behaviour of dominant foreign firms in developing countries. Moreover, international antitrust rules that supersede investment regulations in developing countries could overcome trade distortions caused by excessive investment controls. However, substantive international antitrust rules are not necessary to control anticompetitive business practices by foreign investors. Instead, adoption and enforcement of antitrust laws in the countries where the restrictions of competition occur will likely be sufficient. Thus, we recommend that all countries adopt antitrust laws at the national level.64 Any country is free to adopt national antitrust laws unilaterally and should be encouraged to do so. Effective antitrust enforcement should substitute for excessive controls of foreign direct investment and the transfer of technology by multinational enterprises, as well as trade and production of their subsidiaries. As a complement, effective enforcement of antitrust law may, in some cases, require cooperation between national antitrust authorities regarding the collection of evidence and enforcement abroad, as discussed in the previous section. We therefore suggest a complementary international agreement facilitating interagency antitrust cooperation and consultations.
E. Spillovers of Regulating Competition in One Country on the Competitive Process of Another Country Anticompetitive behaviour and antitrust regulation of one juridiction may have spillover effects on competition in another jurisdiction. Such may be the case when there is a merger between two worldwide dominant firms in a global market. Although both firms might be located in the same jurisdiction, the effects of possible monopolisation are not limited to this jurisdiction, but may occur in other jurisdictions where those firms conduct business. Likewise, the regulation of such a merger under antitrust law of the home country, including its approval (with or without conditions) or rejection, will generate effects on 63
Blackhurst, supra note 9, at 227; Petersmann 1994a, supra note 8, at 9. Similarly the E . U . G r o u p o f Experts stated: '[o]ne should commence with the introduction of an a d e q u a t e set o f competition rules by those countries that d o not yet have o n e ' : V a n Miert Report, supra note 27, at 13. Adoption a n d enforcement of national antitrust laws in member countries was also a requirement in the draft I.T.O., as well as in the N . A . F . T . A . 64
4 - Competition Policy Objectives
223
competition in foreign markets. Such issues were raised in cases such as the merger between DeHavilland and Aerospatiale-Alenia. Spillovers of restrictive business practices and of antitrust enforcement may also occur in other areas, such as various forms of anticompetitive behaviour by globally dominant firms that create or strengthen a globally dominant position. Potential concerns in this regard were raised by the practices of Microsoft, including alleged abusive licensing practices. This section does not discuss strategically intended spillover effects aimed at giving domesticfirmsa competitive advantage over foreign firms, which is covered in the section on strategic antitrust policy. Rather, it will focus on situations where a market is global and a firm is globally dominant, which may sometimes create spillover effects of anticompetitive behaviour, and its antitrust regulation from one country to another. 1. Merger Regulation in Global Markets Mergers may produce spillover effects in foreign countries wherever international trade occurs. The more global a market is, the larger will be the spillover effects between countries. No external effects will result where national markets are closed or no trade occurs. However, a merger in a closed economy today may also have future effects on a foreign market once trade between countries becomes liberalised. Spillover effects of a merger on other countries may occur regardless of whether the merging firms are located in different or in the same jurisdiction. Table 1 summarises the circumstances under which mergers and merger regulation produce external effects, and the nature of such external effects. Facing possible multiple effects of mergers, many countries may claim jurisdiction over a merger. This, in fact, often occurs.65 The one or two countries where the merging firms are located are likely to control the antitrust implication of a merger. Furthermore, some of the effects are not restricted to the home countries of the firms involved, but spread to the territory of other jurisdictions,66 which may initiate a merger investigation under their own antitrust laws. 65
See, e.g., the Whish/Wood study, supra note 23, which examined 9 international merger cases involving investigations by more than 2 national national antitrust a u t h o r ities. T h e merger between Gillette and Wilkinson was examined by the authorities o f 14 countries. 66 T h e U . S . , for instance, claims subject matter jurisdiction under Sec. 7 of the Clayton Act over any merger that h a s effects on U . S . imports. U . S . D . O . J . / F . T . C , Antitrust Enforcement Guidelines for International Operations, (1995) 119-20, especially 'Illustrative Example H \
Working Paper V
224
Table 1: Effects of Mergers on International Trade and Their Antitrust Regulation Closed Markets
Merging firms are located in the same jurisdiction
Mergingfirmsare located in different jurisdictions
Segmented Markets Global Market
Effects: Domestic effects + effects in foreign countries Possible regulation by: home country of the merging firms foreign countries affected by the merger Problems: lack of consideration of anticompetitive effects on foreign markets diverging evaluation of the competitive effects of a merger procedural differences collection of data and enforcement action abroad Cases: Gilettel Wilkinson Effects: Effects: No effects on competition Effects in both home markets + effects on third countries in either home or third Possible regulation by: countries; effects might spread to other countries home country of one firm home country of the other firm after trade liberalisation third countries affected by the Regulation: merger No regulation required Problems: Similar to above case Cases: DeHavillandIA lenia
Effects: Effects are only in the home country; future effects might occur abroad after trade liberalisation Regulation: by the home country
A variety of problems may arise when antitrust authorities of more than one country attempt to control a merger under national law, including problems resulting from procedural and substantive differences in national merger regulations, the lack of national antitrust authorities to consider effects in foreign
4 - Competition Policy Objectives
225
markets, and the collection of evidence and enforcement action abroad.67 Similar issues might arise with global abuses of dominant positions or monopolisation strategies, as discussed in the following subsection.
2. Different Merger Evaluation by National Antitrust Agencies Different approaches toward antitrust policy across jurisdictions may lead to divergences in the evaluation of mergers. National antitrust policies vary with respect to their basic objectives, their economic approach and the weighing of positive and negative effects of a merger. As a result, national antitrust authorities may reach different conclusions regarding a merger, even though they have the same information on the merger transaction and the market, and the effects of the merger are similar in both markets. For example, country A might prohibit a merger because it creates a monopoly, while country B does not anticipate output restrictions because it expects discipline by potential competitors. As a result, antitrust authorities in country A might permit the merger, while authorities in country B see important anticompetitive effects that cause them to prohibit the merger. Furthermore, the objectives of national antitrust policies differ with respect to the degree to which they focus on consumer welfare, total economic welfare or general public interest. Whereas a consumer welfare approach does not allow mergers that restrict output, a total economic welfare approach allows consideration of gains in productive efficiency, and a public interest approach permits consideration of a variety of factors beyond economic efficiency. Whether a merger will be approved depends on those countries that have effective control of the mergingfirms,and that have the strictest antitrust standards. A decision to prohibit a merger may be contrary to the wishes of the affected countries with no control over the merging firms or with a less restrictive antitrust approach (see Box 5.1). Insofar as differing evaluations of the long-term effects of a merger stem from different schools of thought in economics and/or different antitrust experience, one approach cannot be preferred over another. Several approaches may be reasonable, diverging only with regard to the estimated future effects of a merger, which cannot be known with certainty in advance. No one approach is therefore prima facie correct. 67
In addition, where strict merger control in a country prevents the acquisition of a domestic firm by a foreign company, this can be regarded as a barrier to foreign investment. Nevertheless, if antitrust law is applied in a non-discriminatory way, this is simply an issue of compliance of foreigners with local rules of the host country.
226
Working Paper V
Presently, no binding remedies are available where antitrust approaches vary toward merger control. Relief is possible only through cooperation between national antitrust authorities, convergence or harmonisation of national antitrust standards. However, harmonisation of substantial antitrust standards, their objectives and enforcement, is not a realistic goal. Rather, mechanisms of cooperation between national antitrust authorities to prevent conflicts Box 5.1: The DeHavilland, Institut Merieux, and Boeing-McDonnell Cases Diverging National Merger Evaluations In 1991, the two European firms Alenia of Italy and Aerospatiale of France attempted to acquire the DeHavilland Division from Boeing of Canada. Investigations were initiated in the European Community and in Canada. While the Canadian Bureau of Competition approved the merger, the European Commission prohibited it, concluding that it would restrict effective competition in the market for commuter aircraft. Although both authorities agreed that the relevant market was global, they disagreed as to the effects of the merger on global competition. The 1989 merger between Institut Merieux of France and Connaught of Canada raised antitrust concern in Canada as well as in the United States. Institut Merieux and Connaught were dominant in the U.S. and Canadian market for a rabies vaccine. Furthermore, Connaught had a monopoly in both countries' markets for a polio vaccine, for which Merieux was a potential competitor, but did not use its license to produce and sell the polio vaccine. The U.S. F.T.C. considered competition endangered in both the rabies vaccine market, where both firms combined had a monopoly, and in the polio vaccine market, because the merger would eliminate the threat of market entry by Merieux. The F.T.C, therefore, allowed the merger only after entering into a consent decree with Connaught requiring the lease of Connaught's Canadian rabies vaccine business to a third party. By contrast, the Canadian Bureau of Competition did not object to the merger, because it considered that the existence of other potential competitors from the U.S. and Europe could discipline the monopoly in the polio vaccine market created by the merger between Connaught and Institut Merieux. In 1997, Boeing Co. proposed to merge with McDonnell Douglas, which would have directly impacted competition in the global commercial aircraft market. Since antitrust officials on the two sides of the Atlantic have different legal approaches to mergers, the proposed transaction almost immediately received a negative reaction from the European Union, while it was approved by the United States. The E.U.'s stance was characterised (by The Economist) as based more on law and politics than economics. The E.U.
4 - Competition Policy Objectives
227
ultimately cleared the merger, but not without extracting commitments from Boeing, including the cessation of existing and future exclusive supply deals; 'ring fencing' McDonnell Co.'s commercial aircraft activities; and licensing patents to other jet aircraft manufacturers. The E.U. also indicated it would strictly monitor Boeing's compliance with these commitments. Sources: [1991] O.J. L-334 42; Director of Investigation and Research (Canada), 1990 & 1992 Ann. Rep.; Will economics bless this union?, The Economist, 2 Aug. 1997; European Union, press release dated 30 July 1997.
stemming from different merger evaluations should be pursued. An international agreement within the W.T.O. could include rules on general and case-bycase cooperation among governments and national antitrust agencies, positive and negative comity, the exchange of general and case-specific information, and transparency of national antitrust policies.
3. Lack of Consideration of Effects in Foreign Markets National antitrust authorities may fail to consider the effects of a merger on the competitive process of other countries. If national markets are segmented, the impact of a merger on different national markets may vary, depending on the competitive pressure of domestic and other foreign competitors or the lack thereof. Furthermore, a potential conflict might occur where efficiency gains of a merger benefit one jurisdiction, while the anticompetitive effects harm another jurisdiction. Antitrust authorities in thefirstjurisdiction might permit the merger if its law contains an efficiency defence, whereas authorities in the other country are likely to prohibit the merger for its negative effects on competition. Such situations may occur even where national antitrust laws are identical. If, for example, segmented markets exist in countries A and B, and country A allows a merger between twofirmsin its jurisdiction, Country B may be negatively affected by the merger evaluation of country A's antitrust authorities. For instance, welfare losses may occur if the new dominant firm decides to restrict output and raise prices in markets where there are no domestic competitors and no other foreign rivals due to market segmentation.68 68
Eventually, market segmentation may be overcome, but as long as markets are segmented, these effects may occur.
228
Working Paper V
Some jurisdictions, such as the E.U. and the U.S., have expressed a willingness to consider, on a unilateral basis, foreign interests in their antitrust enforcement activities.69 Moreover, some jurisdictions have concluded bilateral agreements to facilitate cooperation between national antitrust authorities, such as the E.U. and the U.S.70 These agreements typically provide for rules on positive and/or negative comity to respect foreign interests in national antitrust enforcement.71 We recommend inclusion of rules on positive and negative comity in a W.T.O. antitrust agreement, requiring that national antitrust enforcement agencies initiate an investigation upon request by a foreign country (positive comity), and take into account anticompetitive effects of domestic restrictive business practices in foreign countries when evaluating the effects on competition (negative comity), possibly in cooperation with foreign antitrust agencies. Antitrust authorities should not discriminate between domestic and foreign competitors and competition.72 Adversely affected countries should have the possibility of challenging foreign antitrust enforcement or non-enforcement before a W.T.O. panel. Dispute 69
T h e 1995 U . S . Antitrust Enforcement Guidelines for International Operations, supra note 67, at 20, provide 'In enforcing the antitrust laws, the Agencies [D.O.J. and F.T.C.] consider international comity. . . . In determining whether to assert jurisdiction to investigate o r bring a n 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'. C o u r t s applied comity previously in the Timberlane and M a n n i n g t o n Mills cases (Timberlane L u m b e r C o . v. Bank of America, 549 F2d 597 (Cir. 1976); M a n n i n g t o n Mills, Inc. v. Congoleum Corp., 595 F.2d 1287 (3rd Cir. 1979)). Likewise, E . U . antitrust agencies respect foreign interests in antitrust enforcement, as shown in the I.B.M. case, where the E.U. suspended the case because enforcement would have h a d an unacceptable impact o n competition in the U . S . C o m m ' n v. International Business Machines C o r p . (1987) 3 C . M . L . R . 147. 70 E.C./U.S. Agreement, supra note 50. 71 Positive comity means that country A m a y ask for antitrust enforcement in country B where anticompetitive practices b y firms in country B adversely affect competition in country A. Negative comity requires that country A takes into account in its antitrust enforcement effects o f anticompetitive conduct by domestic firms o n competition in c o u n t r y B. T h e following bilateral antitrust cooperation agreements contain negative comity: G e r m a n y / U . S . , supra n o t e 50; Australia/U.S., supra note 50; C a n a d a / U . S . , supra note 50, F r a n c e / G e r m a n y , supra note 50; E.C./U.S., supra note 50. Positive comity is p r o vided for in the E . C . / U . S . agreement, supra note 50, and the A . N . Z . C . E . R . T . A . , supra n o t e 50. F o r an overview, see Petersmann 1994a, supra note 8. 72 Such a d o u b l e non-discrimination rule has been suggested by Petersmann 1994c, supra note 6, at 270.
4 - Competition Policy Objectives
229
settlement should cover (1) the failure to initiate an investigation upon request by a foreign country, which should be considered a violation of the positive comity rule,73 and (2) the failure to investigate anticompetitive effects in foreign markets. In evaluating such cases, the W.T.O. panel should be particularly sensitive to whether a transaction will result in the reduction of global output and/or trade, or otherwise impede market access.
4. Collection of Evidence and Enforcement Action Abroad The firms involved in transnational mergers may not have subsidiaries in all countries where effects occur. For this reason, antitrust authorities in some countries may encounter difficulties in collecting information required to reach an informed decision on the impact of a merger. They have no legal powers to demand information from firms located abroad, with no subsidiaries in their own jurisdiction. Likewise, once antitrust authorities have reached a decision not to permit a merger or to impose certain conditions, the enforcement of such a decision may be impossible if the companies are located abroad with no domestic subsidiaries. These problems may result in the unsatisfactory situation that a merger may produce substantial anticompetitive effects in a country, with antitrust authorities being unable to block the merger or require modifications. Therefore, some countries, although adversely affected, may lack effective control over a merger. This problem has already been discussed above, in the section which addressed anticompetitive behaviour by exporters in the import market. We concluded that inter-country cooperation agreements should be extended to the global level, including giving foreign firms legal standing in other jurisdictions, and with appropriate safeguards, providing for the exchange of information, the collection of data through foreign authorities, and the enforcement of decisions by authorities abroad.
73
The E.U. Group of Experts has suggested a similar construction of positive comity enforced via international dispute settlement. However, the Group suggests extraterritorial enforcement of national law as an alternative to dispute settlement where a foreign country refuses to initiate an antitrust investigation. By contrast, we consider extraterritorial enforcement of antitrust law only as a retaliatory measure upon authorisation by the W.T.O. after a panel found violation of positive comity obligations. See also Petersmann 1995, supra note 28, at 20, 22.
230
Working Paper V
5. Costs Due to Procedural Differences The initiation of merger control proceedings in multiple jurisdicitons inflicts high costs on administrative agencies and on merging firms. These costs are even greater when procedural differences exist.74 For example, merging firms must notify authorities in multiple countries, provide documents to all national authorities, deal with different forms, different procedures and deadlines, different timetables for approval, and possibly even different conditions for approval.75 Cooperation between national antitrust authorities, as well as enhanced transparency of merger control proceedings and harmonisation of some elements of procedural law, are desirable. Cooperation and harmonisation of procedural elements may reduce costs to enforcement agencies merging firms, as well as to their competitors, without jeopardising national sovereignty regarding substantive merger standards. The 1994 O.E.C.D. report by Richard Whish and Diane Wood identifies several areas where procedural cooperation may be beneficial, including coordination of timetables, development of model filing forms, and alignment of notification requirements.76 Moreover, the O.E.C.D. Committee on Competition Law and Policy has suggested convergence in merger review pro74
T h e occurrence of high costs involved with multiple national merger proceedings has been recognised in the O.E.C.D. Interim R e p o r t on Convergence of Competition Policies: '[i]ndeed there is concern that diverging competition laws and enforcement practices result in administrative, compliance a n d enforcement costs for government and the private sector, although these cost burdens have yet to be properly documented and are based for the most part only on anecdotal evidence'. O.E.C.D. Committee o n Competition L a w a n d Policy, Interim R e p o r t o n Convergence of Competition Policies, O . E . C . D . (95) 139/Final. 75 A s noted above, the Whish/Wood Report, supra note 23, examined 9 internationally i m p o r t a n t merger cases, in which more than 2 antitrust agencies were involved. In the 1992 merger between Gillette a n d Wilkinson in the market for wet-shaving razor blades, 14 antitrust agencies got involved, imposing tremendous costs on the merging firms as well as o n their competitors, Warner-Lambert a n d Bic. 76 W h i s h / W o o d R e p o r t , supra note 23, at 97-115. Harmonisation of procedures o n the basis of the W h i s h / W o o d Report found support by the 1995 E.U. G r o u p of Experts. Van Miert R e p o r t , supra note 27, at 17. Similarly, the American Bar Association Special C o m m i t t e e o n International Antitrust suggested in its 1991 report that 'sovereign states should strive for greater harmonisation regarding the timing and content of their various premerger reporting requirements. I n particular, the same types of information should be required in each jurisdiction'. Reprinted as annex to Whish/Wood Report, supra, at 121. See also F o x , supra note 5, at 33, who suggests c o m m o n merger notification a n d report forms, as well as enhanced transparency of national merger policies.
4 - Competition Policy Objectives
231
ceedings, such as early notification and contacts between national antitrust authorities involved in a case, model merger notification forms, and model forms that permit the waiver of confidentiality, thus facilitating the sharing of confidential information between national antitrust agencies.77 6. Dominance of Large Markets' Antitrust Regulation Smaller countries must generally accept merger control of multinationalfirmsby bigger countries. Small countries fear that a company may stop doing business in their market if merger enforcement is stricter than in larger markets. Rather than respecting higher merger control standards in a small country, a multinational firm might chose to abandon that market and proceed with a merger that is more likely to be approved in its major markets.78 As a consequence, merger standards of big markets tend to dominate the worldwide regulation of competition regarding global strategies, such as mergers, of multinational firms. Only convergence or harmonisation of merger control rules could remedy this imbalance between large and small countries. However, even in harmonisation negotiations, small countries would probably have to make concessions in order to reach a compromise. Therefore, the globalisation of markets results in the loss of power by smaller countries in the regulation of mergers, whether merger rules are harmonised or differ from country to country. 7. Regulation of Abuses of Dominant Position in Global Markets Various abusive practices by globally dominant firms may produce similar problems to those produced by mergers.79 Table 1, as well as the list of problems that can arise with respect to mergers in global markets, are valid not only 77
See O.E.C.D. Committee on Competition Law and Policy, 'Interim Report on C o n vergence of Competition Policies', 49 Aufienwirtschaft 334, 346 (1994); O.E.C.D., Revised Recommendation of the Council concerning Cooperation between M e m b e r Countries on Anticompetitive Practices Affecting International Trade, C(95) 130/Final (July 1995). 78 If there is continuous demand in such an ' a b a n d o n e d ' small c o u n t r y , other enterprises might attempt to trade the affected goods from a n e i g h b o u r m a r k e t t o the small country. Eventually, the small country must accept t h a t it is n o t able t o enforce stricter antitrust standards against multinational firms than major markets do. 79 The issue of abuse of dominant position in segmented markets has been discussed above, in the section addressing anticompetitive behaviour by exporters in the import market, which dealt with predatory pricing and abusive practices of multinationals in developing countries.
232
Working Paper V
for mergers but also for many global monopolisation cases or monopolistic strategies. An example for globally anticompetitive behaviour by a dominant firm is Microsoft's alleged abusive licensing practices, which in 1994 were simultaneously challenged by the U.S. and the E.U. Box 5.2: Microsoft—Monopolistic Selling Practices, Abusive Licensing In 1994, the U.S. and the E.U. filed antitrust actions against Microsoft, which resulted in two similar settlements between Microsoft on the one hand, and the U.S. and the E.U. on the other hand.80 It was the first time the U.S. and the E.U. cooperated in an antitrust investigation under the 1991 antitrust cooperation agreement. The charges against Microsoft, the dominant producer of computer operating software, included exclusionary per processor licences, unreasonably long licences, and non-disclosure agreements which restricted the development and sale of competing software products. Under its exclusive license agreements, Microsoft required PC manufacturers to pay a fee per PC sold, regardless of whether it contained pre-installed Microsoft software, thereby discouraging PC manufacturers from installing other operating software. In the settlements, Microsoft committed to discontinue the anticompetitive practices. Besides being an example of successful cooperation between antitrust agencies in the U.S. and the E.U., the Microsoft case also shows that there were no major differences in substantive antitrust laws between the E.U. and the U.S., which resulted in identical settlements between Microsoft and the two authorities. Nevertheless, the nature of Mircrosoft's practices did not prevent the U.S. or the E.U. from concluding different settlements with Mircosoft if they had followed different antitrust approaches. Although Microsoft's conduct was global, it could have been split up in marketspecific strategies in order to comply with different national antitrust laws, without directly affecting competition in foreign markets. Differences in national antitrust laws would not have produced negative spillover effects abroad. The evaluation of global abusive or monopolising behaviour depends on whether a company can pursue different strategies in different countries, while complying with different national antitrust rules. Generally, where market80
U.S. v. Microsoft Corp. 56 F.3d 1448 (D.C.Cir. 1995) Microsoft settles accusations of monopolistic selling practice, 67 Antitrust & Trade Reg. Rep. 106 (1994).
4 - Competition Policy Objectives
233
specific behaviour is possible, national antitrust authorities can continue to regulate those practices under their respective antitrust rules. Thus, products will be sold in each national market at a price and in a manner that is in accordance with the different national antitrust laws. However, if a business strategy is generically not divisible between markets, a firm will either pursue this strategy on a global level or drop it globally. This implies that large jurisdictions with strict antitrust enforcement will eventually set the standard. Smaller countries are not likely to be able to control global strategies, for if they attempt to do so, a multinational firm may abandon such a small market. The problems discussed above with respect to merger control apply also to regulation of abuses of dominant position, such as (1) difficulties in the collection of evidence and enforcement action abroad, (2) jurisdictional conflicts as to differences in the evaluation of positive and negative effects, (3) failure to consider varying anticompetitive effects abroad, and (4) procedural inconveniences for companies.
F. Strategic Antitrust Policy Governments sometimes attempt to give domestic industries a competitive advantage over foreign rivals. This is often called 'strategic trade policy'. The usual instrument to support domestic industries facing global competition is subsidies. Antitrust policy can be another means to enhance the power of domestic industries vis-a-vis foreign competitors. Instead of limiting market power of firms through strict antitrust enforcement, a government may grant certain exemption from antitrust discipline to firms facing global competition in order to increase their dominance in world markets. For instance, this may occur by exempting export cartels from the general prohibition of cartels, or allowing certain mergers. Such strategically motivated arguments in antitrust enforcement may be called 'strategic antitrust policy. In economic terms, such strategic antitrust exemptions aim at shifting rents from foreign companies to domestic industries, which may increase profits as the result of more sales in foreign markets, thereby benefiting national economic welfare. However, costs may occur as well if competition in the domestic market is substantially lessened by granting exemptions from general antitrust discipline, thus inflicting costs on consumers due to a possible price increase.
234
Working Paper V
1. Exemptions for Export Cartels Many jurisdictions, including Canada, the U.S., the E.U. and Japan,81 legally exempt export cartels from the general prohibition of cartels. Countries permit exporting firms to join export associations, which may range from common marketing to price fixing, in order to prevent domestic firms from competing against each other in foreign markets, and allowing them to join forces against foreign competitors, and to shift rents from foreign to domestic firms. As long as the colluding exporters continue to compete in the home market, there are no negative effects in the domestic markets. For instance, a deal has been entered between Siemens and G.E.C. Alsthom in Europe for joint marketing of high-speed trains in non-European markets against their common rival in Japan. Furthermore, several international cartels regarding certain raw materials, such as diamonds, in practice may have similar effects as export cartels, since domestic consumption in the producing country is unimportant. Toleration of an export cartel under antitrust law of the home country to exploit its market power abroad at the expense of foreign companies and consumers is a 'beggar-thy-neighbour' policy.82 One country tries to improve its economic welfare at the expense of another. There are two economic effects of such a policy. First, in an already high-price oligopolistic market, rents may be shifted from foreign to domesticfirms.The mere transfer of rents does not necessarily involve global economic welfare losses; rather, it is a 'win-lose strategy' or 'zero sum game. Secondly, export cartels may reduce competition and lead to output restrictions. Where output is being restricted and prices increase, consumers lose more than producers win, thus incurring total economic welfare losses. As a result, if export cartels reduce competition instead of merely shifting oligopolistic rents, national antitrust exemptions of export cartels
81
In the U.S., the Webb-Pomerene Act of 1918 exempts export associations from application of the Sherman Act, which prohibits cartels. In the E.U., the cartel prohibition in Art. 85 of the Treaty of R o m e applies only to restrictions of competition that have an effect on competition in the E.U., thus excluding export cartels that may have anticompetitive effects only abroad. Likewise, the Japanese Antimonopoly Act applies only to cartels that restrict competition in the Japanese market; A . M . A., supra note 26, Art. 6(1). 82 T h e O . E . C . D . recognises that 'there is growing consensus that export cartel exemptions, at least with respect to naked restraints, may be inappropriate beggar-thyn e i g h b o u r policies': O . E . C . D . Committee on Competition Law and Policy and T r a d e C o m m i t t e e , 'Joint R e p o r t ' , 18 World Competition 185, 188 (1994).
4 - Competition Policy Objectives
235
Box 6.1: Siemens! G.E.C. Alsthom—Export Cooperation in the Market for High-Speed Trains In July 1995, Siemens of Germany, the manufacturer of the InterCity Express train, and G.E.C. Alsthom, the Anglo-French producer of the Train Grand Vitesse, announced a project on the joint export marketing of high speed trains outside Europe and North America. The cooperation plan aims to end rivalry in export markets, and to join forces to compete with the Japanese rival, which is building the Shinkansen train in Japan. Mr. Heinrich von Pierer, chief executive of Siemens, predicts that at the end of the decade, 'you will have competition between a high-speed train made in Europe and a high speed train made in Japan. The project would include agreement on dividing up the world market among both European companies, ruling out that one company sells in the assigned territory of the other. It is most likely that the conclusion of such agreement would not violate European competition law, because the anticompetitive effects of reduced competition will occur only in foreign markets, whereas the increased profits will benefit the European companies. Source: the Financial Times, 25 & 26 July 1995. Andrew Baxter, Michael Lindemann and John Ridding, A transport of delight in the air: Will an 'Airbus on rails' plan get off the ground? Financial Times, 26 July 1995, p. 14.
result in a 'lose-lose strategy' or 'negative sum game, where economic welfare is lost internationally.83 We recommend two complementary measures as a remedy against export cartels. First, countries should agree reciprocally to abolish export cartel exemptions and abandon beggar-thy-neighbour policies.84 Although a ten83
There is, however, no 'race to the b o t t o m ' regarding antitrust standards, because any agreement a m o n g firms entailing anticompetitive effects in the domestic market remains under antitrust discipline. Only where anticompetitive effects are exclusively felt abroad does the exemption apply. Therefore, there is n o tendency to lower antitrust standards for domestic trade. 84 See also Petersmann 1994c, supra note 6, at 269; F o x , supra note 5, at 22; Van Miert Report, supra note 27, at 17. In contrast, Scherer suggests prohibiting export cartels in general, but allowing each country exceptions for three well-defined sectors. Those permitted export cartels are, however, actionable in importing countries, which are allowed to countervail a foreign export cartel through a domestic import cartel in order to balance market power. F . M . Scherer, Competition Policies for an Integrated World Economy 93 (1994).
236
Working Paper V
dency may exist for cartels to be unstable and to collapse,85 especially in a global market, export cartels should not be treated with different standards from domestic cartels.86 However, once the exemptions for export cartels have been abolished, there may be little incentive in the exporting country to initiate an investigation against an export cartel whose anticompetitive effects are in a different jurisdiction.87 As a result, the initiative to enforce antitrust law against export cartels must be by the importing country, where the negative effects occur.88 We therefore suggest that importing countries harmed by export cartels apply their antitrust law to foreign cartels according to the effects doctrine. Enforcement in the importing country requires that investigative tools be improved in order to become effective against foreign export cartels. As most jurisdictions contain per se prohibitions of domestic cartels, an investigation focuses on finding evidence whether an agreement exists, may might require extensive search in the exporting country. It is therefore essential that authorities in the importing countries have tools available to collect evidence in the exporting country. Cooperation between antitrust authorities is indispensable, allowing the importing country to request the collection of evidence by antitrust authorities in the exporting country and to use their compulsory powers to this
85
Even relatively stable cartels, such as the diamond cartel, seem to weaken eventually. 'It is the most durable cartel in recent history. But after 60 years it seems in danger of tearing itself a p a r t ' Kenneth Gooding et al., Diamond Cartel Cuts Up Rough: Relations are strained between the main producers, Financial Times, 24 Aug. 1995, p . 23. 86 Petersmann 1995, supra note 28, at 18, suggested a 'national treatment obligation in the sense that domestic competition laws must treat transborder cases (e.g. export cartels) n o less favorably t h a n domestic cases.' I n addition, as price-fixing, market-sharing, a n d supply-restricting cartels are per se prohibited in most jurisdictions, a complementary international rule could provide that such export cartels are n o t enforceable at law. Sir Leon Brittan, A F r a m e w o r k for International Competition, Address before the World Economic F o r u m in Davos, Switzerland, 3 F e b r u a r y 1992, at 10; Petersmann, 1994c, supra note 6, at 270. 87 'It is recognised that even with the exemption eliminated, the exporting country may find it difficult o r lack the incentive t o assert jurisdiction over an export cartel whose anti-competitive effects are felt only in foreign markets': O . E . C . D . Interim R e p o r t , supra note 78, at 344; O.E.C.D. Joint Report, supra note 83, at 188. 88 T h e U.S. D.O.J. a n d F.T.C. have clearly stated their will to enforce U.S. antitrust law against export cartels affecting the U.S. by asserting that subject matter jurisdiction is given under the Sherman Act. See 'Illustrative Example A ' in the 1995 Antitrust Enforcement Guidelines for International Operations, supra note 67, at 13.
4 - Competition Policy Objectives
237
end.89 Similarly, enforcement action in the exporting country requires assistance by authorities there. Cooperation between antitrust agencies regarding both the collection of evidence and enforcement action has been discussed in the previous section on anticompetitive behavior of exporters in the import market. 2. Strategic Evaluation of Mergers There is a widefieldfor including strategic arguments in the evaluation of other anticompetitive practices. Wherever antitrust law allows the balancing of the positive and negative effects of restrictive business practices, it is possible to include the argument of increasing international market power of domestic firms as a positive consequence of an otherwise anticompetitive practice. Such considerations are important in the area of merger control. The reasoning is similar to that of export cartels: domestic rivals should not waste their forces battling against each other, but instead join their forces to compete in global markets against foreign rivals. As a result, there is a tendency that merger control becomes weak against firms acting in global markets, in order to improve their market power vis-a-vis foreign rivals. Some national merger laws explicitly allow consideration of strategic aspects of mergers. For instance, the Canadian merger law takes into account whether a merger is likely to increase exports.90 In the E.U., efficiency gains and
89
'Effective enforcement against export a n d other cartels will necessarily centre o n the jurisdiction(s) in which the anticompetitive effects are felt although much of the information necessary for successful prosecution will often be located in a n o t h e r country. This m e a n s in turn that if M e m b e r countries wish to facilitate action against such agreements, they would need to focus on developing for competition officials the legal mechanisms for cooperation in international cartel investigations and especially for the sharing of information a m o n g national competition offices': O . E . C . D . Interim Report, supra note 78, at 344; O . E . C . D . Joint Report, supra n. 83, at 188. Diane W o o d , former D e p u t y Assistant Attorney General in the Antitrust Division of the U . S . D.O.J., pointed to the necessity of cooperation in the area of export cartels, where a 'strict territoriality' a p p r o a c h and non-cooperation would create 'antitrust havens'. 'If cooperation with the legitimate investigations of the countries where the effects of such conspiracies are felt is not forthcoming, the general cause of strong antitrust enforcement is h a r m e d ' : W o o d , supra note 54, at 6. 90 C a n a d i a n antitrust law permits consideration of efficiency gains of mergers, including increases in exports: '[i]n considering whether a merger or proposed merger is likely to bring a b o u t gains in efficiency . . ., the Tribunal shall consider whether such gains will result in (a) a significant increase in the real value of exports': C a n a d i a n Competition Act, Art. 96(2).
238
Working Paper V
industrial policy have an ambiguous importance in the 1989 Merger Regulation: they are permitted as long as competition is not restricted and they benefit consumers.91 Unfortunately, merger control practice is not sufficiently transparent as to the degree to which strategic arguments have played a role. By contrast, the U.S. merger guidelines and merger control practice allow for efficiency gains, but not for strategic reasons.92 However, in contrast to export cartels, which have no direct negative domestic effect, lax merger control for strategic trade policy reasons may have anticompetitive effects in the domestic market. Consequently, there is a danger of a 'race to the bottom', meaning that merger discipline may become increasingly relaxed in order to give domesticfirmsa competitive advantage in foreign markets, but at the expense of domestic consumers. A collective action problem arises from a 'prisoner's dilemma,' where every country would be better off giving up strategic merger control, but no one has an incentive to do so unilaterally. Only international cooperation can solve the dilemma through reciprocal commitments. We recommend eliminating strategic arguments in merger control proceedings through a reciprocal international agreement. The prohibition of strategic issues should cover all arguments aiming at an increase through a merger of international market power, international market share, the volume of exports and profits at the expense of a possible reduction of competition and output in the domestic market. The prohibition of strategic arguments should not include issues relating to generic efficiency gains through a merger, such as cost savings or technical rationalisation.93 Consequently, there should not be dif91
T h e E.U. merger regulation states: '[t]he Commission shall take into a c c o u n t : . . . the development of technical and economic progress provided that it is to consumers' advantage and does not form an obstacle t o competition': 89 OJ L395 f 1989] 1, Art. 2: l(b), R e.g. 4064/89. Jacqemin concluded: 'The regulation apparently does not provide for authorization in derogation from the prohibition, o n the basis of the efficiency effects of the merger': Alexis Jacquemin, Mergers and European Policy, in P.H. Admiraal (ed.), Merger and Competition Policy in the European Community (1990), 32. Furthermore, the rejection of the merger by Aerospatiale-Alenia and de Havilland ([1991] O.J. L334/42) is often interpreted as showing that industrial policy has little importance in merger control. 92 T h e U . S . Merger Guidelines by the D.O.J. and the F . T . C . d o not include a provision allowing consideration of'efficiency gains' due t o enhanced international competitiveness or export sales. An inexhaustive list of efficiency gains from mergers is included in A r t . 4 of the 1992 D . O . J . / F . T . C . Merger Guidelines. U.S. Dept. of Justice/U.S. Fed. T r a d e C o m m i s s i o n , 1992, Horizontal Merger Guidelines, 41 Fed. Reg. 552, 563. 93 T h e 1992 U . S . D . O . J . / F . T . C . Merger Guidelines include some generic efficiency a r g u m e n t s t o be allowed for in merger proceedings, such as cost savings, economies of scale, better integration of production facilities, plant specialisation and lower t r a n s p o r t costs: 2d.
