INTERNATIONAL ANTITRUST IN THE 21ST CENTURY:
I am also very pleased and a bit awed to be speaking before colleagues from 55 countries from every part of the world. For roughly 30 years, the OECD, through its Competition Law and Policy