4 - Competition Policy Objectives
239
ferential treatment of mergers with international importance versus those of primarily domestic importance.94
3. Exemptions for Research and Development Cooperation Antitrust exemptions for research and development cooperation are another area where countries do not strictly enforce antitrust law, with the goal of promoting the positive effects of inter-firm cooperation on competition and economic welfare,95 as well as increasing the international competitiveness of domestic industry. Antitrust exemptions for research and development joint ventures may be abused for strategic trade-policy reasons, and may lead to a tendency increasingly to ignore anticompetitive effects. Antitrust exemptions for cooperation in research and development exist in many jurisdictions. In the U.S., the National Cooperative Research Act of 1984 recognises the pro-competitive effects of R&D joint ventures, and provides that those positive effects should be permitted in antitrust evaluation. Similarly, the legally non-binding 1980 Antitrust Guide Concerning Research Joint Ventures by the D.O.J. emphasises the procompetitive effects of R&D cooperation. The E.U. has adopted a research block exemption from the
In the E.U., it is unclear whether, and to what extent, efficiency gains may be considered in defence of a proposed merger. The 1989 Merger Regulation apparently allows only for efficiency gains that are not in conflict with competition. The Canadian Competition Act states that increases in exports shall be considered as efficiency gains would not comply with its definition of efficiency. Competition Act, R.S.C. ch C-34, Part VIII, Sec. 96(2) (1984). 94 Petersmann 1995, supra note 28, at 18, suggested, however, with regard to export cartels, a 'national treatment obligation in the sense that domestic competition laws must treat transborder cases (e.g. export cartels) no less favorably than domestic cases The Max Planck Draft International Antitrust Code contains a provision permitting a cost-benefit analysis in national merger appraisals, if an approval of the merger does not harm the interests of another country: D.I.A.C., supra note 21, Art. 12, Sec. 1; Drexl et al., supra note 27. This implicitly outlaws justification of a merger for strategic arguments. 95 Governmental promotion of R&D activities is economically justified by market failure due to insufficient appropriability of the benefits of R&D activity: Alexis Jacquemin, Comments, in F.M. Scherer, Competition Policies for an Integrated World Economy (1994), 99-105.
240
Working Paper V
general prohibition of cartels.96 Furthermore, development joint ventures are likely to obtain an individual exemption under Art. 85(3). International minimum standards should be established in order to limit possible anti-competitive effects. In general, R&D cooperation between firms has positive economic effects. However, exemptions from antitrust discipline may cause external effects if national antitrust administrations tend to disregard domestic anticompetitive effects, where domestic firms may win a competitive advantage over foreign firms. Countries may be unable to prevent excessive allowance of cooperative agreement between firms on a unilateral basis, due to a collective action problem. Therefore, an international agreement is necessary. An international agreement on antitrust treatment of R&D cooperation should define minimum standards for the kind of inter-firm cooperation that should be allowed because it is likely to bring about generally positive effects, and the kind that should be prohibited because it is likely to bring about domestic anticompetitive effects and distortion of international competition.97 A general prohibition of R&D cooperation between firms is neither desirable nor adequate, and would ignore positive economic effects brought about by those agreements. In consideration of the similarities between antitrust exemptions and governmental R&D subsidies, our suggestion is a parallel to the G. A.T.T./W.T.O. Uruguay Round Agreement on Subsidies and Countervailing Measures.98 The W.T.O. subsidy agreement takes a 'traffic light' approach in establishing three categories of subsidies: 'red': prohibited subsidies; 'yellow': allowed, but actionable subsidies; and 'green': non-actionable subsidies. The green category contains subsidies for R&D,99 which are considered procompetitive, not trade 96
[1985] 85 O J L53/5, Reg. 418/84, amended by [1993] 3OJ L21/8, Reg. 151/93. A l t h o u g h the M a x Planck Draft International Antitrust C o d e suggested comprehensive international m i n i m u m substantial s t a n d a r d s , it failed to set detailed rules for R & D c o o p e r a t i o n : D . I . A . C . , supra note 28, A r t . 4, Sec. 2; Drexl et al., supra note 28. This, however, is the only area where we perceive a need for harmonised international minimum standards. 98 F o r a discussion of the Uruguay R o u n d Agreement, see Jeffrey J. Schott, T h e U r u g u a y R o u n d : A n Assessment (1994), 8 6 - 9 3 , 158-61. There has been some convergence between the E . U . a n d the U.S. on the issue of R & D subsidies, and that this was one of the factors allowing a subsidies agreement to be reached. T h e U.S. accepted the m o r e interventionist a p p r o a c h of the E . U . 99 Agreement on Subsidies and Countervailing Measures, Art. 8. G. A.T.T., Results of t h e U r u g u a y R o u n d Multilateral T r a d e Negotiations: T h e Legal Texts (Geneva, 1994). T h e agreement m a k e s a distinction between research subsidies, which may cover up t o 75 per cent of the costs, and 'pre-competitive' development subsidies, which m a y cover 50 per cent of the costs. Further specifications are included in the agreement. 97
4- Competition Policy Objectives
241
distorting, and not injurious to foreign competitors.'100 Likewise, an international antitrust agreement should include a provision establishing how broadly to define R&D, which will be granted antitrust exemption (green category). All other horizontal agreements betweenfirms(red category) may not be exempted from national antitrust laws, most of which contain per se prohibitions regarding agreements on price-fixing, output restriction, and market sharing.
G. Concluding Remarks This paper has focused on market access restrictions due to restrictive business practices and competition (antitrust) law and policy. Market access is a basic tenet which both trade and competition policy experts will embrace. However, a distinction must be made between restrictions to market access by governments versus those by private firms. International restrictive business practices should be addressed only where market access is restricted and there is insufficient relief available through existing legal-institutional arrangements. We made suggestions of possible approaches for dealing with restrictions of market access without interfering with the freedom of each country to regulate internal commerce. We have, until now, respected the economic principle of competition among rules, in the absence of external effects, as well as the G. A.T.T./W.T.O. principle of national sovereignty, to adopt non-discriminatory internal regulations including antitrust law. Harmonisation or adoption of minimum standards was suggested only where national antitrust regulation has negative spillover effects on the competitive process of other countries. However, this approach has limits where market access is being restricted due to either differing substantive antitrust standards or discriminatory enforcement or non-enforcement of existing antitrust law. Where a country either has no rules against anticompetitive behaviour by private firms that restrict market access to foreigners, or does not enforce its law, on a nondiscriminatory basis, the previously suggested remedies are not sufficient to foster market access. There is no relief withfilingan antitrust case in the country where the restrictive business practices occur. In addition, restricted market access which stems from anticompetitive business conduct in a given country may be tolerated by government(s), since 100
However, if 'green' subsidies cause injury to a foreign country, they may nevertheless be actionable. The foreign country would have to prove that those subsidies have a 'serious adverse effect' that results in 'damage which would be difficult to repair': id., Art. 9.
Abolish antidumping International cooperation of antitrust agencies: collection of evidence and enforcement action abroad Comity rules
Adoption and enforcement of national antitrust laws in the importing country (effects doctrine)
Predatory pricing Import and export restrictions for subsidiaries of M.N.E.s Abusive licensing of technology
Anticompetitive Behavior by Exporters in the Importing Country
Legal standing of exporters without subsidiaries Legal standing of private parties Strengthen private action suits Political independence of administrative agencies International dispute settlement if enforcement is discriminatory
Enforcement of antitrust law in the country where market access is restricted via filing of cases by affected exporters
Exclusive dealership Long-term business relationships Vertical integration Distributor boycotts Abusive practices by trade associations
Restriction of Market Access by Import- Competing Firms
Complementary Measures through an Agreement at the International Level
Practices
Category
Remedies at the National Level
Table 2: Remedies to Antitrust Problems in International Trade
f
Level Playing Field
Accept different approaches in antitrust policy as the deliberate choice of a nation
Abolition of export cartel exemptions Prohibition of strategic arguments in merger control International minimum standards for R&D cooperation Discuss convergence of national antitrust laws
Export cartels: enforcement of antitrust laws in the importing country (effects doctrine)
Strategic Antitrust Policy
Antitrust exemptions for export cartels Exemptions for R&D joint ventures Strategic evaluation of mergers Less strict antitrust rules or enforcement may give companies a competitive advantage over rivals in foreign markets
International cooperation of antitrust agencies: collection of evidence and enforcement action abroad Comity rules Non-discrimination rules International notification, consultations, and dispute settlement Selective procedural harmonisation
Enforcement of national antitrust laws; diversity of national antitrust laws
Spillovers of Antitrust Mergers Regulation Global monopolisation strategies
-fc.
I
re
I
Working Paper V
244
Table 3: Three Pillars for an International Antitrust Code International Standards
Requirement to enact national antitrust laws Allow for private action suits Guarantee political independence of administrative antitrust authorities Grant legal standing to private parties and foreigners No discrimination against foreigners Abolition of antidumping Prohibition of strategic antitrust policy: Prohibition of export cartels and strategic arguments in merger control as well as minimum standards for R&D cooperation Procedural minimum standards for merger control
Cooperation Between Antitrust Authorities
Collection of evidence including confidential information abroad through foreign agencies Assistance in enforcement action abroad by foreign authorities Negative and positive comity
International Dispute Settlement
W.T.O. dispute settlement regarding all above obligations
domestic firms benefit from a competitive advantage vis-a-vis their foreign rivals. Cases may arise wherefirmsgain a competitive advantage over others in domestic and/or international markets. In other words, an 'uneven playing field' between competingfirmsin different countries may emerge due to differences in antitrust policy. Two alternatives should be considered to deal with an uneven playing field due to differences in antitrust policy. One is to harmonise antitrust law internationally, or at least set common minimum standards. The second is to accept nationally diverging approaches toward anticompetitive conduct, even when market access of foreign firms may be restricted through government toleration of restrictive business practices. We believe that harmonisation is not feasible or necessary. Convergence in competition law principles, and the development of minimum standards, are preferred. Moreover, countries should strive to adopt, modernise and effectively enforce competition law, and forge cooperative links among antitrust authorities. In order to assist the process offindingcommon grounds for resolving international competition conflicts, this paper has attempted to identify a 'road-
4 - Competition Policy Objectives
245
map' of case-specific issues that are likely to arise. This should be examined by the Working Group on Trade and Competition being formed within the W.T.O. The suggested approaches towards resolving the various problems are summarised in Tables 2 and 3.
Other References H. G. BROADMAN, G.A.T.T.: The Uruguay Round Accord on International Trade and Investment in Services, 17 World Economy 281 (1994). ANNE K. BINGAMAN, The Role of Antitrust in International Trade, Address before the Japan Society in New York on 3 March 1994, Washington, D.C.: U.S. Department of Justice, 1994. COMMISSION OF THE EUROPEAN COMMUNITIES, Towards an International Framework of
Competition Rules, 18 June 1996. JONATHAN FAULL, The Future World Trade Agenda: The New Issues of Competition Policy, Address presented at The Royal Institute of International Affairs, 18 November 1996. HEINZ HAUSER and RAINER E. SCHONE, IS There a Need for International Competition
Rules?, 49 AuBenwirtschaft 205 (1994). BERNARD M. HOEKMAN, Competition Policy and the Global Trading System: A Developing Country Perspective (1997). INTERNATIONAL BUSINESS LAWYER, Internationalisation of Competition Law, November
1996. FREDERIC JENNY, The Interface Between Competition Policy and Trade, Investment and Economic Development, Paper presented at the Global Forum Conference on Competition Policy in a Global Economy, New Delhi, 17-18 March 1997. NAHEED KIRMANI, et al., International Trade Policies: The Uruguay Round and Beyond. Volume I: Principal Issues; Volume II: Background Papers (IMF, 1994). JOHN E. KWOKA and LAWRENCE J. WHITE (eds.), The Antitrust Revolution: The Role of
Economics, (1994). PHEDON NICOLAIDES, Towards Multilateral Rules on Competition. The Problems in Mutual Recognition of National Rules, 17 World Competition 5 (1994). DAVID PALMETER, Antitrust Rules and the G.A.T.T., Journal of Commerce, 18 March 1994. ERNST-ULRICH PETERSMANN, International Competition Rules for the G.A.T.T.-M.T.O. World Trade and Legal System, 27 J. World Trade 35 (1993). W. STEPHEN SMITH, U.S. Antitrust Law and Practice: Monopolisation and Related Offenses, in Garrett James J. (ed.) World Antitrust Law and Practice (1995). JOEL P. TRACHTMANN, International Regulatory Competition, Externalisation, and Jurisdiction, 34 Harv. Int'l L.J. 47 (1993).
VI Joel I. Klein Assistant Attorney General Antitrust Division, U.S. Department of Justice Washington, D.C., U.S.A.
Introduction Last December, the World Trade Organisation held a Ministerial Conference in Singapore, which I attended, along with many others. There, the European Union and other W.T.O. Members proposed the creation of a Working Group to initiate development of a trade and competition agenda at the W.T.O. The government and non-government proponents of this approach believe that significant market access problems exist, whose solutions lie in the application of competition law, and that negotiation in the W.T.O. of multilateral rules on the application of competition law would help to alleviate these problems. After much discussion, Ministers agreed to 'establish a working group 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 W.T.O. framework'. After two years, the W.T.O. General Council will determine how (or whether) the work of the group should proceed; in particular, the Singapore Declaration stated that '[i]t is clearly understood that future negotiations, if any, regarding multilateral disciplines in th[is] areafj, will take place only after an explicit consensus decision is taken among W.T.O. Members regarding such negotiations.' The new Working Group will operate under the expert Chairmanship of Frederic Jenny of France. I have quoted the precise Ministerial language because my goal in this paper is to explain why the United States concurred in the Singapore Declaration's cautious approach to a W.T.O. role in this area, and why the Antitrust Division of the Department of Justice fully endorses that approach. In making the case for caution, this paper will begin by explaining recent efforts in international antitrust enforcement, how those efforts relate to traditional tradeliberalisation concerns, and what would be lost if a misguided W.T.O. agenda were to derail or detract from these ongoing efforts.
* This paper was presented by Connie Robinson, Director of Operations and Merger Enforcement, United States Department of Justice, Washington D.C.
248
Working Paper VI
A. The Intersection of Trade and Competition Policy The events that have sparked an intense interest at the intersection of trade and competition policy are well known. Two factors have converged to bring this matter to the fore: first, an increasingly globalised economy, spurred largely by technological advances, has made markets throughout the world economically available even to previously domestic businesses; and secondly, the successive reductions of government-imposed barriers to trade (resulting from the various G.A.T.T. rounds) has made entry into foreign markets not only economically feasible, but also practically feasible. Taken together, these developments have led to an explosion in worldwide trade—in 1996, there was over $5 trillion in merchandise and nearly $1.5 trillion in services traded beyond national borders; and in the U.S., for example, nearly one-quarter of our G.D.P. is comprised of export and import trade, which is double the figure for 1945. This ongoing process of globalisation has important consequences in terms of conventional trade concerns. Elimination of governmental restraints has brought into focus private business practices that inhibit market access. Globalisation also has great significance in terms of international competition issues, which fall into three categories: • the tremendous growth in transnational mergers has increasingly led to premerger review of the same transaction by several different countries' competition authorities; • international cartel cases, where competitors in various countries get together secretly to fix prices or allocate territories on a worldwide basis, have assumed increasing prominence; • market-access cases, in which anticompetitive horizontal or vertical restraints prevent foreign competitors from being able to compete on a level playing field, have also become more frequent. An international antitrust agenda should focus on all three areas, whereas a sensible trade programme need only focus on the third. We should consider what is at stake in each of these areas of international antitrust enforcement, what specific problems are encountered in addressing each area, and what efforts are already underway to deal with those problems. Transnational merger review presents fewer and less urgent problems than the other two. The Antitrust Division has worked with its counterparts abroad in several recent merger cases, including the Scott PaperlKimberly Clark1 and the 1
United States v. Kimberly Clark Corp. 1996-1 Trade Cas. (CCH) 171,405 (N.D. Tex. 1996).
4 - Competition Policy Objectives
249
Georgia Pacific/Domtar2 mergers. Transnational mergers are not being inhibited by the necessity for multiple reviews. Moreover, at least on an informal level, the various enforcement agencies are already engaged in some cooperation that is likely to increase over time and likely to lead to more formalised merger review cooperation agreements in the years to come — initially on a bilateral, and then, perhaps, on a plurilateral, basis. In addition, merging firms that are subject to multiple reviews can facilitate coordination and cooperation among the various competition agencies by authorising them to share otherwise confidential information. There is, nonetheless, room for further improvement in this area and, in particular, additional measures may be devised to make the multi-agency review process more efficient and less burdensome. The O.E.C.D.'s Competition Law and Policy Committee is currently analysing proposals of this sort. The two other areas on the agenda of international antitrust enforcement where I believe that new arrangements between governments have the potential to make significant advances are cartel enforcement and market access cases. Both have important, although differing, effects on international trade. International cartels typically involve arrangements among manufacturers or producers of goods that sell in international markets. Just as occurs within domestic markets, the sellers of such goods in international markets sometimes conclude that collusion is preferable to competition, and decide to agree either on prices, volumes, or the markets each will sell in. Two such cartel cases were recently prosecuted in the United States. One involved the $600 million market for lysine (a farm feed-additive product), and the other involved the $1.2 billion market for citric acid. The defendants in these investigations have agreed to pay a total of nearly $200 million in fines thus far, and the citric acid investigation is continuing. A prominent U.S. corporation, Archer Daniels Midland (A.D.M.), has been sentenced to pay $100 million for its participation in these cartels, a fine that is almost seven times larger than any we had previously obtained. The Antitrust Division would have sought a larger fine if A.D.M. had not agreed to cooperate fully in our ongoing international investigation in citric acid. The international nature of these two investigations is apparent, as they have resulted in guilty pleas from A.D.M., two Japanese firms, a Korean firm, a U.S. subsidiary of another Korean firm, a U.S. subsidiary of a German firm, and two Swiss firms, and from two Japanese, one Korean, one Austrian, and two German nationals.
2
United States v. Georgia Pacific Corp., 1996-2 Trade Cas. (CCH) 1J71, 560 (D. Del 1996).
250
Working Paper VI
Three present or former A.D.M. officials, all U.S. nationals, have been indicted and are awaiting trial, and one Japanese national has been indicted and is a fugitive. Such cartel cases can have an impact on trade by taking enormous amounts of money out of the pockets of consumers around the world. However, they rarely engage the attention or concern of the trade community. The companies that participate in such cartels, regardless of their country of origin, actually benefit from them. However, consumers (including consuming firms) are hurt. For example, at the U.S. sentencing hearing in the lysine case, Ajinomoto, a Japanese participant in the conspiracy, asserted that almost three-quarters of its sales occurred in Japan, where Japanese consumers seem also to have borne the brunt of this illegal cartel. Moreover, consumers in Europe and throughout the world who purchased these products are all likely to have suffered by having to pay artificially inflated prices as a result of the cartels' practices. Finally, market access cases involving private business restraints present not only the traditional antitrust concern posed by anticompetitive business practices, but also a significant trade concern arising from the impact of such practices on exports. In such cases, consumers in a domestic market are normally harmed because foreign companies are blocked from becoming effective competitors as a result of practices by domestic businesses that may violate domestic antitrust laws. From an antitrust perspective, this is the same as a situation where a new domestic entrant is prevented from being an effective competitor as the result of private restraints. In either case, domestic consumers are harmed because competition has been diminished. From the trade perspective, however, there is a special concern when foreign competition is kept out: the excluded country's economy is hurt by the limitations placed on its businesses, leading to the loss of jobs and overall domestic economic well-being. This set of trade and competition concerns in market access cases is motivating much of the current effort to find new mechanisms to provide redress. In an increasingly globalised economy, stimulated in part by the consistent reduction of government-imposed restraints on market access, it is hardly surprising that trade officials would have hit against private market restraints that impede foreign access. Some of these restraints may have existed even when government-imposed trade restraints were more numerous, while others may have arisen in response to the ongoing removal of government restraints. B. The U.S. Agenda for International Cooperation In searching for a solution to the market access and international cartel problems, the U.S. Department of Justice has focused largely on bilateral agree-
4 - Competition Policy Objectives
251
ments and working relationships with other antitrust enforcement agencies. This section will discuss the reasons for this approach.
1. Cooperation in Gathering Evidence Regarding international cartels, the procedures available to antitrust enforcers for investigating and prosecuting these cases are not commensurate with their international scope. International litigation often raises questions of personal jurisdiction and service of process, and normally presents great difficulties in terms of an enforcement agency's ability to obtain documentary and testimonial evidence located abroad. Antitrust enforcement is fact-intensive, almost invariably placing a high evidentiary burden on enforcers. When they cannot get access to the evidence needed to prosecute a violation, the world's consumers and businesses ultimately bear the cost. Several recent cases illustrate these problems. For example, three years ago in the GE/DeBeers3 case, the Antitrust Division filed criminal antitrust charges against a U.S. company, General Electric, a Swiss affiliate of DeBeers, and two foreign nationals, for conspiring to raise the price of industrial diamonds. Much of the alleged conduct relating to the cartel took place in Europe. Thus, much of the evidence was located overseas and consequently beyond the Justice Department's reach, although we did seek and receive some assistance from the government of Belgium. The case proceeded to trial but, in December 1994, the court entered a judgment of acquittal, observing that much of the 'missing' evidence presumably was located outside the U.S., and beyond our reach. Efforts of the Antitrust Division to cooperate effectively with other antitrust authorities can also be stymied by the absence of arrangements that allow the sharing of our own evidence with those authorities. In criminal investigations, for example, U.S. federal rules of criminal procedure are very strict in protecting the secrecy of grand jury proceedings. Fortunately, we have powerful new legislation that allows the sharing of cartel evidence, but to take advantage of its provisions, other countries must be willing to cooperate on a reciprocal basis. For example, in the lysine and citric acid cases, we have uncovered and continue to develop evidence of price-fixing and market allocation that should be of vital interest to competition authorities in other countries. Our ability, however, under current law to share such information with countries that are not parties to cooperation agreements is very limited. United States v. General Electric Co., 869 F. Supp. 1285 (S.D. Ohio 1994).
252
Working Paper VI
The Clinton Administration and the U.S. Congress, both recognising the critical importance of international cooperation and information sharing necessary for high priority investigations, recently gave us explicit authority to negotiate bilateral antitrust cooperation agreements in the International Antitrust Enforcement Assistance Act of 1994 ('I.A.E.A.A.').4 Once adopted, these agreements will allow the U.S. antitrust agencies to exchange evidence on a reciprocal basis with foreign antitrust agencies, for use in antitrust enforcement, and to assist each other in obtaining evidence located in the other's country, while assuring that confidential information will be protected. Negotiation of properly tailored I.A.E.A.A. agreements with our trading partners is a top priority for the Department of Justice. In April, the D.O.J. and the F.T.C. announced the first such agreement—with Australia.5 Both we and the Australians have requirements we must satisfy before the proposed agreement can be finalised. We are delighted with it, both because it will strengthen our already excellent relationship with the Australian authorities, and because we hope it will serve as a model for similar bilateral agreements with other important trading partners around the world. In further pursuit of this agenda, the United States also recently proposed an initiative in the O.E.C.D.'s Competition Law and Policy Committee to work towards a recommendation urging the adoption of bilateral agreements directed at hard-core cartel activity, which involves the most widely accepted antitrust violations. I believe this proposal will be adopted, after appropriate consideration, for several reasons. First, identifying the type of egregious anticompetitive conduct that constitutes a hard-core cartel is relatively straightforward. In most instances, there is little or no economic debate about the competitive effects of this conduct. Secondly, business community concerns about the sharing of information with foreign antitrust authorities are far less germane to the sharing of information for cartel enforcement purposes. This involves evidence of flagrant wrongdoing: discussions related to price-fixing, market allocation or bid-rigging, and evidence of agreements to pursue types of conduct that competition authorities generally agree can have no legitimate business purpose and therefore should not benefit from rules intended to protect business planning. There is, in short, almost never a need for antitrust authorities to examine, much less to share, the sensitive trade secrets or 4
15 U.S.C. §§6201-6212, Pub. L. No. 103-438, 108 Stat. 4597. Agreement Between the Government of the United States of America And The Government of Australia on Mutual Antitrust Enforcement Assistance. See also Department of Justice Press Release, International Enforcement to be Boosted by New Agreement with Australia, 17 April 1997. 5
4 - Competition Policy Objectives
253
prospective business plans of the kind that may be needed in connection with mergers and other economically complex inquiries. Perhaps for these reasons, we have begun to meet with considerable success over the last year in our requests to numerous foreign governments (in addition to Canada) for assistance in criminal cartel cases, pursuant to both M.L.A.T.s and more traditional letters rogatory. In those matters, foreign government authorities have provided us with information, interviewed witnesses and/or conducted searches and seizures on our behalf. Nonetheless, we recognise that convincing nations that it is in their interest to enter into such bilateral agreements will not be easy, and that differences in the substantive and procedural rules in different countries will have to be carefully worked through. Most countries, for example, do not impose criminal penalties for violation of their competition laws. Moreover, important cultural and sovereignty issues must be resolved when such agreements are contemplated. I believe that such differences ultimately will not stand in the way of cooperation aimed at eliminating cartels. Other models of law enforcement cooperation support this conclusion. For centuries, governments have worked together in law enforcement when it has been in their mutual interest, as when fugitives seek to evade punishment by fleeing the jurisdiction. The United States signed its first extradition treaty (with the United Kingdom) in 1794. Since the 1970s, the Department of Justice and the Department of State have made it a high priority to negotiate mutual legal assistance treaties (M.L.A.T.s), which provide for comprehensive reciprocal assistance between the United States and foreign governments in criminal matters. Currently, M.L.A.T.s are in force, and many others are signed and awaiting ratification. In addition, the United States has recently begun to work cooperatively with other governments through mutual assistance agreements in other areas of law enforcement: tax and securities fraud. Accordingly, the Securities and Exchange Commission has entered into roughly twenty such agreements, which have significantly enhanced its ability (and that of its foreign counterparts) to deal with transnational securities fraud. Moreover, in the antitrust area, the U.S.-Canada M.L.A.T. has permitted the Department of Justice and the Canadian antitrust authorities to conduct a series of joint criminal investigations into price-fixing and market allocation conduct that affected both countries. These matters are extremely important. For instance, in the Fax Paper cases,6 the D.O.J. charged six Japanese firms, 6
E.g., United States v. Kanzaki Specialty Papers, Crim. No. 94-10176NMG (D.Mass. 1994); and United States v. Mitsubishi Paper Mills, Crim. No. 95-10296MLW (D. Mass 1995).
254
Working Paper VI
one U.S. firm, two U.S. subsidiaries of Japanese firms, the U.S. subsidiary of a Swedish firm, five Japanese nationals, and one U.S. national with price-fixing in the fax paper market. Eight defendants agreed to plead guilty and pay fines totalling nearly $10.5 million, and a jury recently acquitted a U.S. firm and a U.S. national. For their part, the Canadians have charged, and obtained guilty pleas and significant fines from, some of the same firms, and I understand that their investigation is continuing. The remaining two Japanese corporate defendants in the Fax Paper cases moved to dismiss the indictments against them, arguing that United States courts had no subject matter jurisdiction over them in a criminal antitrust case because none of the alleged overt illegal acts occurred in the United States. The trial court agreed, and dismissed the indictment.7 In March, however, a U.S. Court of Appeals reversed the district court, holding that controlling Supreme Court precedent establishes that the Sherman Act 'applies to wholly foreign conduct which has an intended and substantial effect in the United States', in criminal as well as civil cases.8 The Court of Appeals recognised that '[w]e live in an age of international commerce, where decisions reached in one corner of the world can reverberate around the globe.... [A] ruling in [defendants'] favor would create perverse incentives for those who would use nefarious means to influence markets in the United States, rewarding them for erecting as many territorial firewalls as possible between cause and effect.' The D.O.J. agrees with this analysis. The Canadians first brought the Fax Paper case to our attention at a time when we were unaware of the conspiracy and of its harmful effects on U.S. consumers. The U.S. and Canadian authorities proceeded to work very closely together, exchanging information within the limits of the M.L.A.T., sharing documents and jointly interviewing witnesses. Both the cooperative analysis of documentation and the sharing of a database created by the Canadians were possible without violating any domestic confidentiality rules. Two other examples of cooperative antitrust enforcement under our M.L.A.T. with Canada are the Plastic Dinnerware case9 and an investigation
7
United States v. Nippon Paper Industries, Co., 1996-2 Trade Cas. (CCH)\11,575 (D. Mass. 1996). 8 United States v. Nippon Paper Industries, Co., 1997-1 Trade Cas. (CCH) 1(71,750 (1 st Cir. 1997), cert denied, Nippon Paper Industries, Co. v. United States, 118 S. Ct 685 (S. Ct. 1998). 9 See Annual Report on Developments in Competition in the United States (January 1-September 30, 1994), Submitted to the Organization for Economic Cooperation and Development, Committee on Competition Law and Policy, pp. 13-14.
4 - Competition Policy Objectives
255
in the ductile pipe industry.10 In the former, the Federal Bureau of Investigation and the Royal Canadian Mounted Police simultaneously executed search warrants on both sides of the border. The evidence thereby obtained ultimately led to U.S. price-fixing prosecutions of, and guilty pleas from, three U.S. firms and seven executives, including two Canadians, with fines totaling over $9 million and jail sentences for all seven individuals, including the Canadians. The seized documents revealed that the conspiracy did not affect the Canadian market, so no Canadian investigation ensued. Lastly, the D.O.J. and the Canadians conducted parallel investigations into anticompetitive behaviour in the ductile pipe industry. The D.O.J. concluded the evidence was not sufficient to prosecute under U.S. law. However, the Canadian authorities assembled a different body of evidence of violations of Canadian law that led to a guilty plea and a then-record criminalfinein Canada, from a Canadian subsidiary of a U.S. firm. The U.S. would like to extend the type of cooperation that has been so successful with Canada to our other trading partners. As our Canadian experience suggests, in order to be effective and sustainable, bilateral cooperation must provide law enforcement benefits for both parties. All responsible antitrust authorities should endorse the first step of the U.S. international antitrust agenda, the promotion of meaningful international cooperation in the prosecution of global cartels.
2. Cooperation to Address Market Access Problems According to the experience of the D.O.J., the most effective way to redress private restraints barring access to foreign markets is to empower competition authorities and to insulate them as much as possible from short-term protectionist influences. This is the situation in the U.S. In several instances, such as the AT&T case," the U.S. markets were open to foreign competition by challenging private restraints that protected domestic competitors. From the perspective of competition policy, this makes sense, since more competition implies greater benefit to consumers. The problem in this area, however, is that not every country has the same history or tradition of independent enforcement in market access cases. With respect to those countries that do have such a tradition, bilateral 'positive comity' agreements are the best way to ensure effective enforcement. 10 11
See Regina v. Canada Pipe Co., 64 C.P.R. 3d 182 (Canadian Federal Court 1995). United States v. AT&T, 552 F. Supp. 131 (D.D.C. 1982).
256
Working Paper VI
Under such an agreement, the antitrust agency of the country that believes its companies are being closed out of another country as a result of private restraints makes a preliminary determination that there are reasonable grounds for an investigation of the matter, perhaps under its own law but, in any event, under the law of the country in which the restraint operates. It then refers the matter, along with its preliminary analysis, to the competition authority in the country whose home market is directly affected. That authority conducts an investigation, then reports back to, and consults with, the referring country as to the nature of its investigation, its findings, and any remedy it is considering. The referring country can accept these conclusions, seek to modify them, or subsequently conduct its own investigation and take actions that it finds appropriate. This positive comity approach has its roots in a 1967 Council Recommendation of the O.E.C.D. on Cooperation between Member Countries on Restrictive Business Practices Affecting International Trade. The term 'positive comity' itself was coined in the 1991 cooperation agreement between the U.S. and the E.U.12 Positive comity is also a prominent feature of the 1995 agreement between the U.S. and Canada.13 We are currently negotiating a new positive comity agreement with the E.U. to clarify the situations that would presumptively call for referrals, and tofleshout the report-back and consultation mechanisms that would come into play once a referral has been made. The E.U. has sought public comment on a draft of the agreement. This approach has several advantages. First, competition authorities tend to have the greatest stake in taking such complaints seriously, even if they involve foreign access. Secondly, such a process facilitates the gathering of evidence for such cases. In the absence of a treaty, the limits of jurisdictional reach for all countries imply a limitation of their ability to gather evidence on another's territory, which thwarts effective investigations. Moreover, there is little chance that country A would use its law enforcement powers to provide evidence to country B in a market access case in which country A thinks country B lacks jurisdiction. Finally, the positive comity approach increases the pressure throughout the world to allow competition authorities to conduct their work fairly, since it enhances the likelihood that these kinds of cases can defuse trade 12
Agreement Between the Government of the United States of America And the Commission of the European Communities Regarding the Application of Their Competition Laws, 23 Sept. 1991. 13 Agreement Between the Government of the United States of America And the Government of Canada Regarding the Application of Their Competition And Deceptive Marketing Practices Laws, 1 Aug. 1995.
4 - Competition Policy Objectives
257
tensions by providing a sensible, systematic approach to fact-gathering, reporting and bilateral consultation among competition authorities. Two recent cases illustrate how positive comity could function. In December 1997, the D.O.J. announced that it had closed an investigation into the way that AC Nielsen, a large U.S. firm, contracted its services for tracking retail sales.14 The U.S. investigation was closed because Nielsen entered into formal undertakings with the European Commission following a DGIV investigation that would alleviate any competitive concerns. The U.S. investigation had focused on whether Nielsen offered customers more favourable terms in countries where Nielsen had market power if those customers also used Nielsen in countries where it faced significant competition. These contracting practices occurred mostly outside the U.S., but they may have had an adverse effect on U.S. export commerce by preventing exports by Nielsen's U.S. competitors. Since most of the conduct occurred in Europe and had its greatest impact there, the European Commission had an obvious interest in dealing with the problem. Although no positive comity requests as such were made, the respective investigative staffs were in frequent contact, and it became clear to the U.S. staff that the Commission would effectively remedy the situation. In April 1997, the D.O.J. made its first formal positive comity request to the E.U. under the 1991 agreement. The D.O.J. asked DG IV to investigate possible anticompetitive conduct by certain European airlines that may be preventing U.S.-based airline computer reservation systems from competing effectively in certain European countries. It became clear that the European Commission was in the best position to investigate this matter, because the alleged conduct occurred in Europe and principally European consumers would be harmed if competition has been diminished. Accordingly, we concluded that this was an appropriate case in which to use our positive comity mechanism. I understand that DG IV's investigation is continuing. Positive comity should not be a controversial concept. An agreement on positive comity would not change U.S. or foreign law, and would not by itself permit the exchange of confidential documents and testimony. It also respects the sovereignty of participating countries, since it recognises that the country whose market is most immediately affected has the principal responsibility for enforcement. Our proposed positive comity agreement with the E.U. provides an excellent opportunity to demonstrate the value of this approach. In the long run, positive comity should occupy an important place on the international 14
See Department of Justice Press Release, Justice Department Closes Investigation into the Way AC Nielsen Co. Contracts its Services for Tracking Retail Sales, 3 Dec. 1996.
258
Working Paper VI
antitrust enforcement agenda, as it offers a real opportunity to deal effectively with antitrust 'market access' cases on a principled, cooperative basis. Nonetheless, positive comity has its limitations. Many competition authorities currently lack the independence, if not the will, to do a proper job. Until that situation changes, positive comity referrals are not a satisfactory solution to the problem of market access that is blocked by private anticompetitive restraints. Moreover, different countries have different substantive law in the competition area. Thus, there is a risk of disparate enforcement. This problem is likely to diminish, though not disappear, through cooperative arrangements. Moreover, the differences are not drastic. A focus on consumer welfare adds a measure of consistency in the competition area that is likely to grow over time.
3. The Connection Between Cooperation on Cartel Enforcement and Positive Comity The ability of competition authorities to cooperate on cartel enforcement will be greatly diminished unless we simultaneously begin to develop positive comity arrangements. This is so because the greatest impediment to cooperation is the fear, or at least the suspicion, that the evidence will be used for trade purposes. Trust among countries, and even among their competition authorities, in this area is not high. Consequently, unless countries can be assured that information is being used only for purposes of international cartel enforcement, and not for trade-related purposes, overall cooperation will diminish and cartel enforcement, as well as market access enforcement, will suffer. This would be a serious loss.
C. The Trade and Competition Agenda of the W.T.O. Anything done at the W.T.O. should not jeopardise, or even detract from, the ongoing efforts described above, which involve bilateral undertakings between countries that have well-established commitments to, and experience in, competition matters. Such efforts are most likely to lead to enduring agreements that can subsequently become a template for other countries. A hasty effort to negotiate rules at the W.T.O. is fraught with risk. First, it will be difficult to reach agreement in the W.T.O. on sound competition rules, which depend substantially on the strict application of neutral legal and economic principles. Even the relatively like-minded Member States of the European Union needed seventeen years to reach agreement on the Merger
4 - Competition Policy Objectives
259
Control Regulation. Indeed, the significant differences in the approach of the U.S. and the E.U. with respect to vertical restraints (which are the result, in part, of different histories and market structures) indicate that in some non-cartel contexts, agreement on common substantive principles may be very difficult. A W.T.O. competition policy debate will have to balance many often diverse national interests, with the possibility of positions shifting in response to tradeoffs in other trade negotiations related to agriculture, services, intellectual property, or any of the myriadfieldscurrently covered by W.T.O. agreements. Secondly, we must guard against a lowest-common-denominator outcome in the development of competition rules by the W.T.O. Efforts to achieve a 'minimum' set of competition principles, or to identify common substantive standards, could legitimate weak and ineffective rules, which would not serve the goals of trade liberalisation. Minimum standards often become the maximum. Thirdly, although a universal commitment to the adoption and enforcement of competition laws, and cooperation in antitrust enforcement, are worthy goals, they go beyond core W.T.O. concerns. This is evident when we consider the problem of how the W.T.O. would even identify, much less devise remedies for, violations of multilateral competition obligations. Over seventy countries, accounting for over 80 per cent of the world's GNP, already have enacted competition laws. Most of these laws would likely meet the requirements of any minimum substantive rules the W.T.O. could adopt. Moreover, competition law enforcement is often fact-intensive, but no government has proposed turning over to a W.T.O. body the kinds of confidential business information typically required for a proper competition analysis in particular cases. Such a process is not likely to be acceptable on a worldwide basis for many years to come. The problem of dispute settlement highlights the difference between competition law and other areas covered by the W.T.O. W.T.O. Members hold differing views on the objectives of competition law and the supporting analysis. Given this diversity, the use of dispute resolution with respect to a general requirement that Members adopt and enforce antitrust laws, and consider requests to investigate from other Members, is likely to have little impact on trade liberalisation. On the contrary, it could give procedural legitimacy to harmful actions masquerading as competition policy. Moreover, if dispute settlement were extended to individual decisions taken by domestic competition authorities, this would interfere with national sovereignty concerning prosecutorial discretion and judicial decision-making, and could also involve W.T.O. panels in inappropriate reviews of the credibility of witnesses and of case-specific, highly confidential business information.
260
Working Paper VI
For these and other reasons, the United States remains 'cautious' about a W.T.O. agenda on competition policy. It might make sense for the W.T.O. to study what is occurring elsewhere and to analyse the significance of those developments, but only if such work is not seen as a precursor to negotiations in the W.T.O. on competition policy. This cautious approach is embodied in the Singapore Ministerial Declaration. The W.T.O. Members that make up the Working Group will determine how it will proceed. However, the strong recommendation of the U.S. will be that the group adopt a work program that: encourages participation by competition experts from all Members; fosters among Members a common understanding of the relationship of competition matters to the W.T.O. framework; and is neutral regarding any conclusions that may be reached at the end of the group's two-year life. Any such programme should be based on sound, realistic judgements about what is reasonably feasible and should not jeopardise the other work that is currently ongoing among antitrust enforcement agencies in other fora, such as N.A.F.T.A., A.P.E.C, F.T.A.A., and O.E.C.D.
Conclusion A busy and important agenda in international antitrust enforcement is currently underway. This agenda is relevant in part to trade issues, but its concerns are substantially broader than such issues. It can be accomplished in the reasonably near future among key trading partners for whom there already exists a broad but untapped policy consensus. These ongoing efforts should not be deferred while awaiting developments in the W.T.O. work programme. Abstract discussions can be useful, even important; but improved international antitrust law enforcement in the present is absolutely necessary in an increasingly globalised economy.
VII Mitsuo Matsushita Professor Seikei University Tokyo,Japan Member of the Appellate Body World Trade Organisation Geneva, Switzerland The W.T.O. is not the only forum in which issues of international competition policy are discussed. They have also been discussed and explored in the O.E.C.D. and U.N.C.T.A.D. There are advantages and disadvantages of discussing competition policy issues in the framework of the W.T.O. ; however, this paper takes the position that competition policy should be included within the W.T.O. framework. A. The W.T.O. and Competition Policy The W.T.O. consists of a series of agreements covering trade in goods, trade in services and trade related intellectual property rights. The basic aim of the W.T.O. agreements is to achieve liberal and fair trade, but not to interfere with domestic policies of Members that have no bearing on trade. To this end, some basic principles are established, such as most-favoured-nation treatment, national treatment, transparency, prohibition of quantitative restrictions and other related matters. Under these agreements, the Members are prohibited from imposing trade restrictions and from discrimination. One feature of competition policy in a broad sense is to remove governmental restrictions that tend to impede competition. In this respect, the W.T.O. and competition policy share the objective of pursuing economic activities without undue hindrance by governmental measures. As in the European Union, as trade liberalisation progresses among W.T.O. members, it is necessary to control private restrictions that offset the effect of liberalisation achieved through the W.T.O. agreements. Thus, competition policy should be introduced in some form into the W.T.O. framework. As tariff barriers and quantitative restrictions are reduced, certain domestic regulatory policies have had a greater impact on trade. For instance, a domestic policy to allow depression cartels is likely to have the effect of protecting domestic producers and reducing imports.
262
Working Paper VII
In entering the W.T.O., the Members have relinquished sovereign powers and agreed to be bound by the agreements, as is reflected in the wording of the agreements. Thus, there is a subtle balance between the power of the W.T.O. to control the domestic policies of Members, and the power of the Members to control their own domestic policies that do not impact negatively on international trade. The W.T.O. incorporates a dispute settlement system, under which a violation of one of the agreements will be met with a recommendation to come into conformity with the agreement. A failure to comply with such a recommendation may lead to retaliation by the complaining member. The dispute settlement mechanism distinguishes the W.T.O. from O.E.C.D. and U.N.C.T.A.D. Thus, any agreement on competition policy should be cautiously conceived.
B. Diverse Objectives of Competition Policy Competition policy is a product not only of political and economic theory, but also of the political, economic and social reality of the country in which it operates. A nation's competition policy and law reflect the legal and administrative structure and tradition of that country. For this reason, the objectives of competition policy differ among nations. Some of the diverse objectives are the following. • Market economy—it is generally accepted that preservation of the market mechanism is the most fundamental objective of competition policy. • Efficiency—promotion of efficiency is also a common objective of competition policy. The Chicago School emphasises the importance of the efficiency goal. The Japanese Fair Trade Commission (J.F.T.C.) has published merger guidelines which include a provision that efficiency is an important factor in determining whether a merger promotes competition. Such a provision is likely to bring about a flexible enforcement policy that follows the rule of reason. Such an approach can also be followed with respect to vertical non-price restraints, which emphasises the importance of interbrand competition. • Pluralism—in the United States, antitrust policy originated following the movement of farmers and workers against large trusts. Currently, this philosophy is maintained by the Harvard School, which is suspicious of concentrations of political or economic power. The Harvard School holds that the goal of competition policy is not only to promote efficiency, but also to maintain a pluralistic society with no predominant powers. This concept is related to U.S. Constitutional notions of federalism, separation of powers, and checks and balances, all of which are designed to divide powers in order to prevent the
4 - Competition Policy Objectives
263
accumulation of oppressive power in society. A competition policy based on this concept takes a strict attitude toward the existence of large enterprises, and mergers that are instrumental in creating predominant enterprises. • Equity in transactions—in Japan, equity in transactions is regarded as a goal of competition policy. The Japanese Antimonopoly Act (A.M.A.) provides that an abuse of bargaining power by one party to a transaction against another is unlawful. For instance, it would violate this provision for a large manufacturer of automobiles to impose harsh conditions (such as unreasonable delay in payment or unreasonable pressure to lower its prices) in a contract with a small subcontractor that supplies parts and components. This concept is important in Japan, where vertically integrated business systems are an important feature of the market. Under this system, subcontractors or dealers would generally have less freedom to switch to other manufacturers than would their counterparts in the U.S. Thus, the lack of horizontal mobility makes the victims of such abuses captives. This provision arguably does not relate to competition policy. However, this illustrates how competition policy and law reflect the realities of the country in which they are implemented. • Market integration—market integration has been the primary goal of E.U. competition policy. Accordingly, vertical territorial allocation of markets is deemed per se illegal. Since market integration is the ultimate goal of the W.T.O. system, the E.U. experiment is important. • Transition to market economy—a number of countries are moving from planned economies to market economies. Competition policy plays an essential role in the transition process. However, during the transition period, competition policy must be sensitive to the specific needs which arise. For instance, in the People's Republic of China, a series of laws and regulations have been enacted that constitute a mixture of competition law, intellectual property law, deregulation law and civil law, which are very different by nature. A civil law system must be created before a competition law system can take effect. Accordingly, if competition policy is incorporated in the W.T.O. framework, its scope must be sufficiently broad that some of the above objectives can be accommodated. These goals often conflict with each other, making it difficult to include all of them within the W.T.O. framework.
C. Utilising the Existing W.T.O. Agreements However desirable it may be to introduce an agreement on competition policy within the W.T.O. framework, it will be a long process, and it is not certain that an agreement will ultimately be reached. Therefore, it is useful to review the
264
Working Paper VII
competition provisions of the existing W.T.O. agreements to determine how they can be used to promote competition policy. Although these provisions are scattered throughout the various W.T.O. agreements, and lack cohesiveness, they should nonetheless be stringently enforced. Some of these provisions are the following. • Agreement on Technical Barriers to Trade (T.B.T. Agreement)1—Art. 3.4 states: 'Members shall not take measures which . . . encourage . . . nongovernmental bodies within their territories to act in a manner inconsistent with the provisions of Art. 2.' Art. 2 provides for national treatment. Accordingly, it would be a violation for a government to encourage private standard control bodies to discriminate against foreign products, or intentionally to tolerate discriminatory practices by a standard body. • Agreement on Trade Related Investment Measures (T.R.I.M.s Agreement)2—Art. 9 states that a review shall be made within five years after the Agreement enters into force to determine whether it should be complemented with a competition provision. • Agreement on Antidumping Subsidies and Countervailing Duties3—Art. 3.5 requires that in order to impose an antidumping duty on a foreign product, the national authority responsible for administering antidumping law must establish a causal link between the dumped product and injury to a domestic industry. In determining causation, the authority must take into consideration trade-restrictive practices of, and competition between, foreign and domestic producers. For instance, an authority must consider whether an import cartel in its own country pressures foreign exporters to lower export prices, thereby causing it to dump; or whether an export cartel in its own country has reached an agreement to drive out competitors in the importing country. Taking such factors into consideration should help the importing authority to enforce competition policy. Art. 15.5 of the S.C.M. agreement contains similar language to Art. 3.5 of the Antidumping agreement. 1
Agreement on Technical Barriers to Trade, 15 Apr. 1994; Agreement Establishing the World Trade Organisation, Annex 1A; Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, Marrakesh, 15 Apr. 1994 in the G.A.T.T. Secretariat: The Results of the Uruguay Round of Multilateral Trade Negotiations, The Legal Texts (Geneva, 1995), at 138-162. 2 Agreement on Trade-Related Aspects of Investment Measures, supra, note 1, at 163-167. 3 Agreement on Implementation of Article VI of the General Agreement on Tariffs and Trade 1994 (Antidumping Agreement), supra, note 1, pp. 168-196; Agreement on Subsidies and Countervailing Measures (SCM Agreement), supra, note 1, at 264—314.
4 - Competition Policy Objectives
265
• Safeguard Agreement4—Art. 11.1 prohibits members from entering a voluntary export restraint agreement. Art. 11.3 states 'Members shall not encourage or support the adoption or maintenance by public and private enterprises of non-governmental measures equivalent to those referred to in paragraph 1. For instance, Art. 11.1 would be violated if the government of a member encourages exporters to create a cartel and limit exports of a product. Such an export cartel could be challenged by the competition authority of either the exporting country, the importing country, or both. • General Agreement on Trade in Services (G.A.T.S.)5—Art. 9.1 requires members to ensure that a monopoly supplier of services operating within its territory not use the monopoly in a manner inconsistent with Art. 2 (the Most Favoured Nation provision), and specific commitments. • Agreement on Trade-Related Aspects of Intellectual Property Rights (T.R.I.P.s)6—Art. 40 provides that members may enact legislation to control anticompetitive practices in contractual licences. D. International Competition Policy 1. Comprehensive Code Approach The Havana Charter,7 which never took effect, contained a comprehensive competition code which was to be enforced internationally. The framers of this charter believed that competition policy was an integral part of the liberal trade order that it was designed to establish. The Charter's competition code has served as a model for discussion of an international competition code in the current timeframe. A group of scholars known as the Munich Group prepared a draft international antitrust agreement, which covered a broad range of competition issues, including cartels, boycotts, vertical restraints, mergers and acquisitions, as well as enforcement and related matters. 4
Agreement on Safeguards, supra, note 1, at 315-324. General Agreement on Trade in Services, 15 Apr. 1994; Agreement Establishing the World Trade Organisation, Annex IB; Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, Marrakesh, 15 Apr. 1994, supra. note 1, at 325-364. 6 Agreement on Trade-Related Aspects of Intellectual Property Rights, 15 Apr. 1994, Art. 40.2; Agreement Establishing the World Trade Organisation, Annex 1C; Final Act Embodying the Results of the Uruguay Round of Multilateral Trade Negotiations, Marrakesh, 15 Apr. 1994, supra, note 1, at 365-403. 7 United Nations Conference on Trade and Employment, Havana Charter for an International Trade Organization, Final Act and Related Documents (1948). 5
266
Working Paper VII
2. Double Track Approach A group of experts commissioned by the E.U. prepared a report presenting a design for introducing competition policy in the W.T.O. The report recommends that more bilateral agreements should be entered. It also suggests that an agreement within the W.T.O. framework should be sought, and some core principles should be adopted, including the prohibition of cartels and boycotts. With regard to non-price vertical restraints, it suggests that due process issues, such as disregard of clear evidence, should be addressed. The experts report thus recommends a two track approach—i.e. simultaneous pursuit of bilateral agreements and plurilateral agreements.
3. Evaluation of the Two Approaches Accordingly, two approaches have been suggested: the comprehensive code approach, such as that suggested by the Munich group, and an incremental approach, such as that suggested by the E.U. expert group. Both have advantages and disadvantages. The comprehensive code approach has the advantage that it covers a wide area, which would have the effect of promoting convergence of domestic policies of those subscribing to the code. However, this approach has the disadvantage that at present, it may be premature, which would make it difficult to rally political support for such an approach. The T.R.I.P.s agreement, discussed above, represents a comprehensive code approach. However, this agreement had a supportive constituency during the Uruguay Round Negotiations, which included the U.S., the E.U., and Japan, as well as a number of multilateral corporations. It is unlikely that a similar supportive constituency would exist with respect to a comprehensive competition code. Moreover, T.R.I.P.s was adopted through the process of the Uruguay Round negotiation, in which many issues were discussed simultaneously. Some developing countries regarded certain provisions of T.R.I.P.s to be adverse to their interests, but were willing to accept them as part of the W.T.O. framework package. Such conditions would not exist with respect to a comprehensive competition code. Accordingly, the W.T.O. should proceed cautiously in introducing competition policy.
4 - Competition Policy Objectives
267
4. Short-Term Goal If the W.T.O. were to establish a competition policy scheme which, from the outset, affected domestic policies that have only an indirect impact on international trade, Members would be likely to react negatively. Accordingly, the initial step should be limited to establishing a set of competition principles that directly relate to international trade. Examples of private restraints that have a direct impact on international trade are: export and import cartels; boycotts which exclude imported products and foreign enterprises; certain vertical restraints, such as exclusive dealing, customer restrictions, and tie-ins, which specifically exclude foreign products; and predatory pricing by which domestic enterprises attempt to block entry of foreign products into their domestic market. Such restraints are an immediate concern to the W.T.O. because they offset the effect of the existing W.T.O. agreements. Accordingly, a general prohibition of cartels may be premature. In some countries, some types of cartels are exempted from the competition laws. For instance, in Japan, small business associations are allowed. Other countries exempt other types of cartels, such as depression cartels. The W.T.O. would be justified in addressing such cartels only if they have a direct impact on international trade. If they do not, the W.T.O. should proceed with caution in addressing them. Moreover, certain hard core violations of competition law that do not affect international trade may be more appropriately handled by national governments in the current timeframe. Examples of such violations are minimum pricefixingcartels which affect the domestic economy, and resale price maintenance contracts that keep prices from going down.
5. Long-Term Goal Once the process of globalisation has progressed further, and domestic and international trade policies are no longer distinguishable, a more ambitious approach can be pursued. National markets will then be more integrated, such that domestic policies of each W.T.O. member will have a much greater impact on the markets of other members. At this stage, the differences in competition policies of the various Members will become the main problem, because they will create incompatibilities of systems and cause frictions. Harmonisation of competition policies will then be necessary. The prohibition of cartels, boycotts and resale price maintenance will constitute the core competition rules to be included in the W.T.O. framework,
268
Working Paper VII
irrespective of whether they directly affect international trade. Members should implement such rules by enforcing domestic laws in such a way that these practices are prohibited. This will have the effect of establishing national markets free from cartels and resale price maintenance, which will have the effect of promoting efficiency within the W.T.O. system. The issues are more complex with respect to non-price vertical restraints, as policies and laws vary from nation to nation in this area. In most nations, including the U.S. and Japan, such practices are dealt with under the rule of reason rather than the per se rule. In the E.U., however, vertical territorial restrictions according to the geographic market of each Member State are strictly prohibited, because they tend to compartmentalise national markets, thereby obstructing achievement of the goal of market integration. Given these differences, an international competition agreement should be limited to general principles in this area. For instance, it could provide that a vertical territorial or customer restraint is harmful if it is exercised by an enterprise with market power, and thus there is no interbrand competition in the market in question.
E. Optional Membership in Agreement on Competition Principles At present, more than 130 nations are Members of the W.T.O., many of which are not yet ready to adopt a competition law. Recent discussions on the relationship between trade and competition have focused on issues such as market access and abusive use of trade measures, such as antidumping law. The market access issue relates primarily to Japan; the antidumping issue relates primarily to the U.S. and the E.U. These issues do not necessarily concern all W.T.O. Members. Therefore, it seems that participation in a W.T.O. agreement on trade and competition should be optional for the Members. Technically, this would mean that a competition agreement would be an Annex IV agreement, in which Members have the option to participate.
F. Developing Countries Most Members of the W.T.O. are developing countries, at varying stages of development. For them, some of the issues of competition policy differ from those of the industrialised countries. Until now, discussions on competition policy in the W.T.O. may have neglected the issues that are important in the context of North-South relationships.
4 - Competition Policy Objectives
269
The U.N.C.T.A.D. draft code of conduct of enterprises and draft code on restrictive business practices were not adopted since they placed too much emphasis on protection of developing countries vis-a-vis industrialised countries. Although this should not be repeated in the W.T.O. context, it is nonetheless important to consider the issues involved in the North-South relationship.
VIII Petros C. Mavroidis Professor, University of Neuchatel Neuchatel, Switzerland
Marc Bacchetta W.T.O. Secretariat Geneva, Switzerland
Henrik Horn W.T.O. Secretariat Geneva, Switzerland Professor, Stockholm University Stockholm, Sweden1
Introduction The case for negotiating multilateral competition rules, and the scope of any such rules, has recently become the subject of intensified debate.2 A number of prominent policy-makers and academics have called for such efforts.3 For 1
The paper represents the views of the authors and is not meant to represent the position of the organisations for which they work. 2 Competition policy in this paper is used in the conventional antitrust sense, i.e. disciplines on anti-competitive practices such as collusion between enterprises, the abuse of dominant market positions or their creation through mergers and acquisitions. 3 See, e.g., Sir Leon Brittan, A Framework for International Competition, Address to the Davos Symposium (1992); European Commission, Competition Policy in the New Trade Order: Strengthening International Cooperation and Rules, Report of the Group of Experts (1995) (hereinafter Van Miert Report); Claus Dieter Ehlermann, The Role of Competition Policy in a Global Economy, unpublished mimeo (1994); Alexis Jacquemin, The International Dimension of European Competition Policy, 31 J. Common Mkt. Stud. 91 (1993); Ernst-Ulrich Petersmann, Proposals for Negotiating International Competition Rules in the G.A. T. T.IW. T. O. World Trade and Legal System, 49 AuBenwirtschaft 231(1994); Frederic M. Scherer, Competition Policies for an Integrated World Economy (1994).
272
Working Paper VIII
instance, the E.U. has recently published a report prepared by a Group of Experts which supports an initiative of this nature.4 The views expressed in this report substantially reproduce opinions already expressed in the literature. Much of the literature calling for multilateral action argues that competition policy issues become increasingly international because of the globalisation of many industries, the unfair advantages for firms that benefit from lax enforcement of domestic antitrust laws, the problems created by extraterritorial application of laws, and the risk that developing countries may be particularly harmed by anticompetitive practices. Proponents of common competition rules have repeatedly stressed the need to take action in thisfieldat the World Trade Organisation (W.T.O.) level. Trade and competition thusfiguresamong the 'new' topics to be discussed in the near future by the trading partners alongside trade and environment, and trade and investment. Indeed, the W.T.O. Ministerial Meeting in Singapore 1996 established a Working Party to discuss this issue. From the perspective of economic theory, how competition policy should be pursued in an international economy is a special case of the more general question addressed in the theory of regulation and delegation. Regulation theory deals with governmental intervention in situations where, because of market failures, private decisions do not contribute to the maximisation of national welfare. The theory of regulation, which would most directly be applicable to the relationship between the national competition authorities and firms, identifies at least three broad categories of reasons for government regulation: externalities between firms, monopoly power and informational problems. However, when considering proposals for international agreements on competition policy, the issue is not only one of regulation, but also one of delegation of regulatory powers from national authorities to an international body. This question has been addressed by Bhagwati in connection with the broadening of the international policy agenda,5 as well as by Gatsios and Seabright6 and Neven7 when discussing regulatory reforms in the E.U. These authors have
4
Van Miert Report, supra note 3. Jagdish Bhagwati, Fair Trade, Reciprocity and Harmonization: The Novel Challenge to the Theory and Policy of Free Trade, Paper presented to the Conference on Analytical and Negotiating Issues in the Global Trading System, University of Michigan (1991). 6 Konstantine Gatsios and Paul Seabright, Regulation in the European Community, 5 Oxford Rev. Econ. Pol. 37 (1990). 7 Damien J. Neven, Regulatory Reform in the European Community, 82 Am. Econ. Rev. 98 (1992). 5
4 - Competition Policy Objectives
ITS
attempted to identify the policies that should be subject to international negotiations or assigned to higher levels of government, identifying international policy spillovers as the key issue in this context. This paper partially follows these approaches in considering the extent to which negative spillovers motivate an international agreement on competition policy. It does not, however, make a thorough regulation theory analysis. Rather, the economic analysis is focused on the source and nature of the spillovers that would possibly justify regulation at the international level. We thus neglect possible adverse aspects of regulation at the international level. The paper distinguishes between 'spillovers' and 'distortions' from national competition policy. Competition policy decisions almost always benefit some agents while harming others. This is also true in an international context. What is special to the latter situation, however, is that the gainers may reside in one country, and the losers in another. In such a situation, there is a negative spillover on the latter agents from the decision, relative to some other policy choice. However, there is not necessarily any inefficiency associated with such a decision from an international perspective, except that winners do not compensate losers, which in any event occurs rarely in the competition policy context. For there to be an inefficiency at international level, there is at least one additional requirement: that a country take actions that are optimal from its own point of view, but inferior for the world as a whole in comparison to some other feasible actions. In such a situation, there is a 'distortion' associated with the national competition policy decision. The importance of this distinction between spillovers and distortions is that spillovers alone do not provide an economic rationale for countries to form a multilateral competition policy agreement, since they are not necessarily associated with any inefficiencies from an international point of view. It is only those spillovers that give rise to distortions that motivate international agreement. Accordingly, a fundamental question that proponents of a multilateral agreement on competition policy should address is whether the negative spillovers from national competition policies also result in distortions. This paper will begin with an investigation of the extent to which there are economic incentives for countries to pursue national competition policies that give rise to distortions for other countries. If these distortions do not exist, or are likely to be small, the case for a multilateral agreement on competition policy seems rather weak. However, in order to remain manageable, this analysis disregards the legal remedies against such beggar-thy-neighbour competition policies. Section B considers the extent to which policies of this type can be addressed under the existing legal framework. Section C argues that, in principle, current trade law could substantially resolve these problems. Finally, Section D draws
274
Working Paper VIII
some conclusions for the question that motivated the analysis: do negative spillovers provide a case for a multilateral agreement on competition policy? We believe that an economic-legal analysis of this type is necessary, but not sufficient, to support the claim that there should be a multilateral agreement on competition policies. It is necessary, because an economic case for an agreement can only be made if basic principles of economics support such an agreement. However, it is not sufficient, since there is a discrepancy between the economic and legal analysis that must be resolved before any particular solution is advocated, if the legal regime can in principle address the problems suggested in the economic analysis. In particular, the economic analysis is based on unrealistic assumptions about how the legal system functions, and the distortions it points to are empirically not apparent, in which case there is little need for a multilateral agreement. Alternatively, the legal system cannot in practice address the problems, in which case the externalities may still be a real-world concern, which would remain to be verified empirically. In this case, the proposed solution should take into consideration the weaknesses of the current legal regime, in order not to duplicate these weaknesses. A. Economic Incentives for Beggar-Thy-Neighbour Competition Policies When markets are imperfectly competitive, there is a general case for government competition policy intervention. The standard approach in economics is to view the purpose of competition policy, like that of any other policy, as the maximisation of 'welfare'. Under a number of stringent simplifying assumptions (for instance, that the policy under consideration affects only a small part of the economy), the maximisation of welfare can be seen as equivalent to the maximisation of the sum of consumer surplus and producer profits. Consumer surplus is a measure of 'consumer profits', in that it measures the difference between consumers' willingness to pay for their consumption and what they actually pay. The essential point for the present discussion is that the welfare criterion weighs the interests of consumers and producers. The precise weights are not material. It is important that profits of nationally owned firms are viewed as a national good, alongside consumer surplus. The implicit assumption underlying the use of the sum of producer and consumer surplus as the objective for the competition authority is that the authority should not be concerned with distributional aspects of its interventions. This could be motivated from a welfare point of view, even if society had preferences with respect to the distribution of the total surplus. For instance, such preferences could be apparent from the existence of redistributional tax
4 - Competition Policy Objectives
275
schemes or from consumers' ownership of significant parts of industry, through pension funds. While neglect of the distributional aspects in the pursuit of competition policy may be acceptable in a closed economy, it has an additional dimension when the policy is pursued in an open economy, given the lack of explicit international redistribution schemes. However, it may be accepted in an open economy if the countries under consideration are relatively symmetric, in the sense that they gain in some sectors but lose in others. Under such circumstances, there is a common interest in maximising global welfare, even if not for altruistic reasons. From a theoretical perspective, it is almost impossible to imagine situations where the choice of competition policy in one country does not affect other countries. For instance, if there is perfect competition in all sectors except one non-tradable sector in one country, there is no foreign direct investment, and there are no imported inputs, then there is no role for competition policy to play in any other sector. It may, at first glance, appear that competition policy pursued vis-a-vis the one sector will only be of interest to the economy which harbours this sector. However, this policy will affect all the trading partners of this economy, albeit more or less directly. For instance, by affecting the conditions under which firms compete in this sector, competition policy will affect the sector's demand for factors of production. This will affect the prices of factors of production used in this sector, which will affect other sectors of the economy, and consequently also trading partners. Hence, at a very general level, nationally pursued competition policies are bound to affect foreign countries. However, these indirect effects in many, if not most, cases would be very small. Competition policy is typically concerned with more direct effects arising in the sector where intervention is considered. We will also restrict our attention to such effects.
1. Negative Spillovers and Distortions from Nationally Pursued Competition Policies The typical analysis of optimal competition policy in the context of closed economies—which is the context within which almost all economic analyses have been performed—seeks to establish the optimal trade-off between the interests of consumers and producers. The optimal policy almost invariably results in gains to some agents and losses to others. This will also be true when the policy is pursued in an open economy context. For instance, a country's policy choice may, on balance, reduce the welfare of another country, relative
276
Working Paper VIII
to some other policy choice. In such a case, the policy gives rise to a 'negative spillover' for the other country. The existence of a negative spillover does not necessarily signal the presence of an inefficiency from a global point of view. Just as a welfare maximising competition policy in a closed economy produces some gainers and some losers, competition policy in an international economy does the same, even if the policy maximises global welfare. It is then possible that gainers reside in the country pursuing the policy, while losers reside in other countries, even if the policy seeks to maximise global welfare. National policies, including national competition policy, should generally maximise national, rather than global, welfare. This has important implications, since competition policy deals with imperfectly competitive markets. Specifically, the existence of market power among firms makes it possible and tempting for national governments to increase domestic welfare at the expense of foreign agents. This gives rise to a strong presumption that interventions undertaken by a national competition authority will differ from those undertaken by a worldwide competition authority. When a national competition authority chooses a different policy than would a worldwide authority, the policy gives rise to a 'distortion'. Distortions are thus a special case of negative spillovers. There are strong analogies between competition policies giving rise to distortions, and 'strategic trade policies', both with respect to the assumed setting (oligopolistic or monopolistic international markets), and with respect to the mechanisms at play. Indeed, the same types of market structures in which strategic trade policy may play a role, also make competition policy interventions attractive. With both types of policies, distortions are the deliberate consequence of policies that seek to improve national welfare, partly at the expense of other countries, rather than as accidental by-products. Moreover, as in the case of strategic trade policy, the interest behind strategic competition policy is not trade flows per se, but the possible negative impact on other countries in terms of loss of consumer and/or producer welfare. While government officials seem to be interested in trade effects, these are not of primary interest, from our point of view. As suggested above, there are two basic sources of distortion from national competition policy. First, such interventions affect, but do not take into account, the interests of foreign consumers. For instance, if twofirmsin country A propose to merge, the merger will be approved by the competition authority in country A if the positive impact of the merger on the merging firms' profits is considered to outweigh its possible negative impact on country A's consumers. In the evaluation of this proposal, the competition authority in
4 - Competition Policy Objectives
277
country A will not take into consideration possible negative consequences for consumers in country B. Similar conflicts may also arise in cases involving business practices that would legally fall under the heading of abuse of dominant position, and that are anti-competitive from an economic perspective. Other examples are coordinated behaviour between firms, such as export cartels, which may increase national welfare as long as they do not have pronounced negative effects on domestic consumers. Even when national competition policy seeks to maximise consumer surplus of domestic residents, and does not attach any weight to domestic producer surplus, it may still have negative repercussions on foreign consumers. For example, the merger discussed above may have beneficial consequences for consumers in country A, and thus be approved (even if producer surplus is not taken into account) despite its detrimental effects on foreign consumers. Such a case may arise when competition policy influences the choice of location of production. Consumers residing in the market where afirmlocates may benefit from lower prices due to lower trade costs, adaptation of products to local standards, etc. However, the more competition policy emphasises the maximisation of consumer welfare, and the less weight that is put on producer interests, the lower the concern for distortions. The second reason that nationally pursued competition policy may give rise to distortions is that the interests of foreign producers are not taken into account. For instance, a merger between two firms in country A may be approved in country A, even though, from a world point of view, it should not be approved because it causes severe harm to producers in other countries. Distortions from competition policy might arise in the home market of the country pursuing the policy, as well as in its export markets. For instance, the former would be the case when a vertical arrangement is permitted, which results in the foreclosure of foreign producers from the national market. The latter would be the case when the foreclosure is in foreign markets. These two situations are often treated differently from a legal point of view. In the former case, competition law may not enable the foreign competition authority to intervene against the foreclosure (unless the law is applied in an unreasonably extraterritorial manner). In contrast, in the latter case, the authority might be obliged to do so. However, from an economic point of view, it is irrelevant whether the negative impact on the foreign producers arises in the domestic or in the foreign country. It would still be a distortion that might provide a rationale for an international agreement.
278
Working Paper VIII
2. Distortions and the Gains from an International Agreement on Competition Policies This section considers whether negative spillovers may motivate a multilateral agreement on competition policy. Negative spillovers on other countries from national competition policies do not, as such, provide a rationale for an agreement. As stressed above, even a policy that maximises global welfare may have negative repercussions for agents residing in certain countries. In such a case, no inefficiency results, and it is not possible to improve on the global allocation of resources by choosing some other competition policy. There is, however, still a distributional dimension to the policy that may call for some international compensation. But, if countries are fairly symmetric, in the sense that they gain in some sectors and lose in other sectors, they can all share the gains of efficient policies. Negative spillover alone does not give rise to a situation where an international agreement affecting the choice of competition policy will improve world welfare by influencing resource allocation. A distortion must also be present. For instance, a distortion resulting from a country's choice of competition policy on a trading partner could be corrected through an agreement between these two countries, which could improve welfare for both. This agreement would stipulate a different choice of competition policy with a more favorable outcome for the partner country, which could more than compensate the policy active country for the lower welfare implied by the new choice. For the agreement to be feasible, there must be some way in which this compensation can take place. In practice, each country makes a large number of competition policy decisions with ramifications for other countries. This has two related consequences. First, a multilateral agreement may become beneficial to all countries, even absent some type of compensation scheme, since countries may be gainers in some sectors and losers in other sectors. The second, and more subtle consequence is that the interaction between the various distortions becomes of interest. The most straightforward case is where there is hardly any interaction, such as when countries' policy decisions concern different industries. In this situation, the logic from the case of a single distortion carries over directly. For instance, assume that two countries may allow cartels among their exporters, and exports occur in different industries. From the perspective of national welfare, it is preferable to allow cartels, regardless of what the other country chooses. However, the countries have a common interest in curbing export cartels globally, since the gains they experience from the cartelisation of
4 - Competition Policy Objectives
279
their own export industry are smaller than the losses imposed by the cartelisation of their imports. This is a simple example of a prisoners' dilemma situation applied to competition policy. The distinguishing feature of such a situation is that in the pursuit of the individual interest, decision-makers end up in a suboptimal position. If the countries could instead make a binding agreement not to allow export cartels, they would both be better off. However, in some situations, countries intervene in the same or in closely related markets. The question then arises as to the combined effects of these decisions. While each decision tends to cause distortions, the combined effect of these decisions is to magnify the distortions, which strengthens the argument for an international agreement. Conversely, distortions may counteract each other to the extent that there are no gains achievable through an international agreement. Hence, when evaluating the gains from a multilateral agreement, the cross-effects from the interactions between the competition policies must be considered, which is a daunting task.
3. Features of Competition and Trade Policy that may Affect Distortions This section will address whether the distortions resulting from national competition policies are important in reality. It will also consider some factors that may influence the size of these distortions. a. Competition Policy that Discriminates Across Countries Whether a particular policy amounts to discrimination is of central importance from the perspective of international trade law. Competition policy interventions involving distortions typically involve some form of discrimination. For instance, the 'national champion' merger, which is a merger that would not be permitted if it involved foreign firms, but is permitted when it involves domestic firms, clearly amounts to discrimination. However, the possibility that competition policy will induce distortions does not necessarily require discrimination in the sense of treating foreign agents differently from domestic agents. For instance, the competition authority may find other characteristics of foreign firms that enables the authority to extract surplus from them. For example, a national competition authority would prefer a merger between two domestic firms to one between a larger foreign firm and a domestic firm in an industry. However, it cannot explicitly discriminate against the foreign firm by disallowing a merger with this firm on nationality grounds. If the authority instead pursued a general policy of limiting the degree of concentration that is permitted to accumulate through mergers, and this
280
Working Paper VIII
policy is said to apply regardless of the nationalities of firms involved in possible mergers, it could set a threshold of concentration at a level such that the only possible merger is between the domestic firms. Until now, the distinction between distortions resulting from discriminatory and non-discriminatory national competition policies has not received attention in the theory literature. However, it may be of practical interest since nondiscriminatory beggar-thy-neighbour competition policies are presumably much harder to identify empirically, and thus to prevent through an international agreement. The distinction might also have important consequences for the appropriate design of an international agreement on competition policy. However, this is beyond the scope of the present paper. b. The Objective of Competition Policy We have conjectured above that the relative weight put on consumer and producer surplus in national competition policy may affect the size of the distortions associated with the policy. Unfortunately, little work has been done on questions concerning the importance of the objective function of competition policy for international distortions. Thus, we can only raise certain issues that merit further investigation. The following issues should be considered. First, where competition policy only serves to protect domestic consumer interests, will foreign firms not be treated differently from domesticfirms,even though they are not formally provided national treatment? Secondly, will extraterritorial application of national competition laws have more favourable welfare properties from a global point of view than those which result when weight is also attached to domestic producer interests (see below)? Thirdly, what are other possible objectives of competition policy such as promotion of domestic rather than national welfare? The national welfare criterion in standard economic analysis, which does not typically view profits generated by a foreign multinational in the domestic economy as part of national income, but does take into account the profits made by domestically owned firms in foreign countries, in many cases does not seem empirically very descriptive. c. Extraterritorial Application of National Competition Laws The basic distortions problem is due to the discrepancy between the coverage of the interests that are promoted by competition policy and the reach of its economic effects. In an attempt to solve this problem from a national point of view, competition policy in certain countries has extraterritorial reach. The existence of these policies has two conflicting effects on the potential gains from international agreements on competition policy. From an economic point of
4 - Competition Policy Objectives
281
view, extraterritoriality enlarges a country's available 'set of decisions'. On the one hand, the extraterritorial reach of countries' competition policies has positive implications from a global point of view, in that it permits countries to offset the discriminatory treatment of domestic interests by foreign competition policies (including lack of formal competition laws, or slack enforcement). This reduces the gains from an international agreement. 8 For instance, this would occur if country B prevented a 'national champions' type merger to be consummated between two firms in country A, i.e., if this merger were intended to shift profits from country B firms to the merged entity. On the other hand, the negative effects result because extraterritoriality allows countries to include foreign agents in the territorial reach of their competition policies, while their interests are not considered. Thus, the country asserting extraterritorial jurisdiction is able to extract more foreign consumer and producer surplus, resulting in new problems with distortions. Continuing with the example, the proposed merger may increase aggregate welfare of both country A and country B, but be prevented by country B because it reduces welfare there. It does not seem possible to draw any general conclusion as to whether extraterritorial application of competition laws increases or decreases distortions. d. Bilateral Agreements on Competition Policy The extent to which there are distortions from competition policies pursued by individual countries is also likely to be affected by the extent to which countries have bilateral agreements on comity. Such agreements would ideally mitigate distortions because they require both countries to take into account the interests of the other party. There are, however, some unresolved aspects of these agreements. First, it is not clear exactly what is meant by taking into account the interests of another country. Does it amount to giving equal weight to the interests of the two countries? If a certain business practice is beneficial for a large number of foreign consumers, but adversely affects the interest of a smaller number of domestic consumers, what would be required to satisfy comity requirements? Would the domestic competition authority be expected to accept the practice? Secondly, a distinction should be made between a situation where there are only a limited number of bilateral agreements, and one where all countries have such agreements with all other countries. In the former case, the bilateral agreements are likely to produce new distortions. For instance, assume that countries A and B have a bilateral agreement, and country A is to make a 8
This type of internalisation of distortions caused the postponement of the discussion on common competition rules. See discussion in Section B, infra.
282
Working Paper VIII
decision on a practice by a country B firm that exploits consumers in country C. Since the bilateral agreement enables country B to ask country A to take its interests into account, and its interests are in this case to exploit consumers in country C, the decision with respect to the practice is more likely to be an approval than it would be if no agreement existed. Hence, while the agreement might mitigate distortions between countries A and B, it may create distortions for country C. To the extent that B and C, and A and C, have common interests, there might be positive spillovers. If, instead, bilateral agreements had been entered into to cover all possible pairs of countries, and the same business agreement is under consideration, then, if both countries B and C ask A to take their respective interests into account, it is not clear how country A should do this. Accordingly, the existence of bilateral agreements may have important ramifications for the degree of distortions that is internalised. More research is needed, however, in order to determine their importance. e. Differences in Competition Policy Across Countries Distortions from competition policies does not presuppose differences between competition policy across countries in equilibrium.9 For instance, two symmetrical countries that have identical competition laws, and that both allow export cartels, would lead to distortions, but these distortions would not be affected by an agreement to harmonise. This is because the source of the problem is not differences across countries in the laws, but rather the design of the law. Hence, harmonisation of competition policies would not remove the problem with distortions, except trivially in the case where countries converge on a competition policy that prevents any type of beggar-thy-neighbour behaviour.l0 / Do Distortions Necessarily have Anti-competitive Effects? Another issue is whether distortions are inherently anticompetitive, so that they could be identified by looking at the competitive effect of competition 9
Competition policies must have the possibility of differing for the policy interaction to be of a prisoners' dilemma type. If they did not have this possibility, then countries would always be forced to be co-ordinated in their choice of policies. 10 There are other economically valid arguments for why an international agreement may be beneficial that depends on the existence of national differences in competition policy. One such argument is that procedural and legal costs are higher whenfirmshave to deal with different rules and procedures in each country. Compliance costs, for example, have become a source of concern for parties contemplating multi-jurisdictional mergers. However, since these issues are not related to the presence of distortions, they are not dealt with here.
4 - Competition Policy Objectives
283
policy interventions. In general, the appropriation of foreign consumer surplus only requires competition authorities to allow particular anticompetitive practices. A competition authority that seeks to reap foreign consumer surplus has no incentive to promote wasteful competition between domestic firms in foreign markets. As to shifting foreign firms' profits, in various cases, beggarthy-neighbour policies will require a permissive attitude from the competition authorities. There are also reasons that foreign profits can best be appropriated under a stringent competition policy. It is in the interest of a group of domesticfirmsto be allowed to employ anticompetitive practices vis-a-vis foreign firms. This may also be in the national interest, even if no foreign consumer surplus is exploited, as in the case of vertical arrangements that foreclose foreign producers. However, competition policies also have structural components, such as the long-run effects of merger regulations. This implies that in a longer-run analysis, one cannot take the group of firms as given. Rather, the possibilities of entry must be considered, which is affected in various ways by competition policy. Long run domestic industry profits may be larger with a stringent competition policy that allows for significant entry of domestic firms. Firms may have smaller profits in such a situation, but on the other hand there will be many domestic firms in the market. Hence, in the long run, foreign firms' profits will not necessarily be extracted through a slack competition policy. National policy in this case is procompetitive, but the policy would still result in a distortion for the foreign country, and would lead to an inefficient outcome for the world as a whole. g. Trade Liberalisation Finally, the relationship between trade liberalisation and the incentive for countries to pursue distortionary competition policies should be considered. It is sometimes claimed that multilateral trade liberalisation, in particular that pursued through the Uruguay Round, makes an international agreement on competition policies desirable. This claim may be interpreted in several ways. One interpretation is that as border protections have come down, other already existing barriers have become more visible. Thus, it is necessary to address the restraints caused by nationally pursued competition policies. Another interpretation, which would make the claim slightly stronger, is that it is necessary to remove the already existing competition policy barriers in order to reap the potential gains from trade liberalisation. Some commentators go further, however, arguing that trade liberalisation might induce countries to pursue policies with a greater beggar-thy-neighbour flavour than had the policies pursued prior to the liberalisation. This argument assumes, in line with the analysis in
284
Working Paper VIII
section A.I, that nationally pursued competition policies are purposely employed to exploit foreign countries, and that the incentives to do this increase as the level of trade liberalisation increases. A growing body of theoretical research exists on this matter.'' Matters are more complicated than they may first appear. From the perspective of economic theory, however, there is no reason to believe that trade liberalisation induces countries increasingly to pursue beggar-thy-neighbour policies. The claim that trade liberalisation fosters beggar-thy-neighbour competition policies appears to be based on the idea that when countries are constrained in their ability to use trade policies for such purposes, they will look for other policies through which they can achieve similar results. Competition policy is one such policy. However, it is the change in the incentives with regard to competition policy caused by trade liberalisation that is significant. This change might depend in part on whether the policies are substitutes. It will also be affected by other circumstances, such as properties of consumer preferences and production technologies, about which limited information is available. For instance, assume an industry in which the number of domesticfirms,and the degree of industrial concentration among domesticfirms,affects their level of profits when they compete with foreign producers. Assume further that the number of domestic firms is determined by the merger regulation. Before trade liberalisation, the domestic government would use tariffs to protect domestic firms; after liberalisation, this is no longer possible. The competition authority chooses the optimal number of firms, such that the benefit from the exit of an additional firm (i.e. the merger of twofirms)in terms of savings onfixedcosts, equals the cost of this exit in terms of consumer welfare. This is the standard economic 'marginal cost equals marginal benefit' condition for any optimal decision. The trade policy stance would affect this choice to the extent that it affects the balance between these costs and the benefits of allowing an additional merger. With a new trade policy, the optimal restrictiveness of the merger regulation will change, and the direction in which the policy must be changed depends on how the balance between the perceived gains and costs of
11
This subsection builds primarily on Henrik Horn and James Levinsohn, Merger Policies and Trade Liberalisation, (1997) (unpublished paper, available through the University of Michigan). Other relevant works are Martin Richardson, Trade and Competition Policies: Concordia Discors? (1996) (unpublished paper, available through the University of Otago); Marc Rysman, Competition Policy in Strategic Trade (1997) (unpublished paper, available through the University of Wisconsin). In all four papers, simple open-economy oligopoly models are employed to study how trade liberalisation affects the optimal degree of concentration among domestic firms.
4 - Competition Policy Objectives
285
an additional merger is affected by the trade policy. In technical terms, this will depend on curvatures of demand and cost functions. Thus, the conclusion is that there is no theoretical reason that trade liberalisation leads to increased use of distortionary competition policy. This does not preclude the possibility that theoretically valid arguments along these lines could be made. The claim is only that they do not exist to date, and they seem, a priori, rather difficult to construct.
B. Existing Legal Remedies Against Beggar-Thy-Neighbour Competition Policies This Section will examine the possibilities offered by the existing regulatory regime to address R.B.P.-related concerns at the multilateral level. It will consider both trade and competition law instruments that can help address the distortions mentioned in the preceding section. The essence of the argument is that, in principle, the existing regime could be employed to deal with distortions created by competition policies in many situations. There is, however, a discrepancy between the theoretical possibilities and actual practice, the rationale for which is examined in Section C. Subsection B.I will consider the extent to which the existing G.A.T.T./W.T.O. regime can address distortions. It will focus on the relevant G.A.T.T. Articles because the interpretation of those Articles through a series of G.A.T.T. panels will lead to some meaningful conclusions. Such an endeavour would not be possible with respect to the G.A.T.S., because none of the relevant G.A.T.S. articles has yet been interpreted by a panel. However, the focus on G.A.T.T. is not meant to imply that there is no link between trade and competition in the services context. To the contrary, a number of G.A.T.S. Articles, such as VIII and IX, explicitly refer to this link. At this stage, however, any general conclusions would be premature. Having investigated the possible use of the G.A.T.T./W.T.O., a possible competition regime will be considered.
1. The G.A.T.T./W.T.O. Option The available legal framework in the tradefieldprovides substantive rules. For instance, Art. Ill requires W.T.O. Members to observe national treatment when regulating. This is not the case of bilateral agreements on cooperation in the field of competition law, which typically aim at sensitising domestic competition authorities as to foreign interests. The most elaborate instrument in
286
Working Paper VIII
this direction is positive (as opposed to traditional) comity. Consequently, such bilateral agreements constitute mere procedural vehicles on which competition-related concerns can travel. However, they do not include substantive competition rules, in contrast to the multilateral trade regime. a. An Historical Perspective
Art. XXIX of the G. A.T.T. (entitled 'Relation of this Agreement to the Havana Charter') provides the institutional link with the history of the treatment of R.B.P.s in the multilateral trade context. Art. XXIX: 1 stipulates that 'contracting parties undertake to observe to the fullest extent of their executive authority the general principles of Chapters I to VI inclusive and of Chapter IX of the Havana Charter pending their acceptance of it in accordance with their constitutional procedures'. Chapter V of the Havana Charter12 (H.C.) is dedicated to R.B.P.s. It contains a potentially all-encompassing list of R.B.P.s that must be addressed by contracting parties because they 'have harmful effects on the expansion of production or trade and interfere with the achievement of any of the other objectives set forth in Art. I'. 13 The G.A.T.T. entered into force through the Protocol of Provisional Application in the expectation that the H.C. would be ratified soon thereafter. This expectation did not materialise, however, as a proposal to delete Art. XXIX was unanimously adopted in 1955.14 Failure by one contracting party to ratify this change to the General Agreement led to the continued inclusion of the Article in the text of the G.A.T.T.15 Although Art. XXIX is technically still a part of the G.A.T.T./W.T.O., it is at least questionable whether Members are still bound by obligations contained in Chapter 5 of the H.C. This point of view has been accepted in the G.A.T.T. case law.16 12
The Havana Charter was designed to establish the International Trade Organisation. It never entered into force because the US Congress refused to ratify it. See John H. Jackson, World Trade and the Law of the G.A.T.T., (1969) 36. 13 See the Havana Charter, Art. 46, reproduced in UN Doc. E/Conf.2/78 (1945). Art. 1 states the objectives of the Charter, which include trade expansion and income growth. 14 B.I.S.D. 3S/240. (1955). 15 See Protocol Amending Part I and Arts. XXIX and XXX of the G.A.T.T., G.A.T.T.-Status of Legal Instruments 2-7.1 (1994). 16 'The Panel noted that the deletion of Art. XXIX of the General Agreement was proposed in 1955 and accepted by all but one contracting party, and that—although this Article is technically still in force—it refers to an instrument which itself has never been implemented and the acceptance of which is no longer pending as is assured in Art. XXIX. This leaves considerable doubt as to the manner in which its provisions would have been interpreted if they had entered into force.' See Canada-Administration of the F.I.R.A., B.I.S.D. 3OS/161 §5.12 (1984).
4 - Competition Policy Objectives
287
Notwithstanding the 1955 attempt to eliminate the link to the H.C., G.A.T.T. contracting parties established a Group of Experts in 1958 to study whether and to what extent the G.A.T.T. should deal with R.B.P.s.17 Members of the Group were unable to reach consensus in their 1960 report. A majority considered that it would be unrealistic at that moment to recommend the negotiation of multilateral disciplines to control R.B.P.s. However, they recognised that R.B.P.s could have a harmful effect on international trade, and proposed that bilateral consultations take place between interested parties on specific practices. The suggested procedure was to keep this outside the realm of Art. XXIII (dispute settlement): '[members of the Group] were not competent to judge . . . whether the provisions of Art. XXIII would be applicable. However, the majority were convinced that, regardless of the question whether Art. XXIII could legally be applied, they should recommend to the CONTRACTING PARTIES that they take no action under this Article'.18 A minority proposed a multilateral procedure to address R.B.P.-related issues. A group of experts would deal with those issues that could not be settled bilaterally, and would submit a report to the G.A.T.T. secretariat, which would in turn report annually to the contracting parties. The disagreement led to a decision to provide a forum for consultations in this area, upon request, and appoint a group of experts on R.B.P.s 'to be convened when appropriate'.19 While the adopted report contains two differing opinions on the possibility of addressing R.B.P.s in the context of Art. XXIII, there was agreement on one crucial point: R.B.P.s could have harmful effects on world trade. Moreover, even the majority refrained from addressing the crucial question of the general applicability of Art. XXIII in this context. b. Preliminaries
This subsection examines the extent to which the current G.A.T.T.AV.T.O. agreement can be employed in order to address competition policy related concerns. The G.A.T.T. agreement offers three legal routes for complaints, the legal reach of each of which is discussed below. Three general issues are first considered: the incorporation theory, the role of previous G.A.T.T. panel decisions, and the necessity of attributability. The distortionary spillovers on foreign producers discussed in Section A exist only to the extent that the authorities deciding on and implementing competition policies do not consider the interests of these producers. However, in 17 18 19
B.I.S.D., 7S/29 (1959). B.I.S.D., 9S/170fT. (1961). Id.
288
Working Paper VIII
most industrialised countries, a firm with foreign ownership is given the same treatment under competition law as a national firm residing in the country, provided it has a local subsidiary through which its sales occur. This is so because most domestic laws accept the 'incorporation theory' for conferring jurisdiction, according to which, to the extent that a company is established under the laws of the host state, jurisdiction is conferred to the host state. Hence, the possibility of direct neglect of foreign producer interests is, at least theoretically, restricted to those firms that sell through independent agents. While this still includes a considerable volume of imports, it is far less than all imports. G.A.T.T. panel reports are relevant in the W.T.O.-context, since they constitute part of the G.A.T.T. acquis. Art. l(b)(iv) of G.A.T.T. 1994 specifies that adopted G.A.T.T. 1947 panel reports are an integral part of G.A.T.T. 1994, which is one of the W.T.O. agreements. Thus, there is a legal requirement for future panels to consider them, while retaining the possibility of deviating from their dictum. This was the position expressed in the 1996 panel report on 'Japan—Taxes on Alcoholic Beverages'. However, the Appellate Body (A.B.) report dealing with the same issue expressed a different view, arguing that adopted panel reports do not constitute subsequent practice in the sense of Art. 31 of the Vienna Convention on the Law of Treaties (V.C.L.T.).20 Unfortunately, the A.B. did not clearly address the issue whether subsequent panels must address previously adopted panel reports that dealt with the same issue. We believe that the position taken in the 1996 panel report is fully in line with the prevailing doctrine in public international law. Sir Gerald Fitzmaurice authoritatively states: '[I]t would seem that, although the Court is not obliged to decide . . . on the basis of previous decisions as such, what it can do is to take them fully into account in arriving at subsequent decisions, and that... it is mandatory for it to apply judicial decisions in the sense of employing them as part of the process whereby it arrives at its legal conclusions in the case.'21 Accordingly, irrespective of whether adopted panel reports are characterised as subsequent practice, they must be addressed by subsequent panels dealing ratione materiae with the same issue. At the same time, subsequent panels dealing ratione materiae with the same issue can always deviate from previous rulings. 20
Reprinted in 8 Int'l Legal Materials 679 (1969). Sir Gerald Fitzmaurice, The Law and Procedure of the International Court of Justice, (1986) 584. For Franck, this is part of the process necessary for all international tribunals to follow in their quest for legitimacy. Thomas M . Franck, Fairness in International Law and Institutions, (1995) 25. 21
4 - Competition Policy Objectives
289
The G.A.T.T. is a contract among governments. The legal consequence is that unless the G.A.T.T. rules are recognised to have direct effect, only governments have locus standi before the competent fora. There has been intense debate in the literature of public international law on whether it is for national constitutions to recognise the direct effect of international law. Until now, with the exception of the European Community legal regime, this seems not to be the case. This means that it is left to individual W.T.O. Members to decide whether to confer locus standi to private parties with respect to W.T.O. rules. In practice, W.T.O. Members systematically refuse to confer locus standi for a number of reasons, discussed in Section C below.22 Hence, in order for the G.A.T.T./W.T.O. agreement to offer an option to address competition policy related concerns, the distortions mentioned in Section A must be attributable to governments. Conversely, to the extent that a distortion cannot be attributed to a government, it cannot be challenged by the current G.A.T.T./W.T.O. regime. Attributability, therefore, becomes the necessary, albeit not always sufficient, condition for G.A.T.T./W.T.O. rules to become an option. There is a discrepancy between economic thinking and actual status of law in thisfield.Generally, in economic thinking, any policy (including, for example, not to enact competition laws) reflects a conscious choice by decisionmakers based on the economic consequences of all the available actions. The status of law on attributability is more restrictive, although international law is still in the process of being codified on this issue. The issue of attributability of a behaviour to a state has recently been the subject of in-depth discussions in the International Law Commission (I.L.C.) in the context of the codification of the law in thefieldof State Responsibility.23 This body of work has already been influential with respect to international adjudication fora24 and G.A.T.T. case law.25 The relevant G.A.T.T. provisions in this context are Art. Ill (national treatment), Art. XI (quantitive restrictions) and Art. XXIII:lb (nullification and impairment of benefits from the G.A.T.T. agreement). 22
F o r an excellent illustration of the E.C. caselaw, see Piet Eeckhout, The Domestic (1997) 34 Legal Status of the W.T.O. Agreement: Interconnecting Legal Systems, C o m m o n M k t . L. Rev. 11. 23 See U . N . G . A . D o c . A/CN.4/L/528/Add. 2. (1996). 24 See, e.g., the arbitral sentence in Rainbow Warrior as presented in Jean Charpentier, L'affaire d u R a i n b o w Warrior, et la responsabilite des Etats (1991). 25 See the unadopted panel report on U.S.—Countervailing Duties on Steel Products from Germany, France and the U.K., S.C.M./185, where the panel considered, inter alia, the question whether private bank debt forgiveness could be attributed to a government.
290
Working Paper VIII
c. The Form of Complaints in the G. A. T. T. IW. T. O. System There are three grounds for a W.T.O. Member to invoke the dispute settlement procedures. According to G.A.T.T. Art. XXIII, a W.T.O. Member can claim that benefits accruing to it directly or indirectly under the G.A.T.T. are being nullified or impaired or that the attainment of any objective of the G.A.T.T. is being impeded as a result of: (1) the failure of another contracting party to carry out its obligations under the Agreement (violation complaints); (2) the application by another contracting party of any measure, whether it conflicts with the provisions of this Agreement (non-violation complaints); or the existence of any other situation (situation complaints). Violation complaints essentially concern cases where a G.A.T.T. provision allegedly has been infringed. Non-violation complaints, in contrast, deal with cases where even though a W.T.O. Member taking a particular action is not violating a G.A.T.T. provision, a benefit accruing to other W.T.O. Members is nonetheless nullified or impaired as a result of the action. Finally, situation complaints deal with all possible cases other than violation and non-violation complaints. As stated above, Art. XXIII: 1 is not applicable to purely private anticompetitive practices (i.e. cases where no government involvement exists). Since they cannot be attributed to governments, they do not violate any G.A.T.T. provision. In what follows we will slightly reverse the order and first examine violation" complaints, next situation complaints (where there is not yet any G.A.T.T. case law), and finally non-violation complaints. Violation Complaints Violation complaints concerning competition policies refer either to the national treatment obligation in Art. Ill or the prohibition of quantitative restrictions in Art. XL Both are addressed below. W.T.O. Members are free to enact any competition law they deem appropriate, or to abstain from enacting such a law. Their only obligation with respect to the other W.T.O. Members is to apply national laws in thisfieldin conformity with the principle of national treatment — that is, they should not discriminate between domestic and foreign products. Art. Ill obliges W.T.O. Members to guarantee national treatment to imported products 'with respect to any laws, regulations and requirements affecting their internal sale, offering for sale, purchase, transportation, distribution or use'. The term 'affecting' has been interpreted quite broadly in the G.A.T.T. case law26; the interpretation 26
According to the 1958 panel report on Italian Discrimination Against Imported Agricultural Machinery, L/833, 7S/60, 64 para. 12 (23 Oct. 1958), the term 'affecting' was used to cover also 'laws or regulations which might adversely modify the conditions
4 - Competition Policy Objectives
291
seems to extend and cover domestic antitrust laws as well, in line with the economic analysis above. Moreover, until now, application of laws has been the subject of disputes in G.A.T.T. cases. In fact, only in exceptional cases have the laws themselves been the subject of disputes, since the judicial review standard applied by G.A.T.T. panels in this regard is quite stringent. Decisions concerning competition policy enforcement can, however, be adjudicated.27 For instance, in the case of a 'national champion' merger, let us assume there are three firms in an industry, firms 1 and 2 which reside in a country A, and a foreign firm 3 that serves the country A market through a sales agent. The three firms are symmetric in all respects other than nationality. A 'national champion' merger would involve the approval of a merger between firms 1 and 2, under circumstances where a merger between one of those two firms with firm 3 would not be approved. Such a merger would arguably violate the national treatment clause of Art. Ill, since it is the result of the execution of a law that affects the 'internal sale, offering for sale, purchase, transportation, distribution or use' of the foreign firm. Moreover, when enacting/applying competition laws, W.T.O. members must also respect the obligations resulting from Art. XI (prohibition of quantitative restrictions). The premise of the argument related to this Article is that private restrictions on competition amount to quantitative restrictions (Q.R.s) which can, in many cases, be attributed to government actions. Consequently, they violate Art. XI. Arts. XI to XIV provide the legal framework for addressing Q.R.s: Art. XI forbids Q.R.s; Art. XII exceptionally permits Q.R.s used for balance-of-payments (B.O.P.) reasons; Art. XIII requires that Q.R.s apply on a nondiscriminatory (erga omnes) basis; and Art. XIV provides that if Q.R.s are applied for B.O.P. reasons, the nondiscrimination requirement may be waived.28 Art. XI: 1 imposes the basic obligation on W.T.O. Members to refrain from introducing or maintaining Q.R.s.29 Import quotas constitute of competition between the domestic and imported products on the internal market'. This view was confirmed in the 1989 panel report on United States—Section 337 of the Tariff Act of 1930, L/6439, 36S/345, 385/6, para. 5.10 (7 Nov. 1989). The legal value of adopted panel reports is considered infra. 27 For a discussion of attributability in the context of Art. XI, see infra. 28 John H. Jackson, World Trade and the Law of the G.A.T.T., 308 (1969). 29 Art. XI: 1 states: ' N o prohibitions or restrictions other than duties, taxes or other charges, whether made effective through quotas, import or export licences or other measures, shall be instituted or maintained by any contracting party on the importation of any product of the territory of any other contracting party or on the exportation or sale for export of any product destined for the territory of any other contracting
292
Working Paper VIII
import restrictions within the meaning of Art. XI: 1, whether or not they actually impede imports.30 Art. XI addresses 'prohibitions or restrictions other than duties, taxes or other charges, whether made effective through quotas, import or export licences or other measures'. Art. XI is, consequently, similar to Art. Ill in that its purpose is to establish competitive conditions, independent of trade effects.31 The extent to which Art. XI can be used to address beggar-thy-neighbour competition policies hinges on interpretation of the term 'measures', since the terms 'quotas', 'import' and 'export licences' are unambiguous, and do not directly relate to competition-policy issues. A G.A.T.T. panel noted that measures must be governmental and that the drafting history of the G.A.T.T. suggested that the drafters were primarily concerned with the effectiveness of the measures.32 The panel did not, therefore, feel bound by the legal qualification of the measure in question, but focused its attention on effect. In this case, the measure in question was a non-mandatory 'administrative guidance' by the Ministry of International Ministry of International Trade and Industry (M.I.T.I.), based on consensus and peer pressure. While the Panel found that this was a measure under Art. XI, it emphasised that its conclusion was limited to this specific case and should not be construed to encompass all 'administrative guidance'. Another Panel concluded that in determining whether specific non-mandatory measures contravene Art. XI, two criteria must be satisfied: 'First, [the existence of] r e a s o n a b l e g r o u n d s t o believe that sufficient incentives o r disincentives existed for n o n - m a n d a t o r y measures t o take effect. Second, [that] the operation of t h e m e a s u r e s t o restrict exports [is] essentially dependent o n G o v e r n m e n t action o r intervention. > 3 3 party.' A r t . X I : 2 lists t h e permissible exceptions. See T h e Results of the U r u g u a y R o u n d Multilateral T r a d e N e g o t i a t i o n s : T h e Legal Texts (Geneva 1994), at 510 et seq. 30
E.E.C.-Payments a n d Subsidies t o Processors a n d Producers of Oilseeds a n d Related A n i m a l - F e e d Proteins, L/6627, B.I.S.D. 37S/86 (25 J a n 1990) ( T h e mere existence o f a Q.R. is presumed t o cause nullification o r impairment not only because of its trade effects b u t also because it w o u l d lead to increased transaction costs and would create uncertainties which could affect investment plans'); Japanese Measures o n I m p o r t s of Leather, B.I.S.D. 31S/113 (15/16 M a y 1984). See also Brazilian I m p o r t Taxes, Vol.II/184-5 (30 J u n e 1949). 31 See 1987 panel report, U n i t e d States—Taxes o n Petroleum a n d Certain Imported Substances, 34S/136, 158, §5.1.9 (17 J u n e 1987). 32 See Japan-Restrictions on Imports of Certain Agricultural Products, L/6253, B.I.S.D. 35S/242, §5.4.1.4 (22 Mar. 1988) 33 Japan-Trade in Semiconductors, L/6309, B.I.S.D. 35S/153^, 154-5, §§108-9 (4 May 1988). The matter of concern was a minimum export price that the Panel found to operate in the form of a mandatory requirement.
4 - Competition Policy Objectives
293
Accordingly, non-mandatory government measures can violate Art. XI. Thus, according to the two G. A.T.T. panels, if a measure can be attributed to a government, and has the effect of a Q.R., Art. XI can come into play. Even if the criteria mentioned in the second panel should change in the future, it is precisely this concept of 'attributability' that is crucial to application of the G.A.T.T./W.T.O. provisions. A number of measures have been found inconsistent with Art. XI: 1, such as: (1) a clause in a U.S. law prohibiting, with certain exceptions, the importation or public distribution in the U.S. of a copyrighted work consisting mainly of non-dramatic literary material in the English language34; (2) the provisions of the U.S. Marine Mammal Protection Act prohibiting imports of tuna if harvested in a way that does not respect the standards imposed by the law35; (3) minimum import price systems (enforced by additional security)36; and (4) minimum export price systems.37 The G.A.T.T. also recognises that Q.R.s may be used by state-trading enterprises. An interpretative note in this regard states that throughout Arts. XI, XII, XIII, XIV and XVIII, the terms 'import' or 'export restrictions' cover restrictions made effective through state trading operations. The note refers to 'restrictions made effective through State-trading operations', and not to 'import restrictions', in an attempt to ensure that contracting parties cannot escape their obligations with respect to private trade by establishing state trading operations.38 In conclusion, to the extent that a practice operates like a Q.R. and can be attributed to a government in accordance with the dictum of the relevant G.A.T.T. case law, the Art. XI option deserves, at least, to be further explored. 34
U . S . - M a n u f a c t u r i n g Clause, L/5609, B.I.S.D. 315/91, §42(i) (15/16 M a y 1984). U.S.-Restrictions o n I m p o r t s o f T u n a , DS21/R, §§5.17, 5.18, 5.36 (unadopted) (1994). 36 E . E . C . - P r o g r a m m e of M i n i m u m I m p o r t Prices, Licences a n d Society Deposits for Certain Processed Fruits a n d Vegetables, L/4687, B.I.S.D. 25S/99-100, §4.9 (18 Oct. 1978). 37 J a p a n - T r a d e in Semi-conductors, B.I.S.D. 35S/153, §105 (1989). 38 As explained in a G.A.T.T.-panel: 'this was a recognition of the fact the in the case of enterprises enjoying a monopoly of both importation and distribution in the domestic market, the distinction normally made in the General Agreement between restrictions affecting the importation of products and restrictions affecting imported products lost much of its significance since both types of restriction could be made effective through decision by the monopoly. The Panel considered that systematic discriminatory practices of the kind referred to should be considered as restrictions made effective through 'other measures' contrary to the provision of Article XI: 1': Canada-Import, Distribution, B.I.S.D. 35S/89, §4.24 (1989). See also Japan-Restriction on Imports of Certain Agricultural Products, B.I.S.D. 35S/229, §§5.2.2.2 (1984). 35
294
Working Paper VIII
This option could be used to address export cartels, which are one of the most common examples of a competition policy distortion, or practices covered by the broad U.S. Noerr-Pennington doctrine.39 Moreover, the governmental link can be easily demonstrated with respect to legislative schemes such as the Webb-Pomerene Act.40 Situation Complaints The wording of Art. XXIII:l(c), which addresses situation complaints, suggests that in principle, it is all-encompassing. Until recently, no panel has ever considered a situation complaint.41 The one exception concerned E.C. allegations that certain measures taken by the Japanese government were inhibiting the importation of products manufactured in the E.C. In its request for the establishment of a panel, the E.C. claimed that benefits of negotiations with Japan had not been realised because a number of factors made it extremely difficult for E.C. products to penetrate the Japanese market. Factors mentioned included the marked concentration and interlinking of the structure of production, finance and distribution in Japan, which made it difficult for for-
39
See Eleanor M . F o x , Toward World Antitrust and Market Access, 91 Am. J Int'l L 1,17 (1997). Essentially through such practices, private parties request from competent authorities lax enforcement (or n o enforcement) of domestic antitrust laws, to the extent that the anticompetitive effects are not felt (or not substantially felt) within their national market. 40 O n e could plausibly argue that Art. XI operates like a super 'effects doctrine', since the plaintiff does n o t even need to show that an R.B.P. produces negative effects in its market. Rather, it need only show that an R.B.P. (operating like a Q.R., for example, a market sharing arrangement) is in place. 41 Other than for the E.U. v. Japan case, G.A.T.T. D o c . L/5479 (1989), a threat of making use of this provision occurred only once. This was following the accession of the United Kingdom to the E.U., when Canada's access opportunities to the U.K. market were substantially reduced. Canada felt that Art. X X I I I : l c was the appropriate provision to handle such issues, since negotiations under Art. XXIV:6 with the E.U. were expected to be unsuccessful. Subsequently, however, Canada dropped the case. G.A.T.T. Docs. L/4107, L/M/101, CMV7250, CAV7251. Ernst-Ulrich Petersmann,
Violation Complaints and Non-Violation Complaints in Public International Trade Law, German Y.B. Int'l L. 175, 227 (1991) has argued that situation complaints had fallen in desuetudo and that they were no longer an appropriate instrument for adjusting the G.A.T.T. framework of rights and obligations. The recent Kodak/Fuji dispute seems to contradict this view. See Frieder Roessler, The Concept of Nullification and Impairment in the Legal System of the World Trade Organization, in, Ernst-Ulrich Petersman (ed.) International Trade Law and the G.A.T.T./W.T.O. Dispute Settlement System (1996), (arguing that situation-complaints should be used in cases where no adoption of a 'measure' is involved).
4 - Competition Policy Objectives
295
eign suppliers to establish distribution channels.42 The E.C. argued that such factors nullified and impaired benefits that would otherwise have accrued to it. The discussion that followed in the G.A.T.T. Council failed to shed light on this case, and the E.C. ultimately decided not to pursue it.43 In this case, the E.C. qualified its complaint as a 'situation complaint' under Art. XXIII: l(c).44 It is questionable whether Art. XXIII:l(c) can provide an appropriate forum to address concerns related to R.B.P.s.45 Recently however, the U.S. invoked Art. XXIII:l(c) in its dispute with Japan (Kodak/Fuji) as the appropriate legal basis, in part, for the panel to address the claim. No conclusions can be drawn since, as stated above, the standard of review of panels in this context remains unknown. Non-violation Complaints
The non-violation provision has an autonomous function in the G.A.T.T. dispute settlement system. Most non-violation cases brought before G.A.T.T. panels have involved the nullification or impairment of a negotiated tariff concession by the introduction of a subsidy that could not have been reasonably anticipated by the other party at the time such tariff concession was made. However, this is not the only type of action that might give rise to non-violation complaints.46 As Petersmann points out, 'non-violation complaints are based on . . . legal principles of effectiveness of concessions, reciprocity and bona fide protection of reasonable expectations'.47 Reasonable expectations are created by concessions negotiated at any point in time. The Vienna Convention on the Law of the Treaties (V.C.L.T.)48 obliges signatory states to 42
See G.A.T.T. Doc. L/5479 (1982). See G.A.T.T. Doc.C/M/167, at 9 (1982). 44 M a r c o Bronckers, Selective Safeguard Measures in Multilateral T r a d e Relations, (1985)151. 45 T h e first sentence of Art. XXIII:2 reads: 'If n o satisfactory adjustment is effected between the contracting parties concerned within a reasonable time, o r if the difficulty is of the type described in paragraph l(c) of this Article .. .' A t t e m p t s t o reach a satisfactory adjustment is n o t a precondition to seek the establishment of a Panel. Since the general rules, as embodied in Arts. X X I I - X X I I I of G.A.T.T., require t h a t for a Panel to be established, bilateral consultations must have taken place, 'situation c o m p l a i n t s ' should be regarded only in exceptional circumstances as a possible m e a n s of requesting establishment of a Panel. T h e conditions under which c o m p l a i n t s could b e qualified as 'situation-complaints' have never been specified in G . A . T . T . c a s e law. 46 In the G.A.T.T. legal system, Art. X X V I I I (modification of schedules) for example, can also provide a forum for such complaints. 47 Petersmann, supra note 4 1 , at 225. 48 V.C.L.T., supra note 20. T h e Convention applies only to s u b s e q u e n t l y concluded treaties. T h e portion governing general interpretation of treaties is considered to be 43
296
Working Paper VIII
refrain from acts frustrating the object of the treaty they signed (Art. 18) and recognises, in more general terms, that all treaties must be implemented in good faith (Art. 26). Since non-violation complaints aim to protect the competitive conditions established by agreed tariff concessions, any government measure that offsets such concessions can, in principle, be brought before G.A.T.T. as a non-violation complaint under Art. XXIII: l(b). For anticompetitive practices to be the subject of a non-violation complaint, three conditions must be met: (1) the measure must be applied by a government; (2) it must alter the competitive conditions established by the agreed tariff bindings; and (3) the measure could not have been reasonably anticipated at the time the tariff concessions were negotiated. Business practices that are not subject to any government involvement are excluded from the scope of a nonviolation complaint. There is some ambiguity, however, regarding the extent of government involvement required. Art. XXIII:l(b) speaks of'application of any measure' that nullifies and impairs benefits. The term 'measure' suggests that not only formal laws and regulations are included, but also other forms of government action that are necessary to make the government choice operative. The term 'application' suggests that positive action is required. If correct, then the mere tolerance by a government of an R.B.P. is not a sufficient ground for a non-violation complaint. However, if tolerance is reflected in a positive (specific) action, the first criterion will be satisfied. For example, an exemption granted by the competent antitrust authority to private enterprises, which effectively reduces market access opportunities for products of third countries by establishing distribution channels that are difficult to penetrate, could constitute such a positive action. In such cases, if the action (exemption) could not have been reasonably anticipated at the time market access conditions were negotiated, a W.T.O. Member might bring a non-violation complaint. Regarding the second condition, there should be a causal link between the measure and the alteration of the competitive conditions established by the agreed tariff bindings.49 Regarding the third condition, 'unexpectedness,' panels have interpreted the scope of protection of 'reasonable expectations' under G.A.T.T. as follows. First, the balance of concessions is the highest ranking codifying customary international law and therefore applies to treaties concluded before its entry into force and is binding even on countries that are not parties to it. 49 So far, disputes submitted in this context to G.A.T.T. panels concerned cases where actual trade damage (i.e. trade effects) were shown. In the future, it will be interesting to see how W.T.O. panels react in cases where the allegedly unexpected measure has not yet produced its results (that is, whether the absence of trade effects remains a privilege of violation complaints or whether it should be extended to non-violation complaints as well).
4 - Competition Policy Objectives
297
norm protected in the W.T.O. system against both legal and illegal actions disturbing this balance. Secondly, W.T.O. Members can reasonably expect that this balance will not be nullified or impaired, 'failing evidence to the contrary'.50 The content of the term 'reasonable expectations' has not been interpreted in G.A.T.T. case law. Panels tend to follow a case-by-case approach. The appropriateness of defining specific criteria for determining the reasonableness of expectations is, indeed, questionable. A procedural rule following the dictum in the quoted Working Party on 'Other Barriers to Trade' could help. If 'failing evidence to the contrary' is always to be interpreted as a shift of the burden of proof to the Member seeking to reject the argument that something was reasonably expected, then 'reasonable expectations' will be perceived as always present and therefore protected when a concession is negotiated, unless the W.T.O. Member that allegedly modified the value of the concession provides evidence to the contrary. If the term 'evidence' is given its ordinary meaning, then the imposed standard of proof on the Member invoking the unreasonableness of expectations is high.51 The substantive content of the term 'reasonable expectations' is therefore arguably of secondary importance, since they exist until proven unreasonable. If this interpretation is accepted, the onus will be on the Member that alters the value of the concession to show that at the stage of negotiations, serious reasons should have led the other W.T.O. Member to believe that the agreed level of competitive conditions would eventually be reduced (modified negatively in the future).52 50
'[T]he recognition of the legitimacy of a n expectation relating t o t h e u s e of p r o duction subsidies in n o way prevents a contracting p a r t y f r o m using p r o d u c t i o n subsidies consistently with the General Agreement; it merely d e l i n e a t e s t h e scope of the protection of a negotiated balance of concessions.' See E . E . C . - P a y m e n t s a n d Subsidies Paid t o Processors a n d Producers of Oilseeds a n d R e l a t e d A n i m a l Feed Proteins, L/6627, B.I.S.D. 37S/128, §148 (25 J a n . 1990). Also, ' a c o n t r a c t i n g p a r t y which h a s negotiated a concession under Art. II m a y be assumed, for t h e p u r p o s e o f A r t . X X I I I , to have a reasonable expectation, failing evidence t o the c o n t r a r y , t h a t t h e value o f the concession will n o t be nullified or impaired by the c o n t r a c t i n g p a r t y which g r a n t e d t h e concession by the subsequent introduction o r increase of a d o m e s t i c subsidy o n the product concerned.' R e p o r t of the W o r k i n g Party on O t h e r Barriers t o T r a d e , B.I.S.D. 3S/224, §15 (1955). 51 T h e French translation of the term 'evidence' in t h e G . A . T . T . A n t i - D u m p i n g Code, for example, is 'elements de preuve'. 52 One may, nonetheless, speculate as to what could come under the term 'reasonable expectations.' For instance, can a W.T.O. Member reasonably expect another W.T.O. Member to respect its declarations in other international fora? If so, then The Set of Multilaterally Agreed Equitable Principles and Rules for the Control of Restrictive Business Practices, concluded at the U.N. (reproduced in 19 Int'l Legal Materials 813
298
Working Paper VIII
2. The Competition Regime This section will address the extent to which bilateral cooperation agreements in the field of competition law can adequately deal with the concerns mentioned in Section A. Further, it will consider whether extraterritorial application of competition laws can help internalise the distortions highlighted in Section A, without at the same time creating new distortions. a. Bilateral Cooperation Agreements
Some distortions can be addressed by means of bilateral cooperation in the competition field among competent national authorities. For instance, Wood offers empirical examples to the effect that bilateral cooperation agreements can, in practice, be an efficient way to address distortions.53 Notwithstanding the legal difficulties associated with its entry, the 1991 E.C./U.S. Cooperation Agreement54 is a good example. The move from 'negative' to 'positive' comity, whereby an interested party can request the competent authorities of another country to become active in a particular case, can greatly improve existing cooperation. However, the newness of these instruments implies that little empirical evidence is available at present as to their utility. (1980)) might be the basis of reasonable expectations between W.T.O. Members that they will not alter the value of negotiated concessions (after 22 Apr. 1980, the date of adoption of the Set) through behaviour inconsistent with the set of agreed principles. Notwithstanding the non-binding character of the U.N. resolutions, the argument can be made that such acts oblige signatories at least to act in a non-inconsistent way. For a discussion of this 'anti-inconsistency' principle, see Hans Baade, The Legal Effects of Codes of Conduct for M.N.E.s., in Legal Problems of Codes of Conduct for Multinational Enterprises (Norbert Horn ed. 1980). Cf. Frieder Roessler, Law, De Facto Agreements and Declarations of Principle in International Economic Relations, G e r m a n Y B Int'l L. 40 (1978) (calls such non-binding rules de facto agreements a n d defines them as 'declarations intended to give ground to expect p e r f o r m a n c e or forbearance of actions without creating legal rights a n d obligations'); G e o r g Berrisch, The Establishment of New Law Through Subsequent Practice in G.A. T.T., 16 N . C . J . Int'l L. & Com. Regs. 497 (1991) (supporting position of Roessler). T h e logical counter-argument is that limits must be placed on the application of this reasoning, as otherwise everything might be brought before the W.T.O. However, the issue here is n o t whether a convention signed outside W . T . O . legitimises a cause of action in the W . T . O . , b u t simply what can be understood to constitute a reasonable expectation. T h e cause of action in the W.T.O. context is the nullification or impairment of the agreed balance of concessions. 53
D i a n e W o o d , Options for the Future, 44 D e P a u l L. R e v . 1289 (1995). Agreement Between t h e Government of the U n i t e d States of A m e r i c a a n d the Commission of the European Communities R e g a r d i n g t h e A p p l i c a t i o n of Their Competition Laws, [1995] O.J. L47 (27 A p r . 1995). 54
4 - Competition Policy Objectives
299
Two issues deserve further consideration in the future in this context: the standing of private parties and confidentiality. Regarding standing, problems analogous to those that exist in the trade field also exist in the competition field. Standing of private parties is a domestic law issue. Bilateral agreements are, among other things, instruments of diplomatic protection. It is left to domestic authorities first to decide whether to support a private concern by requesting a foreign competition authority to become active in a case, and then to argue the case on behalf of this party. On the issue of confidentiality, Wood observes that this has proved to be a major obstacle to a more meaningful cooperation among national competition authorities.55 None of those issues, however, reflects a need to negotiate new substantive international competition rules. They rather concern procedural impediments that are better addressed at their source. Such agreements derive their credibility from 'tit-for-tat' considerations: 'y° u take into account my interest today, I might do the same with respect to yours tomorrow'. For increasingly intertwined economies, repetition adds to credibility. b. Extraterritorial Application of Competition Laws The argument has been raised that establishment of common competition rules would be the appropriate way to address the extraterritorial application of competition laws by some countries, especially the U.S. The argument, which assumes that such extraterritorial application of competition laws runs counter to public international law obligations, is rooted in the manner in which U.S. domestic courts have applied the 'effects' doctrine. However, this critique often ignores that acceptable extraterritoriality of the same laws can often address a number of their other concerns.56 At the risk of over-simplification, the European Court of Justice (E.C.J.) has arguably applied the 'effects' doctrine more cautiously than U.S. domestic courts, as the E.C.J. will assert jurisdiction only if there is some form of implementation.57 There is practically no difference in how the U.S. and the E.U. would handle 55
W o o d , supra note 53. See the analysis by Eleanor Fox, supra note 39, who correctly makes the point that the active use of the Effects Doctrine over an extended period removed the need to address transnational restrictive business practices through international negotiations. See also Aubry D. Smith, Bringing Down Private Trade Barriers—An Assessment of the United States' Unilateral Options: Section 301 of the 1974 Trade Act and Extraterritorial Application of U.S. Anti-trust Law, 16 Mich. J. Int'l L. 241 (1994). 57 See Walter v a n G e r v e n , E. C. Jurisdiction in Antitrust Matters: the Wood Pulp 56
Judgment, in Barry Hawk (ed.), Proceedings of the Fordham Corporate Law Institute 1989(1990).
300
Working Paper VIII
cases in which the anticompetitive effects in the domestic market are felt through imports.58 Differences arise with respect to cases where importers in third countries block exports of the E.U. or the U.S. through allegedly anticompetitive practices.59 In such cases, U.S. domestic courts have already asserted jurisdiction. Davidow cites cases (e.g. the C. Itoh case) that have been adjudicated in the U.S. and could hardly come under the jurisdictional reach of the E.C.J, unless the E.C.J changes its position on the issue.60 The appropriate benchmark to establish whether the U.S. violates its international obligations in this respect is public international law, be it treaty or customary international law.61 Any dispute in this context is best subjected to the International Court of Justice (I.C.J.). The I.C.J., or any international third party adjudication, has the competence to determine the legitimate reach of domestic jurisdictions. Accordingly, the benchmark to determine the legitimacy of the jurisdictional reach of U.S. competition laws exists, and there is no compelling reason to 'invent' a new benchmark. The existing regime can provide the solution to a legitimate concern. If this were not so, the position that the best way to address irregularities is to overregulate would be correct. Moreover, empirical evidence shows that bilateral cooperation agreements can provide an appropriate forum to address claims of extraterritoriality.62 In fact, one of the most important objectives to be accomplished through bilateral cooperation agreements is to define a mutually acceptable jurisdictional reach. Finally, some important market access restricting policies are not yet covered by the W.T.O. For instance, Low and Subramanian have discussed the role of investment liberalisation in rendering non-tradable sectors more competitive.63 They conclude on the issue whether there is a case for common competition rules that 'except for the so-called Japan problem, which can be addressed directly, there does not seem to be the same commercial imperative for new rules as there was for rules on services or T.R.I.P.s'. 58
A l t h o u g h it is still questionable whether, in cases where no implementation of conduct takes place in the E.U. territory (i.e. in the case of an export restriction), the E.C.J. would still assert jurisdiction. See id. 59 See Eleanor Fox, supra note 39 (explaining the function of the famous 'Footnote 159' a n d t h e consequences of its deletion). 60 Joel Davidow, Keiretsu and US Antitrust, 24 L & Pol. Int'l Bus. 1035 (1993). 61 N o t e t h e parallelism with Section A, where the point was made that extraterritorial use of competition laws counteracts certain distortions, but creates other distortions. 62 W o o d , supra note 53. 63 Patrick Low and Arvind Subramanian, T.RI.M.s in the Uruguay Round: An Unfinished Business?, in The Uruguay R o u n d and the Developing Economies (Will M a r t i n a n d L. Alan Winters eds. 1995) (Discussion Paper N o . 307, The World Bank, W a s h i n g t o n , D.C.).
4 - Competition Policy Objectives
301
3. Per se versus 'Rule of Reason' The existing regime offers two approaches. The trade regime contains substantive rules that can be used against distortions stemming from domestic competition policies. The competition regime provides a procedural vehicle for competition related disputes. The latter approach has been used more often in practice. The question arises whether private parties believe that competition authorities can deal more effectively with such cases, or whether the approach followed in bilateral competition agreements is better suited to deal with the issues at hand. Here, only the second question is relevant. Arguably, existing differences among domestic competition laws, even among ideologically similar states, are substantial enough to make an international agreement on substantive rules quite unlikely in the near future.64 However, even if we assume that it is possible, it is at least questionable whether substantive rules at the international level would be a positive step. There is an ongoing tendency at the domestic level to move away from per se prohibitions towards the rule-of-reason approach. The list of per se offences is therefore shrinking. As mentioned above, existing substantive rules in the trade field can, in principle, help address the most flagrant per se violations. How can new substantive rules help address the issue ofjudicial review at the international level of decisions by domestic authorities which are based on the rule of reason? Independent of the economic argument in favour of a competition in rules approach, the very nature of decisions at the domestic level provides an impediment to the negotiation of substantive international rules. C. The Use of the Trade and the Competition Regimes in Practice The previous Section suggested that the current legal regime in the trade and competition fields can, in principle, deal with many of the potential problems concerning distortions identified in Section A. This Section considers the extent to which these possibilities have been used in practice, starting with an examination of the trade regime, and then turning to the competition regime. Empirical evidence shows that sparse use has been made of the existing trade regime, even though globalisation of the world economy leads to a proliferation of alliances between firms that affect many markets simultaneously. The question arises how the sparse use of the existing regime should be interpreted. 64
For a synthetic comparison of U.S. and E.U. competition law, see Eleanor M. Fox and Robert Pitofsky, U.S. and E.U. Competition Law: A Comparison, unpublished mimeo(1995).
302
Working Paper VIII
1. Evidence Regarding Use of the Existing Regimes a. The Trade Regime In the trade regime, we distinguish between G.A.T.T./W.T.O. law, which is available exclusively to governments, and instruments of 'diplomatic protection,' which can be invoked by private parties. As stated above, private parties do not have locus standi before the competent G.A.T.T./W.T.O. bodies. The KodaklFujica.se, which was submitted to the W.T.O. dispute settlement system, is the first competition related case submitted to the W.T.O.65 The U.S. arguably believes that the existing regime (non-violation and situation complaints) suffices to address the issue presented in that case. Regarding diplomatic protection, the U.S. and the E.U. experiences are relevant, since they both have enacted instruments of diplomatic protection the 'Section 301' and the 'Barriers to Trade' regulation, respectively. Finger and Fung66 examined U.S. 'Section 301' investigations initiated upon complaints by private companies. Out of the dozens of cases that had been filed over the years, they found only one in which a positive '301' determination was based on the existence of anticompetitive practices.67 The 'anticompetitive clause' only became a part of the law in 1988. Nevertheless, some cases had been brought on grounds of anticompetitive behaviour in foreign markets prior to 1988. Other legislative attempts to attack anticompetitive practices in foreign markets have been exclusively intergovernmental in nature (e.g. Super 301), and may not therefore be predicated on narrowly defined market access concerns.68 A comparison of these figures with the number of complaints in other fields (e.g. intellectual property or anti-dumping) strongly suggests that confronting anticompetitive practices in export markets has not been a priority for the U.S. private sector. With regard to the E.U., Mavroidis69 provides a list of all invocations of the E.U. instrument for diplomatic protection to address trade barriers in foreign markets, and finds that no competition-related disputes are reported.
65
See E l e a n o r F o x , supra n o t e 39, at 11. J. Michael F i n g e r a n d K . C . F u n g , Can Competition Policy Control '301'?, 49 AuBenwirtschaft 379 (1994). 67 T h e recently s u b m i t t e d K o d a k / F u j i case is only the second instance where a positive Sec. 301 d e t e r m i n a t i o n w a s m a d e . 68 See N . D a v i d P a l m e t e r , Regulatory Reform in the European Community, 82 A m . Ec. Rev. 98 (1994). 69 Petros M a v r o i d i s , Handelspolitische A b w e h r m e c h a n i s m e n der E . W . G . und der U . S . A . u n d die F r a g e I h r e r Vereinbarkeit mit den G.A.T.T.-Regeln (1993). 66
4 - Competition Policy Objectives
303
b. Bilateral Competition Agreements As stated above, bilateral cooperation agreements in the competitionfieldhave been used more often than the options offered by the trade regime. Wood studied the application of such and concluded that they can provide an adequate means to address distortions from nationally pursued competition policies.70 As further stated above, a word of caution is appropriate when evaluating the application of such agreements to date, given their newness. Nonetheless, such agreements have been used more frequently than the existing G.A.T.T./W.T.O. rules to address competition-related concerns.
2. Interpreting the Sparse Use As shown above, it is formally possible to address many of the distortions set forth in Section A through the existing texts of the G.A.T.T. The reasons that this possibility has hardly been used are not entirely clear. We must first consider whether distortions are important. At present, it is impossible to provide a definitive statement on this issue. Assuming that they are important, Subsection C.2.b examines whether, for systemic reasons (that is, for reasons unrelated to the adequacy of the existing substantive rules), the existing legal regime could be deemed inadequate. a. Distortions are Unimportant We first consider the possibility that, in reality, distortions are not very large, and that, consequently, there has not been a need to use the existing legal regime. While this may explain why the system has not been used on a large scale, it does not explain why there has not yet been adjudication of a single case. In some cases, it is likely that distortions have been large, even if, on average, they are not. Moreover, the very existence of bilateral cooperation agreements in the field of competition suggests that distortions are present. Another reason that distortions may not be important, or may not exist, is that the models which have been developed, and which rely on a number of stringent assumptions, do not reflect the essential aspects of reality. For instance, the models presuppose that governments have sufficient information to identify the sectors and firms where gains are possible from these types of policies. With regard to competition policies seeking to extract foreign firms' profits, this requires not only information about cost conditions in domestic firms, but also about foreign firms whose profits are to be shifted. Dynamic 70
Wood, supra note 53.
304
Working Paper VIII
elements are also missing from the reasoning above, which may render the models of little use. For instance, a prime consideration for governments contemplating the use of strategic competition policy should be whether their aggressive competition policies might not trigger similar responses from other countries, so that the competition policies become more strategically interrelated. Turning to empirical evidence, some studies suggest specific instances when certain countries appear to have pursued a beggar-thy-neighbour competition policy. For instance, horizontal export cartels are not only allowed, they are even encouraged in some countries. For example, Scherer,71 citing Mirow and Maurer,72 notes the Canadian government's actions to induce a Canadian subsidiary of Gulf Oil Corporation to join an international uranium cartel in the early 1970s.73 In the context of mergers, there are many cases where decisions have reflected desires to promote 'national champions'.74 As to vertical arrangements, there are cases where restraints pose important barriers to imports.75 There is also evidence regarding the spillovers of policies dealing with joint ventures.76 In the parallel case of trade liberalisation, many studies have systematically revealed the costs of protectionism, and that, together with basic trade theory, has served as an analytical foundation for the drive for international trade liberalisation. In contrast, in the case of competition policies, there is no welldeveloped body of theory that can be employed when advocating, or devising, an international competition policy agreement.77 The discrepancy between the empirical literature on competition and trade policies is even more dramatic: 71
Scherer, supra n o t e 3. K u r t R. M i r o w a n d H a r r y Maurer, W e b s of Power: International Cartels and the W o r l d E c o n o m y (1982). 73 This case is also interesting because it suggests that in making its decision, the C a n a d i a n g o v e r n m e n t considered the profits m a d e b y a foreign multinational firm. 74 See Frederic Scherer, supra note 3; Neil Campbell et al, The Role of M o n o p o l y L a w s in the I n t e r n a t i o n a l Trading System, P a p e r presented at the Taipai Symposium on C o m p e t i t i o n Policy in a Global Economy, V I I T r a d e Policy F o r u m of the Pacific E c o n o m i c C o o p e r a t i o n Council (Taipei 1995), a n d the references therein. 75 F o r recent evidence, see William C o m a n o r and Patrick Rey, Competition Policy Towards Vertical Foreclosure in a Global Economy, in L e o n a r d Waverman et al. (eds.), C o m p e t i t i o n Policy in the Global Economy, Modalities for Cooperation (1997). 76 See L a r s - H e n d r i k Roller and Mihkel T o m b a k , Exclusive Research Joint Ventures, P a p e r presented at the Taipai Symposium on C o m p e t i t i o n Policy in a Global E c o n o m y , VII T r a d e Policy F o r u m of the Pacific E c o n o m i c C o o p e r a t i o n Council (Taipei 1995). 77 E c o n o m i s t s have initiated research in this area only recently. See, e.g., Keith H e a d a n d J o h n Ries, I n t e r n a t i o n a l Mergers a n d Welfare U n d e r Decentralized Competition Policy, unpublished p a p e r available at t h e University of British Columbia, Vancouver 72
4 - Competition Policy Objectives
'
305
we are unaware of any systematic study of the quantitative significance, in terms of losses in consumer and producer welfare, of distortions or negative spillovers. In sum, there is very little systematic empirical evidence that would allow us to assess the importance of spillovers, let alone distortions, from national competition policy. b. Inadequacies in the Current Legal Regime This section will attempt to explore the systemic reasons that the current legal regime has not been fully exploited. One reason the trade regime has not been utilised is that absence of direct effect has complicated matters.78 However, the limited use of diplomatic protection instruments indicates that this is only a partial explanation. If private parties had locus standi before the W.T.O., it is possible that more cases of the sort submitted to the multilateral dispute settlement system would be brought. A second reason for the lack of competition cases before the W.T.O. is that governments do not view distortions one-by-one, which occurs when private parties bring cases to the W.T.O. Rather, governments take a broader look. A country may be hurt by facing distortions in some sectors, but may gain from beggar-thy-neighbour policies in other sectors. Hence, the reticence to submit some potentially interesting cases to the W.T.O. can perhaps be explained because governments constitute sums of divergent interests. A third reason may be that too much discretion is needed to adjudicate competition related disputes, and governments do not trust W.T.O. dispute resolution panels. Since 'non-violation' complaints can be all-encompassing, some have cautioned that Art. XXIII: l(b) procedures not be overburdened. However, each time a W.T.O. Member takes action (even if G.A.T.T.-consistent) that nullifies or impairs benefits accruing to other parties, and thus defies 'reasonable expectations', the preconditions for the filing of a 'non-violation' complaint have been satisfied. This interpretation is dictated by Art. 31 of the V.C.L.T. (ordinary meaning of the term). Those who might argue that this was not the intent of the drafters of the G.A.T.T. must reverse the order of means of interpretation as embodied in Art. 31, and give priority to the intent of the parties over the ordinary meaning of the term. Such an approach is not
(1995), who investigated the conditions under which individual competition policy authorities would approve mergers that reduce world welfare. 78 For a comprehensive analysis of this issue, see Piet Eeckhout, supra note 22.
306
Working Paper VIII
accepted by public international lawyers.79 This legal possibility can only be meaningfully exercised if panels are entrusted with substantial discretion. However, in practice, governments seem not to trust the existing W.T.O. dispute settlement system with the exercise of substantial discretion. Although a number of significant steps have been taken to establish trust, their likelihood of success is questionable. This is because legitimacy is the basis of trust,80 and ad hoc adjudicating bodies generally lack legitimacy. G.A.T.T. panels, however, have not behaved as ad hoc bodies. They have regularly referred to G.A.T.T. case law, and the establishment of a legal office has helped to create a feeling of permanency and continuity in the G.A.T.T./W.T.O. case law.81 Moreover, establishment of the Appellate Body (A.B.) suggests that this trend is to be maintained.82 However, it seems that this is not enough: there is an undeniable discrepancy in the G.A.T.T./W.T.O. regime between the possibilities offered by the legal texts and the manner in which such possibilities can be adequately served by the existing institutional infrastructure. Notwithstanding the recent positive modifications, the current institutional infrastructure is still far from permitting a full exploitation of the possibilities provided by the current legal texts. Finally, the W.T.O. dispute settlement system does not have a fully developed procedural law. Some of the existing procedural impediments must be overcome for competition related disputes to be treated adequately.83 In addition, the G.A.T.T./W.T.O. legal regime is moving from negative to positive integration. The conclusion of the Agreement on Trade Related Intellectual Property Rights (T.R.I.P.s) is an excellent illustration of this phenomenon. A growing number of obligations contracted under the W.T.O. auspices require positive action by the W.T.O. Members, and negative obligations are disappearing. This does not mean that the traditional trade agenda has been completed. On the contrary, slow progress with respect to agriculture and 79
Jimenez de A r e c h a g a , I n t e r n a t i o n a l L a w in the Past Third of a Century, 159 Recueil des C o u r s 1, 4 2 - 4 8 (1978), states t h a t t h e message of the V.C.L.T. is that the a p p r o a c h a d v o c a t e d b y t h e p r o p o n e n t s o f t h e ' i n t e n t i o n s of the parties'-school was p u t aside. 80 See T h o m a s M . F r a n c k , supra n o t e 2 1 . 81 See R o b e r t E. H u d e c , Enforcing I n t e r n a t i o n a l Trade L a w (1993). 82 T h e A B deals only with legal q u e s t i o n s a n d therefore, de facto, its rulings will have some precedential value. 83 Petros C. M a v r o i d i s a n d Sally V a n Siclen, The Application of the G.A. T. TAW. T. O. Dispute Resolution System to Competition Issues, 3 1 J. W o r l d T r a d e 5 (1997) (identifying five specific areas, namely standing, confidentiality, expertise of panels, standard of proof a n d remedies, where additional w o r k is n e e d e d ) .
4 - Competition Policy Objectives
307
textiles, among other sectors, indicates that the agenda is still unfinished. Thus, the W.T.O. is at a crossroads: on the one hand, it must complete the unfinished business; on the other hand, it is 'pushed' to the realm of positive integration. This can prove to be a tough test, and is precisely what the E.U. experience can offer as an example. The current inadequacies in the institutional infrastructure do not favour additional burdens. Rather, the strengthening of the W.T.O. infrastructure should be a priority before any new negotiations on substantive issues occur. c. But What About Telecoms? The conclusions of this paper seem to be contradicted by the recently concluded W.T.O. Agreement on telecoms, where sector-specific common regulatory principles were used to liberalise the world telecoms market.84 Once basic telecommunications were on the negotiating table at the W.T.O., negotiators were compelled to decide whether and how to deal with the determinants of market access attributable to firm behaviour, which normal G.A.T.T./W.T.O. access commitments would be unable to reach. Broadly, three options suggest themselves. First, the 'harmonisation' approach would involve agreement on uniform rules at the multilateral level. Secondly, the 'procedural obligations' approach would require governments to accept a multilateral obligation to enforce their own rules, without any obligation to make changes in substantive or procedural requirements at the national level. Thirdly, the 'do nothing' approach would require that negotiators restrict themselves to market access and national treatment commitments, thus relying (implicitly or explicitly) on independent national regulatory interventions and/or market forces to ensure that markets are contestable. The harmonisation approach has the advantage of guaranteeing legal security, but agreement may be difficult to obtain. The procedural obligations approach presupposes acceptance by W.T.O. Members of each other's national regulatory or competition frameworks. The do nothing approach assumes either that national regulatory or competition frameworks are acceptable and will be appropriately enforced, or that the degree to which the relevant market can be contested is sufficient for aspiring market entrants. The harmonisation approach was selected by W.T.O. negotiators. It was, however, constrained harmonisation, in that not all the W.T.O. members intend to subscribe to the regulatory principles, and certain modifications will be made by 84
See Barnard M. Hoekman, et al., Regulatory Competition Policy and Market Access Negotiations, in Competition and Trade Policies, Convergence or Conflict? 115 (E. Hopp and P. Mallary eds. 1998).
308
Working Paper VIII
some parties to the common text on regulatory principles that has been negotiated. Nevertheless, negotiators in the W.T.O. process chose an option at the higher end of the cooperative scale. This raises the question whether those emphasising the difficulty of reaching international agreement on substantive competition policy rules have exaggerated this argument. The procedural obligations approach was not favoured precisely due to the substantial differences among domestic antitrust and regulatory regimes, and the absence of 'trust' among the various jurisdictions. The do nothing approach, on the other hand, is problematic in an area where the addressees of the international commitments are, in most cases, not governments, but privatised giants. Two reasons may be advanced why a harmonisation approach was possible in basic telecoms. First, the negotiations are taking place at a time when many governments are moving in the same direction, and seeking to attain the same outcome, namely opening up their basic telecom markets to competition. The effort to liberalise via joint action in these circumstances probably made it possible to go further than would have been the case in a less dynamic setting. Secondly, the negotiations involve very specific issues in a single sector. Governments are unconcerned about the implications of what they are doing in telecoms for what they might wish to do elsewhere. Thus, it is easier to cooperate in a narrowly-defined sectoral context to design common regulations than it would be to draw up uniform competition principles of general application. Finally, the trading partners implicitly rejected the option to negotiate broad common competition rules, since they did not embark on a negotiation on Arts. VIII and IX G.A.T.S. By choosing negotiations on sectorspecific regulatory principles, they thus recognised the particular aspects of telecoms.
Conclusion The paper has argued that while national competition policy is likely, in many cases, to have negative consequences for trading partners, this does not warrant an international agreement. For there to be scope for gains from such an agreement, it is also necessary that decisions that seek to promote the national interest also imply distortions in the global allocation of resources. Unfortunately, there is little systematic empirical work that provides any guidance on the magnitude of spillovers in general, and, even more troubling, as to the magnitude of possible distortions. Therefore, at the moment, we are left with the possibility that such distortions could represent either a small or a
4 - Competition Policy Objectives
309
large problem. Of course, if they are small, there is no good reason to negotiate a multilateral agreement. In contrast, if they are large, and since we have shown that the existing legal regime can in principle deal with most such problems, the existence of significant distortions would strongly suggest that the existing legal regime is unsatisfactory. Section C suggested several possible reasons that the current regime has not been used, but it is not clear which explanation is correct. From a policy point of view, this means that it is premature to propose a multilateral agreement. There are still too many uncertainties concerning the size of the distortions problem, and the sources of the lack of trust in the existing legal regime. Before these uncertainties are resolved, it seems inappropriate to advocate one solution as superior to all others.
Other References NEIL CAMPBELL and MICHAEL J. TREBILCOCK, International Merger Review: Problems of
Multi-Jurisdictional Conflict, in Competition Policy in an Interdependent World Economy, (Kantzenbach, Scharrer and Waverman eds., 1993). ANDREW DICK, Japanese Antitrust: Reconciling Theory and Evidence, 9 Contemporary Pol. Issues 50 (1993). SIR GERALD FITZMAURICE, The Law and Procedure of the International Court of Justice (1986). BERNARD M. HOEKMAN and PETROS C. MAVROIDIS, Competition, Competition Policy and
the G.A.T.T.,\1 World Econ. 121 (1994). KAI-UWE KUHN et al, Competition Policy Research: Where Do We Stand? C.E.P.R. Occasional Paper, No. 8 (1992). DAMIEN NEVEN and PAUL SEABRIGHT, Trade Liberalisation and the Coordination of
Competition Policy, in Competition Policy in the Global Economy, Modalities for Cooperation (Leonard Waverman et al. eds., 1997). MICHAEL PORTER, The Competitive Advantage of Nations, (1990). FRIEDER ROESSLER, The Concept of Nullification and Impairment in the Legal system of the World Trade Organisation, in International Trade Law and the G.A.T.T./W.T.O. Dispute Settlement System, (Ernst-Ulrich Petersmann ed., 1996).
IX Dr. Frangois J.L. Souty1 Rapporteur charge des affaires multilaterales Conseil de la Concurrence Paris, France Associate-Professor, University of La Rochelle
Introduction One of the major challenges in the W.T.O.'s future activities will be to unify sometimes varying conceptions of what Europeans call 'Competition policy' and Americans call 'Antitrust policy'. Even if one accepts the most common concept of 'competition policy,' different views still exist as to the significance and coverage of that policy. One way to reconcile the two concepts would be to emphasise the converging views of how to increase market efficiency at the international level. Even our understandings of market dynamics vary from country to country. In some countries, such as the United States, significant changes have occurred. Recent Chicago Schoolfindingsreflect these changes, following a succession of antitrust schools, since enactment of the world's first modern and original antitrust laws. In other countries, there is greater emphasis on the notion of 'fair trading' rather than competition. This is the case, for instance, in the United Kingdom and Japan. In Europe, a process of convergence has occurred over the last several decades among a group of countries with very different legal and economic traditions. The European Union has promoted a strong approach to competition policy, as defined in Arts. 85 to 94 of the Treaty of Rome. In the future, it may be useful to examine the extent to which the E.U. Member States have been able to converge in the area of competition policy. After World War II, the European Coal and Steel Community, and later the European Economic Community, engendered a complex process of legal and economic convergence among Member States, the effects of which can be seen at both national and Community levels. The complexity of the Community's competition regime lies both in the institutions of competition law and in the objectives pursued by those institutions. 1
The views expressed in this paper are the author's own and do not reflect any official position. Thanks to Ann Ayers for comments and suggestions.
312
Working Paper IX
After discussing the above areas in further detail, this paper will consider cartels, the sole area where a consensus could be reached in the W.T.O. context, since competition policies worldwide are all basically designed to fight against cartels. A. A Process of Convergence in Europe from an Institutional Perspective Over the last thirty years in Europe, a set of competition policy regimes has emerged. These regimes encompass general competition laws and specific authorities or bodies devoted to the implementation of those laws. The Treaty of Rome has contributed to this general trend, and the system thus created was modernised in the 1980s. The paradigm of'Competition Law and Policy' is distinctive, and determines the quality and authority of those institutions that have the duty to analyse the evolution of market structures and to preserve their efficiencies by eliminating anticompetitive behaviour. The experience of the Community and the three Member States with the oldest competition regimes (Germany, the U.K., and France) illustrates these developments. Germany has an independent high federal cartel authority, the Federal Cartel Office (F.C.O.), which decides on the validity of behaviour, subject to review by the Court of Appeal of Berlin. The F.C.O. is independent in the sense that the Ministry of Economics, which has jurisdiction over the appointment of the Chairman of the F.C.O., may not give the F.C.O. administrative guidance or direction with respect to specific cases. Moreover, the F.C.O. makes decisions via a collegial process in its divisions. Independence and collegiality are the two main features of this institution. In addition, a 'toothless watchdog' has been created: the Monopolies Commission. It plays an expert and advisory role with regard to the merger control and abuse control activities of the F.C.O. and the Minister of Economics, but does not have enforcement powers. This expertise is the third feature of the German enforcement system.2 The United Kingdom also has a tri-polar system of control with regard to its competition and fair trading laws. The Office of Fair Trading is an independent agency outside the monitoring power of the Minister of Trade and Industry, although each of its decisions is subject to review by the Department of Trade and Industry, especially in the area of mergers. The two institutions are likewise supplemented by a 'toothless watchdog' acting in an expert capacity, 2
Ingo L.O. Schmidt, Wettbewerbspolitik und Kartellrecht : Eine Einfiihrung (5th ed. 1996); see Francois J.L. Souty, La politique de la concurrence en Allemagne Federate (1996).
4 — Competition Policy Objectives
313
primarily with regard to mergers and monopolies. This watchdog is called the Monopolies and Mergers Commission. Its reports are submitted to the Director General of Fair Trading following a collegial deliberation. The basic features of independence, collegiality and expertise are, therefore, present in this system as well. The only missing feature in the British system concerns, so far, the role of the judges and the absence of a dissuasive fining system.3 Over the past ten years, France has gained similar experience through its Ministry of Economics, commonly seen as the major investigating body, acting through its directorate for competition, the Direction Generate de la Concurrence, de la Consommation et de la Repression des Fraudes ('D.G.C.C.R.F.').4 While the D.G.C.C.R.F. is primarily responsible for investigations, cases involving prohibited behaviour must be referred to the Conseil de la Concurrence, which is a collegial board that renders judgments as a specialised tribunal composed of senior magistrates who impose orders and fines upon corporations. The decisions of the Conseil de la Concurrence are subject to appeal in the Court of Appeal of Paris. The Conseil can institute a procedure sua sponte and request that the Ministry of Economics undertake the necessary investigations. With respect to mergers, the Conseil de la Concurrence becomes a 'toothless watchdog' because it can only give its advice to the Ministry, which may choose to follow it and which can decide to publicise the decisions or keep them secret. Again, independence combined with specialisation, collegiality and the toothless expertise drive this system.s The European Community seems to have promoted these basic features of independence, specialisation and collegiality. The only feature missing from the European system is the 'toothless watchdog'. The E.U. system has been severely criticised, especially with regard to the independence of the Commission. However, Commission decisions are subject to several important controls. First, from the outset, the Commission's decisions have been closely scrutinised by a very independent authority: the European Court of Justice (currently the Court of First Instance). Secondly, the E.U. Commissioner in charge of competition policy benefits from a large delegation of fellow Commissioners, and is aided by the independent body of DG IV, which is itself assisted by consultative committees of national experts representing the expertise of their national agencies. Hence, here again, the idea of a body vested with 3
Richard Whish and Brenda Suffrin, Competition Law (1993); Francois Souty, La politique de la concurrence au Royaume-Uni (1996); significant legislative changes are expected for 1999. 4 Ordinance no 86-1243 (1 Dec. 1986). 5 On French competition law and policy, see Guy Canivet and Marie-Chantel Boutard-Labarde, Droit francais de la concurrence (1994).
314
Working Paper IX
a specialised authority, a specific expertise and collegiality controls the system to enforce competition law.6 If one focuses on purely institutional aspects, these four experiences follow a general pattern modeled on the U.S. matrix, incorporated more in the Clayton Act7 and the F.T.C. Act8 than in the Sherman Act.9 The former include a specific set of legal prohibitions defining extensively and in detail the economic order to be protected: a competition system based on a free market and an efficient economy. They also establish the tools to ensure such protection: specialised authorities under the supervision of the federal court system. By comparison, the Sherman Act simply defined a set of prohibitions to be enforced by common law judges. The intervention of the Department of Justice and its specialised unit, the Antitrust Division, emerged more than a decade after the signing of the Sherman Act. In contrast, the Clayton Act coincided with, and the F.T.C. Act gave birth to, the specialised authority with responsibility to implement the law, the F.T.C.
B. European Competition Policies are Converging to a Limited Extent Regarding Objectives We are currently faced with the issues of whether a detailed multilateral code should be entered, and whether an autonomous enforcement and adjudication system should be created; or, on the other hand, whether a bilateral or multilateral agreement should be entered specifying a few consensus principles of constitutional generality, which would be implemented and enforced at national level. The examples discussed above provide some guidance. To a limited extent, European competition policies have been converging. This process has been facilitated by the definition of a core of constitutional principles at the Community level, which has allowed Member States to agree to shared enforcement, while maintaining a certain degree of autonomy of E.U. law preempting national laws. The German experience is closely associated with a strict efficiency balance, existing within the framework of a workable competition concept. This concept has only been adopted following an animated debate involving the business community in the 1950s. However, in 1869, the German business 6 On E.U. law, see Whish and Suffrin, supra note 3; Franpois Souty, Le droit de la concurrence de l'Union Europeenne (1997). 7 15 U.S.C. §§12-27. 8 15 U.S.C. §§41-58. 9 15 U.S.C. §§1-8.
4 - Competition Policy Objectives
315
community had been given full freedom of trade within the Zollverein, that is, without interference by governmental licence or permission. The creation of a nation-state where the same rule applies, notwithstanding the Land in which a firm operates, dates from the same period. Over the ensuing 100 years, the German business community learned to operate freely within Germany, with the notable exception of the Third Reich. The paradigm 'free operation' must, nevertheless, be cautiously considered, since co-ordination between operators, and even cartelisation, was widely admitted by the legal and business traditions during the last decades of the nineteenth and the first decades of the present century. Italy's experience was similar to Germany's, specifically with respect to the interstate free trade implemented by the Zollverein.10 In any case, the intricate legal and political processes have permitted successful economic integration stemming from a plurality of former sovereign states and juxtaposed economies. The present situation in Germany was initiated in 1957 with adoption of the Gesetz gegen Wettbewerbsbeschrankungen (G.W.B.).11 Since then, the long term process of integration of the German economy has forced a transfer of sovereignty from the classical states to the Federation, be it an empire or a republic. At the centre of the workable competition model, the federal cartel authority is vested with the mission of protecting the efficient competitive order of the market by focusing its enforcement actions chiefly on the restrictive practices of cartels and those deriving from some market-dominating corporations. This enforcement is backed by a powerful fining system, adhering to the German tradition of a strong imperial state, but under review of an appellate court. The British experience is different, largely reflecting the habeas corpus tradition which characterises the role of the state and severely limits the sovereign's powers with regard to individuals. In the United Kingdom, competition policy emerged in a mature economy in 1948, as a Labour reaction to a very highly concentrated economy and the exercise of market power, harmful to British consumers. This occurred with the support of the U.S. antitrust authorities, who were then seeking to dismantle global cartels involving major British and American corporations, on the heels of dismantling cartels in Germany and Japan. The British law followed the public interest approach, although this 10
Thus, the Italian legislative and business community may also be familiar with the association of two concepts related to sovereignty transfer on the one hand and creation of a unified free trading market on the other. 11 Gesetz gegen Wettbewerbsbeschrankungen (Kartellgesetz) in der Fassung des 5. Anderungsgesetzes vorn 22 December 1989 (B.G.B.I. 1989, Teil 1, 5.24.86).
316
Working Paper IX
public interest is subject to varying interpretations. The most recent understanding of the public interest relates to efficiency aspects, as in Germany. Despite regular updating of the British competition law, however, enforcement authorities have never been granted sufficient sanctioning powers. Further, in the U.K., the competition authority has never had to share power with local authorities. Thus, it is not in the constitutional tradition to transfer sovereignty. The only break from this tradition has been participation in the European Community, which implies the acceptance of the acquis communautaire. In France, a strongly centralised state has not always been taken for granted. Although there has never been an experience similar to that of Germany, some intense provincial or regionalist feelings exist, making the French version of subsidiarity a form of regionalism. With respect to the market and economic law, a heavy industrialist influence was felt in France until the late 1980s. The French understanding of competition policy, therefore, is associated with the notion of a bilan economique, although the Conseil de la Concurrence has retained the bilan concurrentiel analysis. The latter is a concept in which efficiency considerations prevail, whereas the former encompasses various political and social objectives. As in Germany, there is no tradition of transferring sovereignty in France. However, acceptance and support for construction of the European Community by the French can be traced to such founding fathers as Robert Schuman and Jean Monet, who familiarised the French institutions with the intricacies of a coexistence of a supranational legal framework integrated into a national legal tradition. This acceptance of European integration has enhanced greatly the bilan concurrentiel school. Still, the existence of two paradigms of competition policy enforcement, influencing the reasoning either with purely efficiency-related concepts or broader concepts, raises some uncertainty with regard to the standards. This is especially perceptible in the merger control area, where the Conseil de la Concurrence gives opinions to the Minister of the Economy which are grounded in the bilan concurrentiel approach, whereas the Ministry's own opinions remain determined by the bilan economique approach. As the case law evolves, the applied standard is likely to become more uniform, and judging from the standards of analysis of the Conseil, the trend is clearly in favour of the efficiency conception {bilan concurrentiel). Finally, in constructing the Community competition law and policy, the European Community has had to take note of the experiences of the founding fathers. This was accomplished by elaborating a policy aimed at harmonising market institutions throughout the Community, eliminating barriers to free trade among the Members States (namely antidumping and other trade-related
4 - Competition Policy Objectives
317
instruments), and submitting state aid and public subsidies to the monitoring power of the Commission. The process was supplemented by the Member States' delegation of sovereignty to the Commission. Two phenomena were induced by the establishment of this process. First, as noted by two European scholars in a recent paper, 'this competition policy, enforced centrally (from Brussels) and implemented in the individual member nations, led to vertical convergence, voluntarily accepted and enforced under law'.12 Secondly, this policy also led to a horizontal convergence that is one of the characteristics of the development of European Law. This was made possible under the auspices of the E.U. Court of Justice, with its case law creating and incorporating the 'general principles of European law' ('principes generaux du droit europeen'), which are comprised of a sampling of common law principles already established in the Member States. This second convergence process moves from the centre to the periphery, in the sense that Commission decisions are imposed on Member States and their nationals, or that the Commission delegates its powers to Member States; and from the periphery to the centre, in the sense that national problems and problems affecting two or more states are referred to Brussels for treatment.13 Horizontal convergence has also led to a narrowing of the scope of national legislation pre-empted by the directly applicable Articles of the Treaty of Rome and regulations derived therefrom. This can be viewed as a consequence of the transfer of sovereignty. E.U. competition law and policy are, therefore, the result of a compromise or trade-off between efficiency and integration. Within the European Community, the convergence process was sometimes more apparent than real, and implied choices of considerable political consequence.
C. International Consensus is Possible With Respect to Treatment of Cartels The Community system and the three national systems discussed above prohibit collective anticompetitive practices. However from a strictly economic and legal formulation, important differences remain even in the clear-cut area of cartels, and how they should be denned and treated.
12
Herve Dumez and Alain Jeunemaitre, The Convergence of Competition Policies in Europe: Internal Dynamics and External Imposition, in National Diversity and Global Capitalism (Berger & Dore eds., 1996). 13 Id.
318
Working Paper IX
Section 1 of the German Competition Law of 1957 prohibits cartels. Following a long controversy with the Supreme Court, however, a new prohibition regime has been introduced by the legislators (section 25(1)), which closely resembles Art. 85(1) of the Treaty of Rome. Cartels are defined as agreements between enterprises or associations of enterprises, which are concluded for a joint purpose and may influence production or normal market conditions for trade in goods or services by restraining competition. The same rule applies to concerted practices that have a purpose and effect similar to that of a cartel. The basic prohibition of cartels is, however, qualified by a number of exemptions. In addition to the classical special exemptions for agriculture, utilities, transports, etc., the F.C.O. is authorised to grant exemptions under strictly defined conditions to cartels which are notified. Exemptions are granted mainly to S.M.E.s (more than 100 specialisation and cooperation cartels in the early 1990s) and to export cartels (around forty notified cartels under section 6 G.W.B. in the 1990s).14 Unlike Art. 85(3), the German exemption provision does not include a general rule of reason and does not grant broad discretion to the F.C.O. to lift the ban where economically appropriate. This is sometimes perceived by the business community as a structural deficiency of the German law, felt particularly in joint venture situations, causing corporations to seek clearance under E.U. rules. Furthermore, illegal cartels are fined even more heavily in Germany than in the United States (fines up to three times the amount of illegal profits in Germany as compared to twice in the U.S.). In the United Kingdom, two laws may allow the Office of Fair Trading to prohibit prima facie cartels: the Restrictive Trade Practices Act 1976 (R.T.P.A.) and the Competition Act of 1980. The Fair Trading Act of 1976 allows investigations when a 'complex monopoly situation' is created by a number of separate companies. No clear-cut definition of anticompetitive agreements exists, however, in British competition law as it does in Art. 85(1). In cases where anticompetitive behaviour has been established, the only action that British competition authorities can take is prohibition. However, they have no power to penalise past behaviour or provide relief to injured parties. Further, these parties have no right to sue for damages. From an economic point of view, the competition laws prohibit agreements between companies which together account for at least 25 per cent of the relevant market in the United Kingdom. The R.T.P.A. is the only legal tool creating the right to seek a contempt of court order for continuing prohibited behaviour, but a defendant can seek relief from the prosecution by using a series of tools ('gateways'
14
For a table of statistics between 1958 and 1993, see Francois Souty, supra note 2.
4 - Competition Policy Objectives
319
and 'tailpieces') analogous to exemption provisions. In practice, this law is rarely used to sentence anticompetitive behaviour. After a long period of delay, the French competition law was introduced in its present form in 1986. Art. 7 of this law contains a clear definition of cartels, modelled on Art. 85(1), which states: 'any concerted practice, whether express or tacit, is prohibited when it has the purpose or effect of preventing, restricting or distorting competition in a specific market.' Unlike Community law, it does not contain a notification provision. The Ordinance is, in principle, enforced by the Ministry of the Economy, the Conseil de la Concurrence and common law tribunals. In reality, however, it is enforced primarily by the Ministry of the Economy's D.G.C.C.R.F, which is allocated the necessary resources to conduct investigations. The Ministry may also refer any case to the Conseil, unless it is covered by a de minimis rule or a clearance procedure. To avoid lax enforcement, all results of proceedings undertaken by either the Ministry or the Conseil under authority of the Ordinance must be published. Fines of up to 5 per cent of the firm's total domestic sales may be imposed. The Conseil has significantly increased the amount of fines in recent years (more than 100 million FF each year, 478 million in 1995). Finally, the European Community's cartel prohibition rule must be considered, as it is one of the best models for the future of competition law in the W.T.O. Indeed, it is the only experiment where several states have agreed to prohibit cartels affecting more than one national economy. This model, however, is not appropriate for adoption by the global community of states although it will inevitably gain broader support as it is adopted in Eastern Europe and Northern Africa. The prohibition articulated in Art. 85(1), and the exemption created by Art. 85(3), are well known. Under Reg. 17, the Commission is empowered to impose fines on cartels after first consulting with the Member State Advisory Committee. As a result, condemnation of cartels in the E.U. is the result of an objective, transparent, coherent and consistent law enforcement process that has been more consistent over time than has been cartel enforcement in non-European countries.
Conclusion First, it is possible to internationalise competition law, as the European integration process in the face of diverse national legal systems has shown. Second, as to whether a detailed multilateral code with an autonomous enforcement and adjudication system should be adopted, or a less detailed set of principles of constitutional generality and contemplating national imple-
320
Working Paper IX
mentation and enforcement, Europeans have nothing to fear by supporting the first alternative. History has nevertheless taught us that competition enforcement rarely starts with detailed codes. Instead, several decades and legislative amendments have often been necessary to build the systems of competition policy presently operating in most major economies. Third, a synthesis of the two alternatives and composed of some common features from the examples above may provide a preferable third solution: a limited prohibition which allows national enforcement under the review of a collective and autonomous body, advised by an independent toothless watchdog with a strong expertise in the field of law and economics, and subject to review by an appeal authority.
PANEL DISCUSSION
COMPETITION LAW IMPLEMENTATION AT PRESENT
GENERAL RAPPORTEUR:
Barry Hawk, Partner, Skadden Arps Slate Meagher & Flom, New York, New York, U.S.A.
PARTICIPANTS:
Ian Forrester, Partner, White & Case, Brussels, Belgium Calvin Goldman, Partner, Davies, Ward & Beck, Toronto, Ontario, Canada Herbert Hovenkamp, Professor, University of Iowa College of Law, Iowa City, Iowa, U.S.A. Martin Howe, Former Deputy Head of Office of Fair Trading, London, England Abbott (Tad) Lipsky, Senior Competition Counsel, The Coca-Cola Company, Atlanta, Georgia, U.S.A.
5 - Competition Law Implementation at Present
323
• PROF. HAWK—Session three will attempt to follow David Hume's distinction between the ought and the is. We will focus on the 'is'—i.e., how policy objectives are implemented in the various jurisdictions. I would like to list nine issues for discussion, which are covered in the papers submitted by the panellists of this session. First, there is a considerable gap between the rhetoric of competition law objectives and the reality of their implementation. Canada and the United States provide two examples of this phenomenon. Why is there this gap? I think much of the reason is a failure to distinguish between the objectives of competition law and the asserted benefits of competition. For example, one asserted macroeconomic benefit of competition is reduction of inflation, but this has little, if anything, to do with competition law after it has been enacted. This reminds me of the excellent distinction made by Mr. Schaub between fundamental, ultimate objectives, and operational enforcement criteria. Although they do not use Mr. Schaub's distinction, all of the papers submitted by the panellists of this session say essentially the same thing. Second, in many jurisdictions, including Canada and the United States, there is a strong contrast between the objectives that are expressed in the statute and the narrower interpretations by the competent authorities. For example, many of the Canadian/U.S. statutory provisions, like the ban on price discrimination, expressly invoke populist, non-economic objectives. In general, those statutes are weakly enforced by the competent authorities. On the other hand, courts appear to be more willing than the competent authorities to follow the statutory objectives when faced with a claim, probably because courts have more limited discretion than a competent authority has. If it disagrees with the purpose of the statute, a competent authority has considerable discretion simply not to enforce the statute. Third, a trend has developed to narrow competition law objectives, primarily to a more economics based approach that accords either exclusive or great weight to welfare/efficiency at the operational level. Distinctions between consumer welfare, producer welfare, etc., have no effect at the working level. However, the papers indicate that there is a very strong trend toward far heavier emphasis on economics-based considerations, which rests upon increasing emphasis on economic analysis. This has had a number of effects, one of which is the non-enforcement, or very weak enforcement, of certain legislation that has broader, non- economic objectives. Fourth, this narrowing of objectives does not result simply from a political disagreement with the legislation. This is very important. Prof. Fox makes the point that choosing to pursue efficiency is a political judgement. Some have argued that Prof. Easterbrook and others are simply Chicago School economists, and are rejecting populist goals for political reasons. However, that is
324
Panel Discussion
not the full picture. Prof. Easterbrook and others are also saying that competition enforcement should be limited to economics for lawyers' reasons, not just for political reasons, as economics provides better analytical tools to produce more predictable results than broader political objectives. The constraints on operating a legal system encourage a narrowing of objectives to economics. That is not a political judgement; that is a lawyer's judgement. Fifth, Prof. Easterbrook would say that an economics-based policy objective has the additional benefit of limiting the discretion of the decision-maker. We may not want to limit the discretion of the decision-maker, but if we do, then narrowing the objectives will help. Despite the debates on the uncertainties, economics at least provides firmer analytical rules to decide a case, and maybe formulate legal rules, than any alternative. Sixth, the E.U. system is the only one with a market integration objective. This is not merely an ultimate criterion; this is an enforcement criterion that results in per se prohibitions on certain kinds of vertical restraints, geographic price discrimination, etc. That, of course, raises tensions between this market integration goal, be it political, social, or something else, and the science of microeconomics. It may be possible to establish an efficiency justification for market integration, but it would take creative thinking. Seventh, the papers seem to suggest that the E.U. has one of the broadest lists of competition policy objectives. The U.K. is an exception, with its 'public interest' test, which is the ultimate ambiguity. Mr. Kobayashi's paper on Japanese antitrust law sets forth a broad list of objectives, but it is unclear whether the trend is the same in Japan that, notwithstanding these broad statutory objectives, the actual implementation is becoming more narrowly focused. Eighth, inclusion of non-economic objectives or considerations as enforcement criteria is intimately related to the institution and the enforcement system being used. Environmental, regional, or social considerations, and concern for national champions or competitiveness, are certainly not microeconomic considerations. As a lawyer, I believe that such considerations are distantly related to the normal microeconomics framework used in analysing cases. Centralised executive or administrative bodies are better able to implement broad, noneconomic policy objectives. In Europe, the Jacobin School of antitrust systems holds that the longer and broader the list of policy objectives, the more power and discretion should be accorded to a central 'integrated' body. For instance, a Commission argument in favour of maintenance of its monopoly over the power to grant exemptions under Art. 85(3) is that the Commission makes its exemption decisions on the basis not only of economics, but also on other regional, social, etc. considerations. The more that non-economic objectives
5 - Competition Law Implementation at Present
325
are included as enforcement criteria, the easier it is to support a centralised, non-independent authority. Finally, legal certainty for me is like Liberty with an initial capital: never have so many sins been committed as those committed in the name of Liberty. Many sins are committed in the name of legal certainty, which is used as an excuse to resist the introduction of economic analysis into antitrust. I believe it is a wildly exaggerated concern. In the U.S. and Canada, where there are many open ended rules based on economic analysis, I question whether there is less legal certainty. • MR. FORRESTER—Being asked to find objectives in European competition law is as easy as being asked to find tasty vegetables in the market-place in Florence: there is a huge choice, it is very rich, very agreeable, there are many interesting things. I was embarrassed to hear that Mr. Schaub regarded my article as enumerating a chamber of horrors in that the Commission seemed to be pursuing so many different goals. I did not view it this way. Rather, the Commission is a relatively young but exceptionally successful competition authority. Therefore, it is natural that it shouldfindjustifications for its actions in many exotic corners. That really does not trouble me. Second, many discussions have occurred over the years on the subject of the market integration goal. The preoccupation of the Commission with market integration has been criticised by most people around this table, either publicly or privately, including myself. However, the pursuit of the goal of eliminating obstacles to cross-border trade was not, in itself, undesirable. Rather, people criticised the relative standard. Why should a modest restriction of cross-border trade be penalised so heavily, while cartels were treated more gently? Correspondingly, it was said that the Commission was too harsh on restrictions of trade, too soft on other things. That does not mean that the essential principle of encouraging cross-border trade, and using the competition rules as a mechanism to this end, was bad. More generally, the Commission has used cross-border trade as an independent mechanism for stimulating competition. Many questions have arisen over the years regarding how far that argument would hold. What about trade with Sweden, when it was not in the Common Market? What about trade with Portugal? What about trade with Poland, Hungary, countries of Eastern Europe, as well as restrictions on transatlantic trade? When the dollar is low, are restrictions placed on a U.S. dealer network with respect to transatlantic shipments caught by Art. 85(1)? Would the countries of central and eastern Europe be interested in using or at least exploring the possibility of cross-border trade as an independent mechanism for encouraging competition in their jurisdictions?
326
Panel Discussion
At what point are we going to draw the line and say that cross-border trade may be insisted upon only up to this point? Third, the Commission should be given great credit, which it does not often receive, for recognising and attacking government distortion of the marketplace. That is almost unique to the European Commission. The law has not been perfectly consistent. The European Court sometimes has said that where parties get together and the government then ordains what they have agreed privately, that is caught by Art. 85(1). On other occasions, the Court has been more cautious and said that the competition rules do not apply under such circumstances. Accordingly, there are inconsistencies, but the principle that state distortions of the market-place are subject to the competition rules is extremely important. It has not been viewed as a problem in the United States, although I believe it is more of a problem there than American practitioners have perhaps recognised. It has been a problem in Europe and other countries around the world. Thus, this is a field in which the European experience should be of great interest, and from which other jurisdictions should borrow boldly. Fourth, in the enforcement of competition law in the European union, there is a strong element of moral righteousness. One might almost say it is a law against unfair competition. There are instances where a complainant presents to the Commission something that seems shocking in its unfairness, so that the Commission has a strong temptation to intervene on behalf of the party. That bears some reflection, because it suggests that even a sophisticated, mature, senior competition agency may be readily distracted from competition law into unfair competition law. Finally, there is no point in having a competition law unless it is actually implemented and enforced. We may have a beautiful, perfect competition law in Ruritania, but if there is no agency to enforce it (as indeed there is no agency in Belgium at the moment to enforce competition law because the officials have all resigned), there is no point in having the law. It's nice, it's there, you can read it, but it doesn't do any good. European law has developed in a pragmatic manner. There is a great deal of law, with a great deal of influence. I submit, however, that its weakness is a failure systematically to address an objective that has not been properly articulated: that the law should be observed and enforced. There is not enough output. Thus, although we have a rich law, we have extremely unpredictable circumstances with respect to the observance and enforcement of the law. That is a weakness that should be corrected in the years ahead. • MR. GOLDMAN—I would like to make six relatively brief points. First, the expressed statutory objectives contained in Canada's antitrust law have played a useful role in the enforcement of that law in three respects. One, from the per-
J - Competition Law Implementation at Present
327
spective of those at the enforcement agency, the objectives help in defining case priorities for the allocation of limited resources. Two, from the perspective of the private sector, the objectives of the Act and the experience under the Act tell a great deal about what kind of submissions are going to be more important. For example, in Canada, it would not be useful to focus on employment issues or on trying to create a national champion, since the law provides for national treatment and nothing else. Three, from the perspective of the adjudicator (the courts or the tribunal), the objectives assist them in trying to interpret the intention of parliament. The objectives do not remove the legal uncertainty at all times; nothing can. However, they provide useful guidance. Second, there is a definite trend in laws such as Canada's to incorporate more economic principles. We have an explicit set of provisions in the 1986 merger law on barriers to entry and innovation, as well as an express statutory efficiency gains defence. They reflect economic thinking and, again, they help to remove uncertainty, but they do not cure it. Third, there is a definite need for greater transparency in competition law at all times. Transparency assists in achieving the objectives, or at least makes the antitrust agency accountable as to whether it has achieved those objectives. Transparency also increases confidence, and educates the public, including the business, legal and consumer communities. When I first went into office, I decided to release detailed statements on cases that we did not challenge, as to why we had decided not to challenge them. Fourth, I tend to agree with the statement in Prof. Hawk's paper, that 'legal certainty is often a shibboleth'. Despite all the express provisions in legislation, legal certainty is like beauty in the sense that it is elusive, it can always be subject to peoplefindingflawsin it, and it does not necessarily get better with age. For instance, Canada had a conspiracy provision that was over 100 years old, and still subject to court challenge on the ground of being void for vagueness. It was set aside a few years ago at the trial division by a judge, but was ultimately resurrected in the appellate proceedings. Thus, anything can happen. Fifth, we must consider how to deal with conflicting competition policy objectives. Objectives are often not compatible. Even with the greatest economic thinkers, some parliaments write objectives that do not always seem to be consistent. The key to dealing with this problem is the balancing process. Comparing harms from anticompetitive acts to efficiency gains is like comparing apples and oranges. In the balancing process, the key phrase is relative certainty, that is, balancing how certain the likely efficiency gains are against how certain the likely anticompetitive harms are. Clear, covert, past cartel conduct presents an easy case, but a case involving prospective, future gains is much more difficult.
328
Panel Discussion
Finally, in defining and implementing competition policy objectives, there is always the challenging task of dealing with competition policy issues on their face, as well as competition policy issues that run into other policies, such as intellectual property policy. Increasingly in this age of information, hi-tech and innovation, competition comes squarely into conflict with the objectives of intellectual property policy—those trying to protect their trade marks, copyright and so on. In Canada, our experience has, to date, been quite limited, yet we see other cases on the horizon where this balancing process will become a daunting task. However, there is no turning back. • PROF. HOVENKAMP—In the last ten tofifteenyears within the United States, there has been a narrowing of the focus of concerns. I would not describe this narrowing of focus so much as the elimination of non-economic concerns, since they had already been eliminated by then. They have some influence at the margin, but very little at the centre. Rather, this narrowing of focus relates to the use of more restrained criteria for acceptable economic theory. The predominant economics has become the traditional neo-classicism, characterised by the following assumptions: markets are all more or less alike, are fairly robust, and tend toward a competitive equilibrium; firms tend to maximise their profits; consumers tend to be well informed; resources are mobile; and finally, and perhaps most importantly, if there are permanent failures in a sufficient number of those criteria in a particular market, then that market does not work and it befalls the legislative branch, rather than the antitrust courts, to find a solution. Incidentally, our conception of the robustness of markets has broadened considerably over the last ten orfifteenyears. However, antitrust policy has not simply adopted neo-classical economic theory. There are many areas where it has not. In the United States, there are relatively severe shortcomings in this regard in some areas. First, the doctrine of antitrust federalism, known in the U.S. as the State Action doctrine, or the Parker v. Brown doctrine, gives states and properly authorised local governments very expansive power to create monopolies, to create or further cartels, or to engage in other significant anticompetitive practices. Thus, if the state articulates its wish to engage in some anticompetitive act, or to permit its local government to do so, and if private conduct is supervised, even if it is hopelessly anticompetitive, then that conduct is entirely immune from federal antitrust law. A very significant set of instances where markets deviate is where the actions of state or local governments accept that immunity and permit the markets involved to deviate. The other major area is our extremely expansive use of private antitrust enforcement. From 75 to 90 per cent of antitrustfilingsin the U.S. are by pri-
5 - Competition Law Implementation at Present
329
vate plaintiffs rather than government agencies. Private plaintiffs have purely private motivations, and do not take the public interest into account. They are generally far less restrained in deciding when cases should be brought, some being quite responsible, but many not. They have the lure of treble damages and attorneys' fees for prevailing plaintiffs. That, coupled with non-expert judges (our federal judges are by and large generalists, not antitrust specialists), results in decisions containing many exceptions to the general principles of neo-classicism. For instance, the Ninth Circuit Court of Appeals recently held in Chroma Lighting v. GTE Sylvania that the Robinson-Patman Act prohibits charging two different dealers different prices, even though there was no proof of any impact on competition. This holding brought the Ninth Circuit into conflict with the Third Circuit. In the Ninth Circuit case, output in the market was not shown to be injured; the only injury was that one dealer had to pay more than another dealer. The court reasoned that when the Robinson-Patman Act was passed in the mid-1930s, competition was not on the mind of Congress. This is true, but does not distinguish this Act from any of the other antitrust laws. For instance, we know today that John Sherman's major constituency when he drafted the Sherman Act was small, Ohio gasoline producers who wanted protection from lower cost Standard Oil. Standard was shipping oil in tank cars; John Sherman's lobby was shipping it in wooden barrels. Sherman initially proposed legislation to the Interstate Commerce Commission forbidding the use of tank cars in the United States, thereby forcing Standard Oil to ship in wooden barrels. When that legislation failed to pass, he proposed the Sherman Act as an alternative. Fortunately for us, that set of motives no longer guides the interpretation of the Sherman Act. Finally, I view a recent set of a half dozen franchise cases as the nightmare of Kodak coming back to roost. Four of these cases, which are all basically the same, have now been approved for trial. Typically, the franchisee is a fast food seller. One of them, for instance, is Queen's City Pizza in New York. Queen's City alleges that Domino's Pizza has market power in the market for Domino's brand pizza dough. Pizza dough is made of flour, salt and water, and it is very hard to see how anyone could have a monopoly in pizza dough. The important element here is that the franchisee is obligated by his contract to buy all of his pizza dough from Domino's, and as a result, he is 'locked in'. Under Kodak, lock-in means that certain customers have no mobility as a result of prior choices. Did we mean, however, to use that case to rewrite contracts? Domino's Pizza is a case about someone who was unhappy with a contract that forced him to buy pizza dough from Domino's at a high price. Thus, he is now claiming that because his contract forces him to buy Domino's pizza dough, he is locked in with respect to that grouping of sales, which has become a relevant
330
Panel Discussion
market. To me, this is a silly argument. In four of these cases, the judge has denied the summary judgment motion of the defendant and the cases are proceeding to trial or perhaps settlement. • MR. HOWE—U.K. law is still different from the laws of most countries. There is nothing in the U.K. competition statutes that indicates what their objective is. The substantive test that must be met before any agreement or practice of an individual firm or merger can be stopped or modified is that the agreement, practice or merger be against the public interest. Therefore, the objective of the law is presumably that the public interest is generally promoted by competitive markets and conduct. However, this is not necessarily the case. The statutes contain some indication of the considerations that the enforcement authorities should have in mind in applying this public interest test. In my paper, I describe the articulation of these considerations as a mishmash, which is far from an economic efficiency standard. The statutes explicitly state that the authorities may declare in a particular case that producer interests dominate consumer interests, that regional considerations are pre-eminent, that costs to jobs outweigh benefits to efficiency. That is clear from how the statute directs the authorities to view public interest. It is a vague test that is hardly operational, and very difficult to apply. Second, British competition law imposes an extremely low threshold for initiating cases. Usually, it is sufficient if a simple market share test of twentytwo per cent is met. Hence, the initiating body, which is usually the Office of Fair Trading, need not establish a strong economic case before it can initiate the proceeding. This very low threshold has, however, had the benefit of permitting us to look into some areas, such as oligopolies, which would not be possible in other systems. Third, except for restrictive agreements (which are adjudicated by a court), all cases are investigated by a tribunal, the Monopolies and Mergers Commission (M.M.C.), which is only an advisory body; a Minister takes the actual decision. Accordingly, the decisive role in almost all cases that come up now in the U.K. is in the hands of a Minister. This has obvious implications for the objectives ultimately pursued. The Minister is not obliged to accept the recommendations of the M.M.C.; rather, he can seek his own solution, which may include an alternative objective, or even a political agenda. Those characteristics lead to a broad, open-ended, pragmatic approach to enforcement in the U.K. However, as in other jurisdictions, there is a tendency to approach these issues in an economics-oriented way in the U.K. This is not immediately apparent from a review of the cases. However, decisions in merger cases demonstrate that the U.K. authorities are now emphasising efficiency
5 - Competition Law Implementation at Present
331
considerations. There are a number of reasons for this. One is that the U.K. government, like many other governments, has become more pro-competition, pro-market forces in its general policymaking. In the U.K., this has manifested itself more in privatisation and deregulation than in competition law. Another reason is that the openness of the British economy and the increasing internationalisation of markets have destroyed any possibility of protecting domestic industry to make it more competitive in international markets. Not many years ago, this was a powerful argument, but it hardly exists any more. A third reason is that many more economists have been brought into the system. Indeed, before 1970, the U.K. competition authorities had no staff economists. However, the public interest test remains the law in the U.K., which provides less incentive to engage in the more rigorous style of analysis in the British system than in some other systems. Moreover, the involvement of Ministers in the process implies that it is not sufficiently transparent. It is transparent to the extent that the reports of the M.M.C. must be published, but where Ministers deviate from the M.M.C.'s recommendations, or decide that the M.M.C. should not investigate, no reasons are given. Regarding institutions, the M.M.C. is quite a good tribunal, well suited to evaluating the conflicting issues that can arise. The difficulty is that it often seeks the middle ground, and it does not like to make explicit trade-offs, such as detriments to consumer surplus, against cost savings from economies of scale. In essence, it does not like to expose these difficult choices, and tends instead to hide them in its reports. Nonetheless, I believe the institutional characteristics of the M.M.C. are better suited to perform the function of tribunal in competition cases than courts. Finally, the new British government, which came into power on 1 May 1997, has made a more robust commitment to competition law reform than its predecessor has. It has promised to propose a bill in the autumn. The likelihood is that it will be a more root and branch reform than the previous government's proposals, which were limited to amendment of the law on restrictive agreements. • MR. LIPSKY—In the U.S., the idea of non-competition-based justifications for anticompetitive behaviour was explicitly eliminated decades ago. Thus, for many years, justifications presented by defendants for their conduct have been presented in terms of economic efficiency, long run output growth, keeping prices low, etc., all of which fundamentally boil down to a long run, economics-based conception of competition. Two cases explicitly state that even though a horizontal pricefixingagreement among competitors has been established, it should be judged under the rule of reason. The two cases approved
332
Panel Discussion
such agreements because they fit into the category of restraints that could contribute to long run output growth. BMI ASCAP v. CBS held legal the collective sales agency for copyrighted musical works and Nabanco v. Visa held legal the fixing of the bank interchange fee, used by the largest credit card systems in order to function. In fact, non-competition goals in the sense of long run objectives of the competitive process play a very healthy and active role in many U.S. competition cases. Mr. Howe has referred to the tendency of authorities to avoid explicit trade-offs. Such trade-offs are never made explicit. In that respect, the use of non-competition objectives at the operational enforcement level is always a kind of show. The spotlight must be moved on the stage such that it covers only those aspects that appear relevant to a competition decision, because if it were characterised as an explicit trade-off, it would not be accepted. I believe the E.U. Commission's attack on government distortions of trade is important and creditworthy. However, I do not agree with Mr. Forrester's assertion that it is unique. In the United States, competition is exposed to the political process, as far as the antitrust authorities are concerned. For instance, Congress could decide to restrain competition under its broad, Constitutionally-based interstate and foreign commerce regulatory authority. The broad State Action doctrine in the U.S. allows a state government to restrain competition by simply explaining that this was its goal. From the standpoint of the antitrust agencies and antitrust enforcement, that is the end of the debate. It is a terribly unfortunate aspect of our law, and I agree that it is an admirable aspect of European law, that the Treaty of Rome reads like a charter of how to control government restraints of trade and how to limit state anticompetitive interference in the economy. That is something that the U.S. would do well to imitate.
Observations and comments • PROF. EHLERMANN—The M.M.C. plays an extremely important role in the U.K. system. I have advocated the transposition of this model to the European level. How well it functions depends on the individuals sitting on the M.M.C. and their courage to speak out. • MR. PITOFSKY—In the second half of this century in the United States, and in other countries as well, economic analysis has taken on increased importance, and has been applied more effectively. In the 1960s in the U.S., there was
5 - Competition Law Implementation at Present
333
great scepticism about economics and economic analysis, but this has changed. The question remains how far we should go with it. Economic analysis was described as relying upon assumptions that consumers are well informed and mobile, that they make rational choices, and that sellers are always profit maximisers. If this were always true, then we enforcers would be wasting our time, because the market would adjust to all problems. In the early 1980s in the United States, that attitude seemed to prevail, as only hard core cartels and a few horizontal mergers were challenged during that period. However, the state attorneys general, the U.S. Congress, and private enforcement provided checks and balances on the federal enforcers, as they believed that non-enforcement was inappropriate. The number of private actions increased considerably. Presently in the U.S., we are attempting to strike a middle ground by learning about economics on the one hand, but recognising that markets are not quite as efficient, cohesive, tidy, and coherent as some would argue on the other. The challenge is to accomplish this at a time when the commercial world is in a process of radical change, including global competition (e.g., ten times more exports and imports in the U.S. than there were just twenty years ago), and the increased significance of technological competition. Thus, the challenge is great, and it is not yet clear where it will lead. It is, however, clear that the 1960s represented a period of over enforcement, and the early 1980s represented a period of under enforcement. We enforcers must attempt to find the middle ground. • PROF. KIRCHNER—I shall raise two methodological points. First, the consistency of the market integration goal and efficiency is questionable, and depends on the type of economic approach being employed. If we apply neoclassical industrial organisation theory, then this problem cannot be addressed. However, if we apply a neo-institutional approach combined with a classical economics approach, then we have difficulties with this simple efficiency test. Second, as to legal certainty, I believe it is not simply an ideological question. Rather, it involves the rule of law and checks and balances on the antitrust agencies by the courts. It amounts to a principal-agent problem. Can the courts really control the antitrust agencies? This is not a matter of economic analysis. We might have good economic criteria and bad economic criteria, which are not justiciable. Therefore, this amounts to a lack of control of antitrust agencies by the courts. • MR. WARNER—Regarding checks and balances, the O.E.C.D. Competition Law and Policy Committee plays a useful role by performing country reviews of administration of antitrust policy. These reviews constitute an important device that could be expanded also to cover market access issues.
334
Panel Discussion
In the last meeting of trade ministers at O.E.C.D. in late May 1997, a regulatory reform project was discussed, a key part of which also involves competition principles in regulatory reform. They agreed to implement a peer review system for domestic regulatory regimes. One aspect of that will be to review the application of competition policy to domestic regulatory regimes. This begs the question that if the W.T.O. has an existing country review of trade policy, this could be an avenue for building competition principles into the W.T.O. framework. • MR. CASTANEDA—Regarding checks and balances, I shall make two observations concerning the Mexican case. First, after 1993, both the government and the private sector found the government to be the main source of anticompetitive conduct. Thereafter, the concept of ancillary functions of the state was created. Thus, Pemex, the large state-owned oil company, would have an antitrust exemption, but would not be allowed to go beyond the limits set. The competition commission then took action against Pemex for anticompetitive conduct on two occasions. This amounts to one government entity, the competition commission, taking action against another governmental entity, Pemex. Second, specific provisions have been included in some statutes, such as the telecom statute, referring all competition matters to the competition law and to the competition commission. In the U.S., the Transportation Commission, which is a special regulatory commission, approved a merger in the railroad industry between Union Pacific and Southern Pacific. This has had effects on trade matters, as it affected 95 per cent of all the transport trade going to the U.S. through the Mexican gate. Competition rules should take this into consideration. It is undesirable to leave the interpretation of competition rules to special regulatory bodies, because as many competition rules will emerge as there are boards or commissions. • MR. SCHAUB—I admire the inimitable way in which Prof. Hawk and Mr. Forrester have illustrated the weaknesses of Commission practice regarding the multitude of objectives expressed in Commission decisions. I would like to recruit the two for training courses for both the young, and in particular notso-young, officials in DG IV. Second, I find Prof. Hawk's explanation of why the Commission is reluctant to decentralise Art. 85(3) to be simplistic. Over the last three years, I have not heard that the multitude of objectives, or in particular that the market integration goal, is the insurmountable difficulty. There are presently some serious difficulties, because nearly half of the national competition authorities still do not have national legislation empowering them to apply Community law. This
5 - Competition Law Implementation at Present
335
poses serious problems. For instance, in the BAIAA case, the British competition authority is, for the first time, applying Community law in parallel to the Commission proceedings and in close cooperation with the Commission. The other problem is that many of the national authorities that have national legislation empowering them to apply Community rules are not interested in doing so, or do not have the resources to do so. For instance, in Belgium, such national legislation has been enacted, but the national authority has disappeared from the scene. Thus, it applies neither national nor Community law. Most importantly, we must assure a coherent application of the law within the fifteen Member States. This is not necessarily an insurmountable problem, but it should not simply be dismissed. Third, regarding legal certainty, I agree that excess use is made of this argument. However, in every competition law, an inevitable tension exists between legal certainty and use of economic analysis. We must continue to seek the proper balance between these two elements. • MR. HEIMLER—Economic analysis is important with respect to the efficiency goal in antitrust enforcement. In addition, economic analysis is important with respect to the Community's market integration goal. Economics is about paradoxes, and one of the most important paradoxes on which we all agree is that a system in which everyone does what he wants does not lead to chaos but to the best results. This is one of the most important contributions of economic thinking. There are many other paradoxes in economics, such as the paradox that vertical restraints promote competition rather than restraining competition. The same type of paradox exists with respect to vertical restraints and market integration—rather than reducing integration, they promote it. Thus, economic analysis is needed to analyse market integration. Preventing firms from introducing restraints is not always the best way of promoting market integration. Rather, in many instances, allowingfirmsto introduce vertical restraints or absolute territorial restraints is a good way to promote integration. • MR. KOBAYASHI—I would like to discuss an example of how the objectives provision affected the amendment of the Japanese law. In Japan, the holding company, defined as a company whose principle purpose is to control other companies, is totally prohibited. The reason for this is historic: to prevent the re-emergence of 'zaibatsu', or big family concerns. The private sector has strongly opposed this prohibition, asserting that it has been deprived of this type of business organisation, and that this prohibition is unique to Japan, and perhaps Korea. Government agencies in charge of industrial policy also favour
336
Panel Discussion
abolition of this provision. The J.F.T.C., however, believes that regulations controlling the establishment of holding companies are still needed because certain economic practices in Japan, such as the custom of mutual holding of shares among companies, still result in the formation of large business groups. Thus, the J.F.T.C. has been rather isolated supporting the provision. Section 1 of the Japanese anti-monopoly Act provides that by preventing excessive concentration of economic power, it aims to promote free and fair competition. The J.F.T.C. has relied on this provision, arguing that the need to prevent excessive concentration of economic power still exists. Thus, it drafted a proposed amendment to the prohibition, which would abolish the total prohibition, but establish certain conditions under which the setting up of holding companies would be prohibited. For example, the total value of shares or assets of the entire group held by a single holding company could not exceed ¥15 trillion. Thus, the lengthy provision on objectives exerted an influence in this situation. • MR. KHEMANI—When the Canadian competition law was being drafted, the question arose as to who gives the competition office the authority to engage in competition advocacy. From the experience of other countries, we saw that since competition does not have a fervent lobby, competition advocates may be closed out of appropriate fora. Accordingly, a provision was included in the Competition Act that gives the competition director the power to intervene before federal tribunals, commissions, and parliamentary committees to advocate competition. The competition director cannot address other matters, such as equity or regional development, which do not fall within his or her mandate, but can specify the damage to competition that can result from various federal as well as other regulatory policies. This is a very important aspect of competition policy, which should be emulated in other countries. Both informally and formally, various other countries have analogous provisions, which may not be statutory. For instance, in the U.S., such a right has been recognised historically as the right of an intervenor in regulatory hearings. Formalising such a right in a competition law makes the competition office a focal point for discussion of competition issues, and of the potential negative spillovers from other government policies and arrangements. The competition office is the logical focal point for competition advocacy. For monetary policy, there is the central bank; for fiscal policy, there is the ministry of finance; for commercial policy, there is the ministry for international trade. However, there is no focal point for competition unless the competition advocacy function is embodied in a competition office. In this regard, the interface between competition policy and trade policy is
5 - Competition Law Implementation at Present
337
very important. One of the major interventions by the Canadian competition office was in an anti-dumping case brought by the three large automobile companies, Chrysler, General Motors and Ford, against Hyundai Motors. We in the competition office argued that the Canadian import tribunal should take into consideration the impact of their decision on competition and consumers. In other jurisdictions, this function of the competition office would be very important in the context of privatisation proceedings. Governments are often in conflict with respect to privatisation. For instance, in the U.K., when British Caledonian was being privatised, both British Airways and Scandinavian Airlines were interested in acquiring it. The Thatcher government accepted the higher bid from British Airways, but could have injected competition into the holiday charter market by allowing Scandinavian Airlines to make the acquisition. In another context, in the Czech Republic, when Volkswagen made investments and an acquisition of Skoda, it also attempted to get increased tariff protection and a moratorium on foreign investment in the Czech republic. But for the advocacy of the competition office, that type of extraction of privileges might have gone through. Many jurisdictions are enacting various regulations in the context of privatising their infrastructure services, and are including competition policies within the regulations, thereby leading to fragmentation of competition. They fail to recognise that a competition office could play a critical advocacy role. • PROF. FORNALCZYK—A very important problem that exists not only in transforming countries, but all over the world, relates to the interdependence between state intervention into the economy and competition law implementation. I agree that a competition agency cannot be absolutely independent, because, with the exception of the Hungarian case, all competition agencies are government bodies. They are therefore all within the government. However, a competition agency must have professional independence. This is closely associated with the need for transparency and responsibility for decision-making. What should be the strategy of a competition agency in a case of state intervention that may strengthen the market position of certain companies? When I was the chair of the Anti-monopoly Office in Poland, in cases in which the government established special measures which strengthened the market position of a company, we did not initiate proceedings for abuse of dominant position or anticompetitive agreements. This would have meant implementation of competition law against the behaviour of companies supported by the government. It would have been undesirable, because the government must act with internal cohesion.
338
Panel Discussion
• MR. FAULL—There is a tension between the idea that competition authorities should be independent, just applying the law and leaving the allegedly dirty business of policy and politics to others, on the one hand, and the view that the competition authority should advocate competition solutions when other policy and political positions are being taken, on the other. In the E.U., having DG IV embedded in the Commission, with its Commissioner sitting at the Commission table and discussing matters with the Parliament, etc., makes competition advocacy much easier than it would be if an independent agency were set up, with the sole responsibility of applying the law. Thus, there is some tension between the most effective forms of advocacy and independence. • MR. PERA—I shall make three points. First, institutional design must be considered in conjunction with how the law functions. In many instances, power is given to governments to intervene in order to allow mergers and agreements. How the law allowing such agreements is structured is crucial. For instance, E.U. law gives considerable powers to the Commission in the evaluation of particular cases. In the case of Italy, the government was never interested in competition. The Italian authority was born as a completely independent authority, and actually took charge of competition matters. This has had two effects. One, it has caused the authority to focus fully on competition, and to apply the law to achieve the competition objective, because that is its sole task. Two, it has given the authority a strong power of advocacy. The authority often intervenes, advocating competition. Second, if an economic approach to market integration or vertical restraints is not followed, this leads towards formal evaluation and therefore centralisation. Thus, either an economic approach must be followed or any change in the application of Art. 85(3) is bound not to succeed. Third, enforcement of Art. 90(2) is usually viewed as the application of regulatory measures by the Commission. I believe, however, that it should be viewed as part of the antitrust policy of the Commission. The Commission has a peculiar character, being the only antitrust authority that can act on state measures, usually via proposing regulations or directives to the Council. However, it can act directly under Art. 90(3). Those measures derive from the perception that they basically involve violations of Art. 86. Thus, they seem to be structural measures, which should perhaps be considered within the usual domain of competition policy stricto sensu of the Commission. • MR. HOWE—I would like to take the floor to react to Mr. Khemani's comment on the British Caledonian case, because I do not agree with his comments.
5 - Competition Law Implementation at Present
339
It was not a privatisation, but a straightforward merger case. The bid for British Caledonian by Scandinavian Airlines would have created no problems, and therefore could have gone ahead. However, the British Airways bid caused competition problems and went to the M.M.C. In applying the public interest test, the M.M.C. had the dilemma of deciding how to weigh the adverse effects on competition, which were rather clear, with arguments about employment, because it was a failing firm. Although I respect the point that many issues that are not strictly competition considerations can be evaluated using economic analysis, I am not sure that when the trade-off is between jobs and competition, economic analysis is very helpful. The M.M.C. did not attempt to apply economic analysis, but instead had to make a judgement whether saving jobs meant more than losing competition. It made its judgement, weighing intractable things. • PROF. AMATO—Regarding the question whether a tension exists between independence and advocacy, it depends on what is meant by advocacy. According to the statutes implemented by the Italian authority, we have both adjudication and formal advocacy powers, in the sense that we can send recommendations to the government and to Parliament whenever we believe an existing statute, bill, or something else might distort competition. We do not view the two as different jobs. I would consider a tension to exist if we were involved in drafting and negotiating the final version of legislation, being present for meetings with ministers and others. However, we are totally foreign to that. Our recommendation is published and considered in the debate. However, we are not present when the final decision is taken, which I believe makes a significant difference. Second, regarding the reliance by courts on economic analysis, I do not view protection of competitors in essential facilities cases as the single most important exception to the judicial emphasis on efficiency. Rather, in Europe, another important exception is abuses of dominant position. Competition is protected by protecting existing weak competitors vis-a-vis a dominant firm that might otherwise destroy them. We do this for many reasons. For instance, we may believe that weak competitors that might survive are worth protecting because competition may be restored through them in the foreseeable future. I believe we continue to protect weak competitors in Europe due to our tradition of regulation of economic activities. We find it natural in Europe to impose rules when a firm reaches a certain level of power. However, we might also view protecting the right of minor competitors to sell as covered by the competition rules, according to the initial American tradition. I am not certain which applies.
340
Panel Discussion
• MR. CASTANEDA—Mexico provides a fresh example because it has been implementing its competition law only since 1993. Two levels of advocacy should be distinguished. One involves sending letters to Congress or to regulatory agencies; the other is perhaps less an advocacy function than a statutory function, where powers have been given by statute to the competition commission. I believe the latter is gaining momentum to be part of a competition policy, which I will attempt to illustrate with two examples. The first involves rate setting, i.e., setting the maximum rates on telecom services, natural gas and other industries. Statutes provide that no maximum is to be set by a regulatory agency, except when there is no substantial competition in that particular area, which is a decision that must be made by the competition authority. Accordingly, the powers are removed from the sectoral regulatory agency, which eliminates the possibility of regulatory capture. The second example involves regulations of specific sectors with respect to mergers within that industry, which do not meet the monetary thresholds of the competition statutes. Such new sectoral regulations generally provide that mergers, transfer of licences, or concessions are to be reviewed by the competition commission. As in the first example, this is a way of decentralising power, and provides the possibility of eliminating regulatory capture. It focuses on the structural problems rather than the behavioural fudging that sometimes is natural when several regulatory agencies are involved in one transaction. The third example concerns the trade area. The new international trade law in Mexico provides that the competition commission may provide input in all anti-dumping cases. Thus, if information is presented in an anti-dumping case that establishes a breach of the competition law, the competition commission is obliged to try the case on the merits of a possible breach of competition law. This creates the possibility to eliminate rent-seeking—opportunistic conduct by people who are attempting to induce the government to file an anti-dumping case—and the competition commission to file an antitrust case. This presents the danger that there will be regulatory capture of both the trade and competition regulators, which supports the argument for more transparency in the system, especially in areas like dumping. • PROF. NEVEN—It is possible for an agency to have too much independence. An assumption seems to be made in this discussion that more independence is always better. However, we should be careful about how we define independence. We could say an institution is independent if it has the ability to fulfil the objectives that have been assigned to it without interference. Denned in this way, independence is potentially dangerous for two reasons: first, the agency
5 - Competition Law Implementation at Present
341
can become lazy and do nothing; second, the agency can become full of ayatollahs, who enforce the law in an excessive manner. Accordingly, the agency can be captured by itself, in a sense, if it becomes too independent. This risk is particularly strong if the agency's objectives are very broad. If the objectives are narrow, then the risk of capture by itself is much less. Some sort of accountability of the independent agency by a third institution is an essential element of the German system, and that is probably what stops the competition agency from becoming either lazy or an ayatollah. Second, there is more than one way of achieving independence of an agency. We all have in mind the German model, which I admire. However, it is not the only way to achieve independence. The agencies in Eastern European countries are extremely telling. For instance, the head of the Czech competition agency was a full minister, who sat in the cabinet. This is no longer the case. The Czech agency was, nonetheless, rather independent because the man who headed it was respected. Therefore, perhaps one way of making the agency independent is to have a strong politician in charge of competition who is going to be respected by his peers in government. • MR. FORRESTER—There is a very profound difference between the U.S. approach and the E.U. approach with respect to how competition law is enforced. In Europe, it is quite common for parties to sign an agreement, and then years subsequently to assert 'Sorry, I was wrong to sign that agreement, I was living in sin all these years, and now I wish to be purged of those sins, I will no longer obey that agreement'. Because of the structure of Art. 85, the European Commission regularly intervenes, and courts intervene, to examine whether the parties were correct in asserting that the agreement was prohibited under Art. 85(1) when it was signed. That is a very distinctive and particular difference between the E.U. and the U.S. approach. Second, leading into the ayatollah problem, the European Commission also has the goal of protecting parties from inappropriate contracts. For instance, the know-how regulation, which specifies how to license knowhow on an exclusive basis, provides that the approval of Community law would apply only if the knowhow was recorded in writing. A competition agency has no reason to make such a requirement. It reflects a belief on the part of the Commission, which I submit should not be there, that it is part of its responsibility to preserve parties from bad bargains. Third, the Commission is concerned about quality; naturally, it is concerned about consistent enforcement throughout the Common Market. I submit that it is impossible to achieve consistency, but that is not a bad thing. Perfection is not the goal that the Commission should be pursuing. Rather, by
342
Panel Discussion
taking a large number of decisions and running some risks, more jurisprudence will be produced, with a wider respect and recognition of Community law. • MR. WARNER—The trend of narrowing competition law objectives primarily towards a more economics-based analysis may present problems for jurisdictions that are adopting competition laws for the first time. As our analysis of competition law is increasingly focused on economics, it is questionable whether the data that are available in some of these countries are sufficient to perform real empirical economic analysis, as opposed to simply telling economic narratives. For instance, in my experience with mergerfilingsin Canada, data such as S.I.C. codes were not available. We relied more on fuzzy narrative telling, as opposed to rigorous economic analysis of market definition, calculation of market share, entry, efficiency, etc. I am both an American and a Canadian lawyer, and am often struck by the differences between the two systems. Thus, implementation of competition law objectives in countries just developing competition laws will be difficult, to the extent that data are required for economic analysis. • PROF. IMMENGA—As to the relation between underlying traditions and values of a country and the objectives and implementation of competition law, I would like to discuss a German example. This system provides for the ministerial authorisation of any prohibition delivered by the federal cartel office. It functions well because competition and liberalism in our society play a very important role and are broadly accepted, not only by the public, but also by governmental institutions, including the cartel office and the Ministry of Economics. The system could not function without broad acceptance of these underlying principles. On the European level, as regards the issue of the creation of an independent agency, I believe that a separation of the kind that exists in Germany would be dangerous, such as if DG IV were independent and its decisions could be overridden by the Commission. Thus, in considering the implementation of law, the context of the general underlying culture and acceptance of values must always be taken into account. • MR. SIRAGUSA—Antitrust authorities can have a dynamic function of bringing about change. The Italian antitrust authority is working in an environment that has a tradition of non-acceptance of principles of competition. Thus, Italy has a very peculiar situation: an authority which is like a Martian who has landed in a country with a tradition of state monopolies and government intervention, and is working to broaden the spectrum of applicability of
5 - Competition Law Implementation at Present
343
the laws. A relationship must exist between the nature of this antitrust agency as an independent authority and the objectives of the law. In a situation like this one, I believe the antitrust authority should have a strict competition law objective. This will allow it to work independently, and to perform the function of advisor to the government on competition law matters. However, once the antitrust authority is responsible for implementing a law with other objectives, such as market integration, and using antitrust laws to liberalise markets, then it is more difficult to give the authority complete independence. It will become almost impossible for the authority to reconcile the tension between the various values. This is why the Commission is an ideal forum, given that it must implement a law with both competition objectives and deregulation objectives. I believe that in Europe values are accepted broadly enough that they could be entrusted to an independent authority. However, such an authority could not also be responsible for pursuing the other objectives. • MR. PITOFSKY—From my experience as a member of an independent agency in the U.S., our independence is much exaggerated. There are many ways by which the rest of the government can influence an independent agency and can replace the Commissioners as their terms expire: the President has the power to appoint the five members of the Commission; the Commission must undergo oversight review by Congress every two years, and Congress is not shy about telling us when they disapprove of what we are doing; through the annual budget approval process, Congress can inform us each year whether they disapprove of what we are doing; and our most important decisions are reviewable in court, and therefore must be ratified through the judicial process. These controls are not merely academic. Fifteen years ago, when the F.T.C. was widely regarded as having exceeded its boundaries, all four of those controls were imposed in such a way that the wings of the agency were clipped. Second, when Jim Rill was head of the D.OJ.'s Antitrust Division, he probably had greater influence in advocacy than someone like me, heading the F.T.C, especially if negotiations were part of the advocacy. However, if the competition advocacy is directed not to another government agency, but to the legislature, then history shows that in the U.S., the perceived objectivity of an independent agency has often been an advantage. An independent agency could have greater influence in the legislature than a member of the existing government, especially in the U.S. today where the government is divided, and the legislature is wary about proposals made by the administration. • MR. FORRESTER—It is easy in browsing through decisions of the Commission and the Court to find a hundred worthy social and economic
344
Panel Discussion
goals that can be pursued under competition law. However, that does not prove either that those were the goals being pursued, or that the decisions taken were bad decisions. Rather, these are the justification for particular enforcement decisions. Accordingly, the richness and the discrepancies are not problematic. In the U.S., difficult decisions taken have probably been justified by an assortment of peculiar, even at times inappropriate, reasons. The problem in Europe is not the setting of incorrect goals, but the lack of resources to pursue the right goals, and too much concern about the possibility of making mistakes. Less perfectionism would allow more law and richer law to develop. There would be less need for agonising about past mistakes. • MR. GOLDMAN—A trend towards greater use and application of economic principles, and toward efficiency as a leading objective, is a positive development. However, in contemplating the implementation of objectives, it is important to consider the weight attributed to those objectives, particularly with respect to cases of cross-border conduct, and cross-border mergers. For instance, we should consider whether, when raising the efficiencies defence, the efficiencies must occur within the jurisdiction where the anticompetitive effects are felt, and whether the efficiencies can completely override the anticompetitive effects. These questions all relate to implementation of similar objectives. If the weight given to each objective differs considerably, such as occurred in the DeHavilland case, then the results may conflict, notwithstanding similarity of objectives. Therefore, even though the number of cross-border mergers will increase, conflicts are likely to decrease if both the objectives and the manner of implementation continue to converge. Fora like this and the O.E.C.D. play an important role to this end. • PROF. HOVENKAMP—The main antitrust objection of those who disagree with the essential facilities doctrine in the U.S., or who believe it should be seriously constrained, is that the term 'abuse' in such cases refers merely to the refusal to share an input. More specifically, in an essential facility case, a defendant refuses to sell something to a rival, which the rival claims is needed to conduct its business. I object to such a use of the antitrust laws because sharing would ordinarily occur at the monopoly price in any event. Thus, nothing more than increasing the number of participants in this market by one would be accomplished by a court order requiring a monopolist to sell to a rival, unless the court wants to act as a price regulator, since market output and market price will remain the same. Of greater concern is that this smaller firm will have acquired a judicially enforceable right to share the monopolist's input, which significantly decreases its incentive to find its own independent
5 - Competition Law Implementation at Present
345
source for that input. However, the role of the antitrust laws is to make markets competitive, not to force firms to share inputs. Markets are competitive not when firms share their inputs, but when they find their own independent sources for them. Second, to clarify my earlier point about the Domino's Pizza case, the pizza franchisees were required by their contract to purchase their dough from the franchisor. They found that they paid more than the market price for pizza dough, because that is how the franchisor receives its franchise fees. Such an arrangement may well present a problem of unjust contracting, but the appropriate solution should then be found within the realm of contract law. It is inappropriate, however, to conclude that because the buyer is required to purchase Domino's pizza dough, then Domino's pizza dough is a monopoly, and therefore raises an antitrust problem. Doing so will have troublesome implications for a wide range of cases where plaintiffs will urge very narrow, idiosyncratic market definitions. Finally, regarding the British Caledonian case, there are economic criteria for analysing failing firms. However, such criteria are not employed systematically. Historically, the U.S. Supreme Court has not relied on them. Congress has been at least as concerned about problems of employment or injury to shareholders as it has with efficiencies. Efficiency rules could, however, be adopted. In some cases, it may be socially preferable to allow certain failing assets to be taken out of the market rather than to continue to use them. Moreover, the second best buyer should always be considered. In the British Caledonian case, there was a second best buyer that may have been able to save the firm with much less severe implications for competition. • MR. HOWE—I have participated in many round table discussions with colleagues from other enforcement agencies, which left me with the impression that we all seem to be approaching problems such as market definition, dominant firm behaviour, mergers, and assessing efficiencies in much the same way, notwithstanding the differences in our laws. The subject of objectives is bound to exaggerate the differences, whereas discussion of how to approach specific behaviour such as collusive tendering or exclusive dealing would tend to emphasise the similarities in our approaches. The British system is different from any other system, especially as a result of its public interest test. Public policy should be driven by public interest considerations. All laws are designed to promote the public interest, whatever their specific objectives. The question is whether the concept of the public interest provides a satisfactory operational objective for the British enforcement agency. I believe it has two main problems. First, a public interest test implies
346
Panel Discussion
that no conduct is prohibited until it has subsequently been held to violate the public interest. However, the agency cannot be given major powers of investigation, and major penalties cannot be imposed on those who have been found to have violated the public interest. Accordingly, the British authority does not have the necessary tools to perform these functions effectively. Second, public interest is a very ambiguous concept. It allows, for instance, that a failing firm merger case will be addressed with all the rigour of the U.S. D.O.J. approach one day, and then in a subjective manner the next day. The track record of the British authorities over the years shows that although even cartels have been allowed if they were found not to be against the public interest, cartels are basically prohibited in the U.K. The law does not condone them, but condemns them. However, the British authority does not have the tools it needs to detect them. Regarding abuse of dominant position, under the British system a much wider range of dominant behaviour may be considered than some purists think is appropriate. This includes, for instance, excessive prices and distribution issues, as well as resource allocation issues. As for mergers, over the years we have moved away from heavily weighing non-economic, nonefficiency considerations. At present, although occasional aberrations occur, it is generally difficult to detect differences between our approach and the Commission's approach, although the statutes are quite different. Finally, the British system has more checks and balances built into it than most, which is an important advantage. However, when part of the checks and balances system involves a Minister, trouble may develop. We have a Minister of Competition and Consumer Affairs, who is not himself in the Cabinet, although his superior is. It is excellent if that Minister is pushing hard for competition, because it implies both that the final decision-maker is committed to competition, and that there is an advocate for competition within the government. However, problems occur when the politician has a different agenda. When there was a Labour government in power in the 1970s, the competition authorities were virtually sidelined, as the government chose to employ price controls and encourage mergers to create large scale British firms. There have been no indications that this will occur again with the new Labour government. On the contrary, we are looking forward to a much more rigorous enforcement. • MR. LIPSKY—We have not explicitly discussed the pursuit of various objectives and the problem of market failure. Even if pure economic efficiency objectives are adopted, in some industries competition may be an inappropriate policy. The essential facilities doctrine, which is a form of regulation of abuse or monopolisation, creates an institutional problem for antitrust enforcement. Once a court requires access, it must be prepared to regulate price. Otherwise,
5 - Competition Law Implementation at Present
the monopolist can simply set a very high price and achieve the same end as that achieved by refusal to deal. Moreover, it must also be prepared to regulate all the ancillary terms of service and quality. It must have the power to force the monopolist to add capacity, or else the monopolist will maintain enough capacity for only its own use. Thus, it is a slippery slope to a form of complete regulation, which may be perfectly justified on economic grounds. However, it is questionable whether federal district court judges in the U.S. system, or the Commission in the European system, should engage in this regulatory function. Second, it is tempting for a competition authority, as for any other institution of government, to improve its legitimacy by invoking themes that become popular in the political debate. However, there is a great danger that the political wind may shift, and the theme pursued by the agency will become unpopular, or that the agency's direction will become incoherent. • PROF. HAWK—This summary of Session III will hopefully lead us nicely into many of the issues to be discussed in Session IV. There seems to be a consensus that there is a gap between the rhetoric and the reality. Mr. Schaub's distinction between fundamental objectives and enforcement criteria is essential to a reasonable debate, because it eliminates much of the confusion. Second, there appears to be a consensus that a strong trend is under way toward economics, at least in industrialised countries. This creates some difficulties, such as insufficient data or imperfections of economics, but the strong trend is toward narrowing enforcement criteria to economics. One exception mentioned by Prof. Amato is the use of abuse of dominant position laws to protect weak competitors. In a Supreme Court case involving the Aspen ski resort, joint ski tickets used to be issued. The main ski lift operator of three of the four hills said: 'You can't get a ski ticket with me anymore', which constitutes a refusal to deal. The plaintiff sued that lift operator. The Supreme Court affirmed a jury instruction, which said: 'You on the jury will find a violation if you find that the intent in discontinuing the ticket was to exclude the competitor'. That decision was subject to extreme criticism by many commentators for reflecting a populist notion that most economists would agree had nothing to do with microeconomic analysis. Thus, such judgments exist in the United States, although they are rare today. Third, why do we have this trend toward economic analysis? I was pleased that no one responded to my assertion, defending Easterbrook, that the people who want to limit antitrust to economics are not simply making a political judgement, which is one way to dismiss them. They are making a lawyer's judgement that economics provides better tools within the legal system to decide a case when an economic law is being applied.
348
Panel Discussion
Fourth, the distinction between competition advocacy and adjudication is crucial in discussing independence. With regard to the assertion that a competition culture is needed in order to have an independent authority, where is the competition culture in Italy to explain the truly extraordinary history since 1990 of the Italian authority, which is independent? As an outside observer, what has occurred with this independent authority is phenomenal. Thus, I am extremely sceptical that a competition culture is needed in order to have an antitrust authority.
Barry E. Hawk Partner, Skadden, Arps, Slate, Meagher & Flom, LLP Director, Fordham Corporate Law Institute New York, New York, U.S.A.
Introduction The principal focus of Session Three is the actual in fact implementation of competition law objectives in the enforcement and application of competition law in specific jurisdictions. Following Hume's distinction between the 'is' and the 'ought', this session is intended to focus on the is, in contrast to Session One, which focused on the ought. Session Three participants were asked to focus their papers and remarks on the following questions: (1) To what extent have the following policy considerations influenced the application of competition law in your jurisdiction (contrast time periods if appropriate) and describe the actual influence (not theoretical debates): (a) Consumer welfare/allocative efficiency in the static sense; (b) Production efficiency in the static sense; (c) Efficiency in dynamic service (e.g. innovation, etc.); (d) Income distribution or wealth transfer effects; (e) Deconcentration/dispersal of economic power and preservation of democratic government; (f) Protection of small or medium sized enterprises; (g) Protection of individual traders/entrepreneurs; (h) Employment effects; (i) Market integration; (j) Environmental, health and safety, and other considerations; (k) 'Industrial policy' (provide its meaning in your jurisdiction); (1) 'Competitiveness' (provide its meaning in your jurisdiction); (m) Promotion of national champions; (n) Promotion of exports and international trading conditions; (o) Fairness/equity; (p) Promotion of opportunity; (q) Protection of consumers from exploitation; (r) 'Public interest' (provide meaning);
352
(2)
(3) (4) (5)
(6) (7)
(8)
(9) (10) (11) (12) (13)
Working Paper I
(s) Macroeconomic effects (e.g. combat inflation); (t) Others? Assess the operability (i.e. useful application in particular cases) of the policy considerations in your jurisdiction; compare their operability with their rhetorical use (e.g. general enforcement policy statements, speeches, annual reports, etc.; assess the legal certainty (and uncertainty) generated by these policy considerations. Is there any priority or hierarchy among these policy considerations in your jurisdiction? How are decisions or conflicts avoided or resolved in general and in particular cases? What is the role/effect of institutions on competition policy considerations and objectives? Are some institutions (e.g. administrative bodies) better able to further certain policy considerations (e.g. promotion of national champions or protection of employment) than other institutions (e.g. courts)? Is there a relationship between 'independence' of the antitrust authority and the implementation and prioritisation of policy objectives? Describe the relationship (if any) between 'federal' or centralised enforcement on the one hand and implementation of specific policy objectives on the other hand. For example, do certain policy objectives or considerations require more centralised or uniform enforcement? What specific legal situations demonstrate or reflect the tensions/conflicts between efficiency (and economics-based considerations) and other considerations (e.g. fairness and protection of small traders)? (exclude E.U. market integration goal, which is separately addressed below); possible examples are: price discrimination, excessive prices, 'exploitative' abuses, essential facility cases, definition of a restraint of trade (e.g. Sherman Art. § 1, German A.R.C. § 1 and Art. 85(1)). Are formal/express exemption systems better (or worse) suited to resolve/avoid tensions and conflicts among policy considerations? Consider the importance of the market integration goal in the E.U. in the present timeframe. Has there ever been a similar goal in your jurisdiction? If so, describe and compare it with the E.U. goal. To what extent does the market integration goal conflict with other competition policy objectives, such as efficiency? What substantive areas have been most affected by the market integration goal, e.g. vertical restraints, geographic price discrimination, intellectual property rights?
6 - Competition Law Implementation at Present
353
(14) To what extent can/should the Treaty's free movement rules today make unnecessary or less necessary competition policy oriented toward the market integration goal? (15) Should the E.U. market integration goal be modified or eliminated? And if so, how should this be implemented? More tolerant substantive rules (e.g. toward vertical restraints)? More relaxed jurisdictional rules (e.g. replacement of a Community dimension test for the potential effect on interstate trade test)?
Discussion Points/Conclusions The discussion was largely structured around a set of propositions, with respect to which there was a surprisingly strong consensus. First, there was apparent unanimity that a considerable gap exists between the rhetoric of competition law objectives and the reality of their actual implementation. This gap is particularly marked where one contrasts general 'political' recitations or litanies of competition law objectives with the (much narrower) list of objectives relied upon to decide particular cases. The Goldman/Barutciski paper discusses some examples of this for Canada. Second, there was apparent unanimity that much of the debate about objectives arises from a failure to distinguish between the objectives of competition law and the asserted benefits of competition. For example, asserted macroeconomic benefits of competition, such as reduced inflation, have at best only a rhetorical role with respect to competition law. Much of the political and economic debate can be explained by this 'category failure', to borrow another philosophical term of Gilbert Ryle. A similar distinction usefully can be made between fundamental/ultimate objectives and operational criteria for enforcement. Alexander Schaub's distinction between fundamental objectives and enforcement criteria nicely captures this distinction. Third, there was acceptance of the proposition that the general competition law statutory provisions (i.e. those concerning restrictive agreements, dominant firm behaviour and mergers) generally do not express objectives. Thus, competition law objectives usually must be inferred, a situation which permits changes in objectives over time, despite the absence of legislative action. One exception is Canada, where the current general statute expressly identifies four objectives: efficiency and adaptability of the Canadian economy; protection of small and medium size firms; international competitiveness; and competitive prices and consumer product choice, as discussed in the Goldman/Barutciski paper. Another exception is Japan, where promotion of employment and
354
Working Paper I
democracy are stated as objectives, among others, as described in the Kobayashi paper. In both jurisdictions, however, it is not clear whether these objectives have played a direct role in the application/enforcement of Canadian and Japanese competition law. Fourth, there was implicit acceptance of the proposition that in many jurisdictions (e.g. Canada and the United States) there is a strong contrast between statutory objectives expressly articulated in the legislation and the significantly narrower interpretations on the part of the competent authorities. For example, as discussed in the Goldman/Barutciski and Hovenkamp papers, many of the Canadian and U.S. provisions (like the bans on price discrimination) expressly invoke 'populist' (Poujadiste) non-efficiency objectives (if not noneconomics-based objectives). On the other hand, as discussed in the Hovenkamp paper, courts have been more willing to follow the expressed statutory objectives, probably because of their more limited discretion when faced with a claim under those statutes. Fifth, there was a strong, if not unanimous, consensus that in most jurisdictions there is a marked trend toward a narrowing of competition law objectives, primarily toward a more economics-based analysis that accords great (where not exclusive) weight to objectives like consumer welfare or efficiency (allocative, static or dynamic). Examples are discussed in the Goldman/ Barutciski (Canada) and Hovenkamp (United States) papers. This is certainly true in the more industrially developed countries. In Canada, the efficiency objective is supplemented with broader objectives, some of which are consistent with efficiency (e.g. the promotion of an open marketplace, free from the abusive exercise of market power by other market participants), and others, such as fairness, that do not fit well with an efficiency objective, as discussed in the Goldman/Barutciski and Matte papers. This trend rests on the increased emphasis on economic analysis in competition law enforcement. Thus, the growing influence of economists and economics in competition law not only has shifted the terms of the debate, but also has de-emphasised historically accepted, non-efficiency, populist or distributive objectives like protection of individual small traders. In turn, this has had the important practical effect of discretionary non-enforcement of certain competition statutes and legislation, which expressly were intended to protect certain groups (like small traders). Although courts have been more willing to follow express statutory objectives, the trend toward efficiency as the predominant objective can also be seen in U.S. court decisions, as discussed in the Hovenkamp paper. Perhaps the most important exception to the judicial emphasis on efficiency and protection of competition over protection of competitors concerns dominant firm behav-
6 - Competition Law Implementation at Present
355
iour, where populist or distributive concerns arguably underlie aggressive application of the essential facility doctrine. Another example is the continued per se ban on resale price maintenance. This narrowing of objectives does not result simply from a 'political' disagreement with certain statutory objectives, such as protection of small traders. The narrowing and consequent emphasis on efficiency and consumer welfare also rests on a 'jurisprudential' or lawyers' judgment that economics (and thus economics-oriented objectives) provides better analytical tools to reach more predictable, well-reasoned results in particular cases (and to formulate more precise general policies) than broader, non-efficiency objectives. In a sense, application of the law to a specific set of facts requires more focused objectives, while the political arena (and speechmaking) can live with broader objectives. There is a greater need for transparency to the extent that broader (nonefficiency based) objectives are used. This is particularly true where the possibly relevant objectives in a specific case appear inconsistent. Published detailed rationales are necessary to explain how the different objectives were applied and reconciled, and to limit the discretionary power of the decisionmaker. Sixth, it was recognised that interesting issues arise concerning the relationship between the institutions applying competition law and the objectives. For example, some institutions, such as administrative bodies, may be more capable of furthering certain objectives (e.g. promotion of national champions or protection of employment) than other institutions (e.g. courts). Moreover, there is a relationship between the 'independence' of the competition authority and the implementation and prioritisation of objectives. Seventh, the E.U. is the only system with a market integration objective. This is certainly not mere rhetoric, or, to follow Mr. Schaub's distinction, market integration is a direct enforcement criterion and not merely an ultimate or fundamental policy objective. The market integration objective has played a paramount role in formulating specific legal rules and doctrines and in deciding particular cases. Indeed, it is this objective, and not welfare or efficiency objectives, that underlie the E.U.'s unique position toward, among other arrangements, many vertical restraints and geographic price discrimination. Moreover, as discussed in the Forrester and Kirchner papers, the tensions between the market integration objective and welfare or efficiency objectives have given rise to vigorous criticism of E.U. competition law as sacrificing efficiency on the altar of market integration. The Commission has its outside defenders, however. For example, Prof. Waelbroeck argues that the tension between market integration and economic efficiency is only temporary, because once markets are fully integrated there will no longer be any reason to fear disparate governmental measures that produce long-term distorting effects.
356
Working Paper I
The E.U. also appears today to have one of the broadest lists of competition law objectives in the world, with perhaps the U.K. and Japan having somewhat comparable open-ended objectives. It is not only the market integration objective that distinguishes the E.U. objectives from those of other jurisdictions. Broader objectives, which are not mere rhetoric or simply fundamental/ultimate objectives (as opposed to 'enforcement criteria'), appear to be used in the E.U. to decide cases. Examples include: • • •
• • • • •
protection of small and medium sized firms; promotion of European champions and exports; promotion of a 'level playing field' between public and private enterprises or undertakings. This last objective is expressed in Art. 90 and can also be seen in other jurisdictions, such as Mexico; 'industrial policy' considerations, such as the 'competitiveness' of European industries; control over state-sponsored distortions of competition; employment effects1; cultural considerations (e.g. approval of arrangement conditioned on investment in European films); and environmental policy considerations, although Forrester concludes that the impact is difficult to see.
Eighth, it was recognised that inclusion as 'enforcement criteria' of noneconomic objectives (like environmental, regional and cultural objectives) and objectives only distantly related (if at all) to economics-oriented objectives (like promotion of national champions and 'competitiveness'—a shibboleth or largely vacuous concept) has important institutional enforcement implications. Centralised executive/administrative bodies probably are better capable than courts at implementing such a broad array of vague and inconsistent objectives. Moreover, in Europe, unlike perhaps any other place in the world, the debate about objectives has a profound impact on the institutional debate. The 'Jacobin school' argues that the longer and broader the laundry list of E.U. competition law objectives, the greater the power and discretion that should be given to a centralised 'integrated' (i.e. non-independent) enforcement authority, with considerable weight accorded the authority's 'expertise' by reviewing courts. Indeed, the Commission has invoked its role as implementer of broad objectives (well beyond consumer welfare or efficiency) in order to justify its See, e.g., Remia v. Commission [1985] E.C.R. 2545, para. 42.
6 - Competition Law Implementation at Present
357
non-independence, its primacy over Member State authorities and its union of investigative and decision-making functions. This contrasts with both the narrowing of objectives in many, if not most, other jurisdictions (including most of the E.U. Member States) and the independence and separation of powers in most other jurisdictions. In these respects, the E.U. appears increasingly isolated from the general situation throughout the world where the strong trend is toward competition 'law enforcement' regimes characterised by independent competent authorities, separation of investigatory powers from decisionmaking and a narrowing of objectives toward a heavy emphasis on welfare or efficiency objectives. The E.U.'s uniquely broad and varied list of objectives probably has had undesirable side-effects. First, economics may have become a bit lost in the crowd of other objectives. This is so at least in the sense that sophisticated economics analysis appears to have been unduly delayed at the E.U. level, when compared with the accelerating emphasis on economic analysis seen in competition authorities far younger than the almost 40-year-old Commission. The 7year-old Italian authority is one example. Secondly, the plethora of E.U. objectives makes formulation of enforcement priorities more difficult. For example, one might ask whether a more focused list of objectives would concentrate the Commission's attention and scarce resources on those agreements and practices (such as price fixing) about which there is a strong world consensus favouring tough competition law enforcement. Finally, the concept of 'legal certainty' frequently has played an unhelpful role, particularly at the E.U. level, in the debate about competition law objectives. Like 'Liberty', many sins have been committed in the name of 'legal certainty' which often is invoked to resist the desirable and appropriate introduction of economic analysis into competition law. 'Legal certainty' also has been invoked to resist an emphasis on or narrowing of the objectives to efficiency or economic welfare. Those jurisdictions (for example, Canada, Italy and the United States) in which efficiency and economic welfare are recognised and used as a paramount objective do not appear to have less legal certainty than the E.U. Bluntly speaking, legal certainty is often a shibboleth.
II Ian S. Forrester, Q. C.l Visiting Professor in European Law, Glasgow University Glasgow, U.K. Partner, Forrester Norall & Sutton Brussels, Belgium
Introduction The purpose of this paper is to examine the goals of competition policy. A striking feature of experience with Community competition law is that the main preoccupation of the law's principal enforcer is market integration, a political goal rather than one which would normally be attributed to an orthodox trustbusting authority. The economic wisdom of this goal, which has amounted at times to a near obsession, has frequently been in doubt. In retrospect, the penalties seem to have been disproportionate to the gravity of the offence viewed in non-political terms, especially when compared to the penalties for what, in competition terms, might have seemed more offensive behaviour. Although the emphasis on market integration has become less predominating, cross-border trading is still a hugely favoured economic activity, more encouraged in the E.U. than in any other jurisdiction. However, before repeating a familiar lamentation about the Commission's preoccupation with helping free riders, I begin by reviewing some other discernible competition law goals. The process of describing these may make the Commission's policy preoccupation with market integration easier to understand.
A. Review of Competition Policy Goals The Commission is not a free agent that has independently set its own goals. The point of departure must be the EC Treaty, as amended. Art. 2 calls for 'a 1
A number of the ideas set forth in this paper have been discussed in the annual reviews of competition law published in the Oxford Yearbook of European Law by Christopher Norall and me. I am grateful to my colleagues for ideas and suggestions. I express with pleasure my particular thanks to my colleague, Gregor Schneider, Assessor and Doctoral Candidate at the Europa Institut of the University of Saarbriicken, for his help.
360
Working Paper II
harmonious and balanced development of economic activities'; Art. 3(g) states that the activities of the Community shall include 'a system ensuring that competition in the internal market is not distorted'. The fourth recital to the Treaty records 'that the removal of existing obstacles calls for concerted action in order to guarantee steady expansion, balanced trade and fair competition'. Art. 85(1) does not provide much assistance in identifying goals, other than the obvious one of deterring the prevention, restriction or distortion of competition. Art. 85(3) contemplates giving the Commission's blessing to arrangements that may lead to: • • •
technical or economic progress; improvements in production or distribution; a fair share of the benefit for consumers.
The law is based upon a prohibition, which has been broadly interpreted, coupled with a wide-ranging power to exempt, which is rarely exercised in specific cases. This makes it difficult to say that the law favours particular social or economic goals other than those already identified in the basic texts, such as a fair share for consumers, economic progress and other bland objectives. It would be strange if Community competition law were as totally lacking in goals as the basic law would suggest. Thus, one would expect the decisions to provide clearer guidance. In individual Commission decisions, it is not easy to detect exactly how these goals are reconciled with other goals, including such notions as efficient allocation of resources, maximisation of prosperity or pure competition. Moreover, these decisions are not precise about the competition theories being pursued, or about difficulties encountered along the way. The Commission's annual reports are more reflective, and deserve to be quoted, but it is difficult to see how the policies espoused are implemented in actual decisions. The small number of individual decisions makes it difficult to trace how policies have been applied or pursued in reality. It is not clear whether E.U. policy favours protection of competitors or protection of competition. At times, when cases appear to present 'moral wrongdoing', the competition analysis can become less than rigorous, and the drafters condemn behaviour which may be economically rational and aggressive rather than abusive. There is always a tendency for lawyers to seek truth in the oldest texts, so I began by looking back at the annual Reports on Competition Policy, thefirstof which appeared in 1972. There, in the earliest days of modern Community competition policy, some familiar themes appear: that competition, on a Community-wide basis, is an essential vehicle for the achievement of economic
6 - Competition Law Implementation at Present
361
regeneration, and that it can be a means of regenerating stagnant economies. Above all, there was to be a parallelism between the action taken at Member State level to eliminate public governmental hindrances to cross-border trade and competition law enforcement to eliminate private contractual hindrances to cross-border trade. In the First Report on Competition Policy, there was even mention of competition policy as an instrument for fighting inflation, although I am far from sure that economists would say vigorous use of antitrust rules is a cure for inflation.2 The Commission was probably observing that the economic problems of the day could not be solved by national protectionism, and that competition makes for healthier enterprises. In the Second Report on Competition Policy, the Commission spelled out its ideas further: The Commission's measures against restraints of competition liable to maintain prices artificially high are part of the policy which must be pursued both by the countries and at Community level to combat inflation. Seen from this angle, the competition policy is a tool that must be used to create conditions under which the monetary and budgetary policies can have their full effect. Its first objective is to ensure that the markets within the Community are opened up so that purchasers can operate throughout the Common Market.
This is why the Commission has expressed its determination—noted by the Council—'to strengthen its action against restraints of competition which may derive either from horizontal price agreements, concerted practices with regard to prices and price discriminatory measures applied by undertakings in dominant positions or from market-sharing agreements or other restrictive practices pursued by undertakings the purpose of which is to maintain fragmentation of the markets, or from self-limitation agreements, wherever such agreements hamper the Community's commercial policy.' 3
1. The Creation of a Common Market The Commission stated in its Second Report on Competition Policy: Side by side with application of the banning principle the Commission continued its policy of strengthening the competitive position of undertakings by exempting the desirable forms of cooperation between them from the ban on restrictive agreements via regulations or case-by-case decisions. This line is followed with regard to European Commission, First Report on Competition Policy, 12 (1972). European Commission, Second Report on Competition Policy, 25, para. 16 (1973).
362
Working Paper II
specialization agreements and exclusive-marketing agreements, especially as far as small and medium-sized firms are concerned. In the distribution field the Commission has still had to combat the tendency on the part of certain undertakings to keep the national markets separated by prohibiting exports or by concerted price discrimination according to country of destination within the common market.4
In the Fifth Report on Competition Policy, the Commission stated: In times of economic difficulty competition policy must continue to make its influence felt alongside all the other Community policies. Its function is to preserve a situation in which the structural changes that are needed can take place. Although competition policy can make only an indirect contribution to solving the economic difficulties now besetting the Community—and then only if it achieves its objectives—there can be no solution without it.5 This was echoed the following year in the Sixth Report on Competition Policy: The illusion that economic and social problems can be solved either by Community or national protectionism, jeopardizing the unity of the common market, cannot be maintained. Competition policy is one of the fundamental means for preserving the unity of the market. Its aim is to ensure that business operates along competitive lines, while protecting the consumer by making goods and services available on the most favourable terms possible. It therefore endeavours to cut monopoly profits, to ensure that the economy remains adaptable to circumstances and to stimulate innovation. In Community terms competition policy has to play an ancillary role to other policies, essentially to preserve or restore conditions in which competition can flourish.6 In short, the Commission was consistently saying that competition policy must be a key part of the policies underlying the creation of a common market.
2. Market Integration If one could review 10,000 pages of advice on E.C. competition law rendered by lawyers over the past few years, I predict that many portions would analyse not only whether cross-border trade would be limited by the arrange4 5 6
European Commission, Second Report on Competition Policy, at 15, para. 2 (1973). European Commission, Fifth Report on Competition Policy, 7, Intro. (1976). European Commission, Sixth Report on Competition Policy, 9, Intro. (1977).
6 - Competition Law Implementation at Present
363
ments in question, but also whether such trade might be said to be limited by a complainant. This is especially true in the context of distribution, but also of technology licences and other fields. Thus, prudent lawyers become very defensive and conservative when there is a risk that cross-border trade could be hindered. When I was a young competition lawyer, acting for complainants, I believed (in accordance with my legal upbringing) that hindering cross-border trade was the most wicked thing one could do. As an older competition lawyer representing defendants, I found that the Commission's preoccupation with smiting export bans was perverse or obsessive. However, the policy was unhesitatingly justified by the Commission as part of its mission to achieve an integrated common market. To quote the introduction to the Ninth Report on Competition Policy: The first fundamental objective is to keep the common market open and unified. The metamorphosis of a heterogeneous collection of isolated national markets into a single vast market could not succeed without the establishment of some basic rules. There is accordingly a continuing need—and this is the primary task of the Community's competition policy—to forestall and suppress restrictive or abusive practices of firms attempting to divide up the market again so as to apply artificial price differences or impose unfair terms on their customers. European Commission, Ninth Report on Competition Policy, 9, Intro. (1980). The formidable non-contractual obstacles to cross-border trade and the formidable differences in market conditions were not denied, but the Commission consistently refused to let them be reinforced by private barriers. DG IV, in a sense, obliged private business reluctantly to participate in creating the common market by forcing it to ignore governmental barriers to trade and discrepancies in market conditions between different Member States. The profits of the supplier might be reduced if the parallel trader were allowed to buy in a cheap Member State and resell in a high-priced Member State without contractual interference from the supplier. The supplier could legitimately react by bringing his/her prices closer together, or by putting a volume ceiling on deliveries to the parallel trader, but he/she could not prohibit the trader from making supplies to the unwelcome exporter. A huge part of the Commission's enforcement efforts concentrated on making contractual barriers to cross-border trade imprudent and potentially costly. Up to the early 1980s, cases involving distribution problems were predominant among all cases in which fines were imposed. One consequence of this political objective
364
Working Paper II
was that, in contrast to other antitrust agencies, the Commission seemed to be soft on horizontal arrangements between major industrial enterprises. It was well known that, in certain sectors of Community industry, cartels existed but were not challenged by the limited Commission resources available. In the late 1980s, a new atmosphere began to emerge. Competition policy seemed gradually to be coming to reflect a more realistic vision of an integrated common market. Competition law enforcers, economists, businesses and even politicians seemed to be pursuing a common goal. Individual decisions concerning horizontal cooperation, especially in high technology sectors, were driven more by encouragement of innovation and less by fear of cartelisation. The decisions were still drafted in an artificial manner, whereby ten theoretical restrictions of competition were followed by five technical advantages justifying an exemption. But the feeling in the air was definitely changing; and in one or two instances, the decisions themselves were commendably radical. There was also evolution on the external trade front. As the momentum towards an integrated Common Market built up, the Community was compelled to make coherent its external relations with third countries, notably Japan, then the world's leading supplier of electronic equipment and technology. This success occasioned severe trade tensions during the early 1980s. Did Japanese companies represent a threat to European industry, or a valuable resource whose technological capacities and cooperation were essential to Europe's efforts to catch up in the technology race? Several leading European manufacturers complained about unfairly cheap (dumped) Japanese exports, and simultaneously entered into long-term cooperation agreements with the targets of their complaints. DG IV had reservations about the enthusiasm of DG I for antidumping and other protective measures, and it is possible that DG IV's encouragement of cooperation between Japanese and Community enterprises reflected these reservations. {CanonlOlivetti1 is an example of the Commission blessing a relationship with a foreign investor/technology partner which was at the same time the target of an antidumping complaint by its partner). In any event, the Community decided politically that the market was to be as freely accessible to competition from third countries as individual Member State markets were to be accessible for competition from elsewhere in the Community. As the barriers to European competition were removed, the Community refrained from erecting new barriers at its external frontiers, and maintained its traditional free-trading stance. However, it had uneven success in persuading the Member States to agree.
7
1988O.J. (L52)51.
6 - Competition Law Implementation at Present
365
The absence of an environment that assumes and encourages vigorous competition ultimately weakens the enterprises that appear to be benefiting from the situation. The progressive feebleness during the 1980s of the French car industry demonstrates this phenomenon. In 1979, the President of the Republic announced that imports of Japanese cars would be limited to not more than 3 per cent of annual new car registrations. This spectacular attempt to defy economic gravity seems to have done little good for Simca, Peugeot, Renault or Citroen, which were deprived of the stimulus to innovate and keep their customers happy because the big threat of those days, the Japanese export tide, was to be diverted to other shores. Great ingenuity was deployed by the French administration to maintain French roads largely free of Frenchregistered cars made-in-Japan. If too many cars of one brand were registered in a particular month, the importer of that brand was punished by having to wait a few weeks for its next type-approval certificate permitting, for example, a new design of carburettor. Consumers responded by deploying many ingenious schemes to bring cars into France from other Member States, including complaining to the European Commission. However, although the facts were notorious, the Commission was unable to achieve consensus to take public legal action against this gross distortion of the marketplace. The risk of losing a policy battle with the French authorities was too great. This was one of the great failures, even scandals, of E.C. policy in the 1980s. Responsibility is certainly not to be attributed only or even principally to DGIV, whose senior officials and Commissioners favoured action. The state of the French automobile industry today demonstrates the consequences of protecting a domestic market, and the tolerance of the European Commission demonstrates that it has had varying degrees of success in translating its goals into action. Another example of the consequences of a protected domestic market is the insurance industry. From the end of the Second World War, only banks and insurance companies could conduct pension fund management in Japan. Insurance company premiums for all non-life insurance contracts were set by a rating agency composed of Japanese insurance companies, which compiled the raw data on all-risk and non-risk factors, casualties, and the like, and produced, in effect, recommendations to insurers on their premiums. Thus, all components of the risk were known, calculated and conveyed, after the inclusion of a 5 per cent profit margin, to the insurance companies, which simply applied them in their premiums. This activity was exempted from the Japanese anti-monopoly law. The competition environment was indeed very placid. European insurance companies, accustomed to a profit margin in the Community of approximately 1 per cent, had little basis on which to compete because there was no rate freedom and no real product differentiation. During
366
Working Paper II
the boom years of the 1980s, the weakness of the financial system was not troublesome; but when the bubble burst, the combination of bad debts, poor investments in rigidly determined funds and hidden losses (Japanese balance sheets not reflecting actual market value for a devalued piece of land) proved very damaging. The Japanese financial economy was hollowing out (like the industrial economy following the appreciation of the yen). The politicians reacted to the crisis not by protective measures, which had failed in the past, but by adopting a plan which would allow the progressive emergence of tougher competitive conditions, permitting hungry foreign investment advisers to enter the market. After modest concessions in favour of foreign investment advisers, policymakers recognised that more radical steps were needed, covering the entire financial services sector. The foreigners did a better job of managing the assets with which they were entrusted than their Japanese rivals. They were better at judging market opportunities through their experience in a highly competitive marketplace, where profit margins were much lower. Once the dust has settled, the marketplace probably will be more lucrative for Japanese and foreign service-providers alike. Thus, in Japan, on the one hand we have an example of ferocious competition, totally deregulated, between the members of a strong domestic industry (consumer electronics and cars) exporting to Europe and making in Europe the products developed on the domestic battleground. On the other hand, we have a clear example of what happens when producers, markets and consumers lose touch with each other. It is easy to draw a moral conclusion to the effect that competition is desirable, and protectionism does not pay. However, it is not so easy to draw conclusions about E.C. competition policy over the past twenty-five years. DG IV consistently spoke up for competition, consistently seemed to support competition in its annual Report on Competition Policy, yet its decisions were few in number, and by no means covered the entire field where Commission intervention was desirable. To use the familiar metaphor, if competition policy was a tree whose branches were constituted by decisions, it was seriously lopsided. The style of drafting also left something to be desired: instead of frankly acknowledging the real difficulties, decisions were often sonorous recitals of undeniable truths. A classic example is Bayer/Gist Brocades, where the Commission stated: The fundamental principle in this respect, established at the time the common market was formed, lays down that fair and undistorted competition is the best guarantee of regular supply on the best terms. Thus the question of a contribution to economic progress within the meaning of Art. 85(3) can only arise in those excep-
6 - Competition Law Implementation at Present
367
tional cases where the free play of competition is unable to produce the best result economically speaking.8
Today, DG IV's heart is in the right place, but its decisional output (other than in the field of major mergers) is an inadequate guide to its heart's sentiments. It is not easy to point to examples where consumer protection or environmental considerations can readily be detected as the motivating forces behind a decision.
3. Policy Considerations in Commission Decisions a. Consumer Welfare and Allocative Efficiency Each favourable exemption decision certainly states (as it must under Art. 85(3)) that consumers will derive some benefit, but the effect on the consumer is usually rather indirect. In 1994, Commissioner Van Miert articulated the link as follows: Another crucial competition policy objective is consumer protection. The single market must first and foremost serve people. It must be ensured, through strict application of the competition rules, that consumers have freedom of choice between quality products at competitive prices.
As has already been noted, a true internal market does not exist in a number of sectors in which competition is restricted, or indeed prevented, by exclusive or special rights granted to enterprises performing certain services. Such services involve the fundamental requirements of all European citizens (for example, telecommunications, energy and postal services). A true internal market cannot be said to exist if, in daily life, firms and individuals in the Community have no choice but to deal with national enterprises in satisfying some of their essential requirements.9
Thus, consumer benefit can be detected in the opening up of choices, even costly ones, to consumers. In Yves St. Laurent, a distribution system limiting fine perfumes tofineshops received a formal exemption as it was:
8
1976O.J.(L30)19. European Commission, Twenty-Fourth Report on Competition Policy, Intro. (1994). 10 1992O.J.(L12)24, 29. 9
368
Working Paper II
intended merely to safeguard, in the public's mind, the aura of prestige and exclusivity inherent in the products in question, thus preventing any association with lower-quality goods.10
The Commission held that the conditions to satisfy Art. 85(3) were present. The rationalisation for what might seem a slightly surprising conclusion is set forth at length: The manufacturer's capacity to create and maintain an original and prestigious brand image is thus a key factor in competition. It follows that a luxury cosmetics brand must be distributed on an exclusive basis. Experience shows that generalized distribution of a luxury cosmetic product can affect the consumer's perception of it and in the long term reduce demand for it.''
The manufacture and distribution of the products were thus enhanced, ensuring that they are distributed under conditions that maintain the image and exclusivity required for luxury cosmetics. As to benefit for consumers from the system: The consumer is thus assured that the luxury product will not become an everyday product as a result of a downgrading of its image and a decrease in the level of creation.12
By contrast, the Commission took a hostile approach to justifications for the high prices of luxury goods in HennessylHenkell, where an exemption was refused for a distribution system involving constraints on price competition: even if the products in question are considered, as Hennessy considers them, to be luxury products. Hennessy stated in its letter of 3 November 1978 that the purpose of Art. 6 of the agreement is to ensure that its products 'which are regarded as luxury products, do not become subject to cut-price selling which would lead to business anarchy'. The Commission has already stated, however, when considering selective distribution arrangements in the luxury perfumes industry, that 'the luxury character of a product could not in itself be regarded as an adequate ground for exemption under Art. 85(3)'.13
Consumer benefit in other cases seemed rather uncertain. For example, a newly-published joint venture decision concerning cooperation between Iridium and Motorola to implement a global digital wireless communications service was blessed, along with pricing guidelines for future gateway operators: 11 12 13
1992O.J. (L12)24, 31. 2d. at 33. 1980O.J. (L383), para. 32.
6 - Competition Law Implementation at Present
369
the principle of uniform prices and other conditions in different territories . . . seems appropriate to fulfil customers' needs.14 A critic could say that customers must tolerate consciously parallel rate fixing as part of the price to be paid for making European industry competitive. This is probably unfair. The explanation may simply be that the Commission considers that the consumer welfare criterion is not very important when considering high-technology alliances, especially when these will create European networks, and may briefly be disposed of by a reference to access to better goods and services. There is little trace of real concern for the welfare of consumers as direct beneficiaries of competition policy. There is, however, more concern for citizens as indirect beneficiaries, residing in a healthy economy. In Magill, it was relevant to the Commission and the Courts that an Irish consumer wishing to plan his television viewing on a weekly basis was obliged, for lack of a Magill multi-channel television guide, to buy three single broadcaster guides: The publication of the Magill TV Guide, albeit brief and of a limited print run, also clearly demonstrates the consumer demand for a comprehensive weekly TV guide in the area concerned. * * * In this connection the Commission considers that the three undertakings are perfectly capable, having regard to their present position and experience on the market, of playing a major role on the market for comprehensive weekly TV guides, if they so wish. Alternatively they may continue to publish individual TV guides on a market where comprehensive TV guides are available if they consider the consumer to be best served by their own guides, as they have stated. By limiting the scope of their licensing policies so as to prevent the production and sale of comprehensive TV guides, however, they restrict competition to the prejudice of consumers.15 The difficulty of finding a benefit for consumers was confronted in the Synthetic Fibres crisis cartel decision,16 where the ten largest producers mutually agreed to a mass reduction of capacity. In assessing consumer benefit, the Commission took a very long-term view based upon the advantages to the consumer of improved production, and a healthier and more competitive industry producing better and more consumer-oriented products. Consumers would be protected against price rises by virtue of a large number of producers (both 14 15 16
Iridium, 1997 O.J. (L16) 87, 94. Magill T.V. Guide/I.T.P, B.B.C. and R.T.E. 1989 O.J. (L78) 43, 49-50. Synthetic Fibres Agreement, 1984 O.J. (L207) 17.
370
Working Paper II
signatories and non-signatories) remaining in the market. The risk of shortterm price increases arising from capacity reduction was acceptable, given the supposed market strength of the consumers (textile manufacturers) who were subject to price pressure on these markets, and given the existence of nonsignatory producers (both E.C. and non-E.C). This reasoning is open to question. That the consumer is a producer on a vertically related market, and is itself subject to economic difficulties, hardly seems a reasonable ground for assuming that it will be able to resist price increases. Nor does this constitute a ground for assuming that it will benefit, at least in the short term, from the restructuring. Moreover, although the Commission pointed to E.C. non-signatories and non-European competitors as alternative sources of supply, it failed to appreciate that membership was, in fact, open to all European producers, and that vis-a-vis non-European producers, transport costs might well eat away much of the saving. b. Protection of Small or Medium-sized Enterprises, Individual Traders and Entrepreneurs, and Consumers
The competition rules have sometimes been used to help the small trader, such as a repair shop cut off from necessary replacement parts (HuginlLiptons11) or the family grocer displeased by an export surcharge imposed by a large whisky producer (Bulloch/DCL18). Kawasaki19 was precipitated by a mother who wanted to order motorbikes for her sons, and in the process caused an export ban to be revealed. ICI/Zoja/CSC20 involved the protection of a relatively small company threatened with elimination, because it was cut off from supply of an indispensable chemical. Thus, there is certainly E.C. sympathy for complainants, but it is rare for individuals to be mentioned. In the case of passenger cars, there is a unique statutory right in favour of the consumer. Reg. 123/8521 instituted a regime whereby a consumer could name an intermediary to act on his behalf in another Member State, placing an order for the model of the consumer's choice at the (presumably lower) price prevailing in that other Member State. The car manufacturer's distribution network could not, on pain of losing the benefit of the block exemption, refuse to supply the order tendered. This has been renewed in the successor Regulation.22 The battle between Eco System and 17 18 19 20 21 22
Hugin/Liptons, 1978 O.J.(L22) 23. Distillers C o . Ltd., 1978 O.J. (L50) 16. K a w a s a k i , 1979 O.J. (L16) 9. 70 J A / C S C - I C I , 1972 O.J. (L299) 5 1 . Reg. 123/85, 1985 O.J. (L15) 16. R e g . 1475/95, 1995 O.J. (L145) 25.
6 - Competition Law Implementation at Present
371
Peugeot 23 was one of the by-products of the legislation. Eco System recruited hundreds of consumers, each of whom mandated it to order a new car in a Member State where prices were lower than in France. Although, in actual practice, it is not clear how many cars were sold in this manner, the initiative had a large impact in the marketplace. It is not clear that D G I V is truly helping small businesses, other than in the important respect of making them welcome as complainants, which indeed is a big exception. The risk of trouble through prohibiting parallel trade is that an unhappy dealer or customer will send copies of the incriminating correspondence to the Commission. Indeed, some traders placing unwelcome orders routinely send a copy to D G IV. Small businesses are not well served by the difficulty of completing Form A/B, a daunting task. Small firms are more likely than very large firms to address themselves to small non-specialist law firms, whose lawyers may wrongly, but reasonably, also believe that there is an obligation to notify any international exclusive agreement unless it precisely fits the criteria of a block exemption. c. Market Integration This extremely important subject has been addressed above. One further quotation from a Report on Competition Policy is added here: Key importance is attached to developing the interaction between competition policy and establishment of the internal market. Such interaction springs from the fact that these Community policies serve the same fundamental objective—reinforcing the wealth-creating capacity of the Community economy through improved allocation and more efficient use of productive resources.24 There is perhaps also merit in quoting the parallel aspirations of the Cockfield White Paper on the Completion of the Internal Market: We recognise that many of the changes we propose will present considerable difficulties for Member States and time will be needed for the necessary adjustments to be made. The benefits to a n integrated Community economy of the large, expanding and flexible market are so great that they should not be denied to its citizens because of difficulties faced by individual Member States. These difficulties must be recognised, to some degree they must be accommodated, but they should not be allowed permanently to frustrate the achievement of the greater progress, the
23 24
Eco System/Peugeot, 1992 O.J. (L66) 1. European Commission, Twenty-Third Report on Competition Policy, 88 (1993).
372
Working Paper II
greater prosperity and the higher level of employment that economic integration can bring to the Community.25
d. Environment, Health, Safety
It is difficult to say that environmental policy as such has had much impact on competition policy. However, the Commission has occasionally accepted the argument that some level of cooperation is appropriate between companies facing strict environmental norms.26 The question has also arisen whether cooperation within an industry in pursuit of green goals could exclude foreign competitors. In SpalMonopolelG.D.B. ,27 a complaint was at issue concerning a pooling system among German bottled water producers covering reusable bottles. Under pressure from the Commission, the relevant Genossenschaft allowed non-German water producers to enter the pool. The deals may have been approved even if they concerned something other than green issues, but the green credentials of the system unquestionably helped. In V. O. T. O.B.,28 tank storage companies had agreed that the cost of maintaining tanks without gas emissions was to be recouped by passing on to customers a uniform fixed amount reflecting the extra cost of installing new safety equipment. The Twenty-Second Report on Competition Policy stated: The case makes clear that the Commission is not opposed to the possible passing on to the customers of 'polluter pays' investment costs, since this makes them more aware of environmental problems and their implications. However, customers should not be barred from challenging price increases and shopping around for the smallest increase.29 e. Industrial Policy; Enhancing Competitiveness; Helping European Enterprises Compete Internationally (Promoting Competitiveness); and Promoting Exports
There has been much discussion about whether E.C. or national industrial policy should concentrate on picking winners or protecting losers. Conventional 25
Cockfield White Paper, Com(85) 310, 7, para. 14. See, e.g., Duales System Deutschland, European Commission, Twenty-third Report on Competition Policy, 94 (1993); IFCO Id. at 95; Press Release IP(93)430; (dealing with the sorting and recycling of packaging waste and the standardisation of reusable fruit containers, where an industry-wide solution seemed appropriate). 27 European Commission, Seventeenth Report On Competition Policy, 70 (1987). 28 European Commission, Twenty-Second Report on Competition Policy, 106 (1992). 29 European Commission, Twenty-Second Report on Competition Policy, 106, 108 para. 185 (1992). 26
6 - Competition Law Implementation at Present
111
wisdom is that it is better to establish conditions in which competition can flourish than to leave it to enterprises to compete. The Commission offered a view on industrial policy in its communication, Industrial policy in an open and competitive environment30: Industrial policy rather seeks to create the essential conditions for the rapid development of an efficient Community industry... the idea is for the Community to act as a catalyst in encouraging innovation and creating an appropriate and stable environment. Seen in this way, the main achievement is certainly the completion of the internal market, which should allow the Community industry to benefit from economies of scale, while at the same time exposing it to greater competitive pressures, which can only boost its productivity.31 In 1991, the Commission was aware of the risks of allowing industrial policy considerations to distort competition law analysis: [T]he Commission is favourable to cooperation between companies and restructuring of an industry in order to create efficiencies and to attain the optimal scale to produce and market within the Community and worldwide. It takes care to implement its policy with due regard to the interest of industry in a stable environment where excessive bureaucracy and unnecessary intervention are avoided. [However,] artificial concentration of resources in one or a few companies to face powerful competitors from third countries is not without risk. If insufficient attention were paid to ensuring effective competition on Community markets, European consumers would normally have to pay higher prices on account of the lack of effective competition on home markets. In addition, experience has shown that, in most cases, maintenance of effective competition on EC markets does not normally preclude the maximum exploitation of economies of scale.32 By the time the Twenty-Fourth Report on Competition Policy was published in 1994, doubts about industrial policy considerations had dispersed. As shown above, the Commission sees competition policy as an instrument of industrial policy, and: intends to maintain a favourable approach to forms of cooperation that strengthen the efficiency and thus the competitiveness of the parties. The opening-up of markets following the completion of the internal market programme and the liberalization of trade at world level, on the one hand, and the need for restructur30 31 32
European Commission Doc. COM(90)556. European Commission, Twenty-Third Report on Competition Policy, 90 (1993). European Commission, Twenty-First Report on Competition Policy, 42-3 (1991).
374
Working Paper II
ing in order to emerge from the crisis, on the other, call for efforts on the part of firms.33 It is easy to say the right thing, more difficult to do the right thing, when considering whether it should be legal for two major enterprises to collaborate in order to be stronger in facing global competition. One source of the trend towards liberalisation of the rules on joint ventures lies in a belief that gained ground in the dark years of the early 1980s. This was the belief that cooperation and restructuring, involving even very large companies with considerable power on the E.C. market, might be beneficial or even indispensable for the survival and restructuring of European industry. During the 1980s, the blessing of cooperative arrangements was achieved by very cautious decisions. However, in 1990, five interesting decisions were issued, two of which seemed to be adventurous and break new ground, three of which34 were more conservative. The first of the more adventurous decisions was ElopaklMetal Box-Odin?5 Odin was a joint venture company created to develop a new kind of container for U.H.T. foods, to develop the machinery and technology for filling the containers, and to produce and distribute the containers and their filling machines. One of the parent companies, Elopak, was a manufacturer and supplier of U.H.T. cartons. It had also supplied U.H.T. filling machines manufactured by others, but this arrangement had recently been terminated. It acquired an American manufacturer of fresh foodfillingmachines, which was trying to develop U.H.T. filling machines. The other parent company was the Metal Box Group, a much larger company with a number of businesses including packaging, notably metal cans. One of its packaging products was an aseptically filled polypropylene container with an aluminium top, used for various products including U.H.T. milk. The Commission found that the agreement was not caught by Art. 85, and granted negative clearance. Specifically, it found that Elopak and Metal Box were not competitors, actual or potential, in the relevant product market, and the development of the product by either party on its own was highly unlikely. This conclusion was defensible, but the intellectual route followed by the Commission in giving its blessing was somewhat startling: Neither party could in the short term enter the market alone as such entry would require a knowledge of the other party's technology which could not be developed without significant and time-consuming investment. 33
Euro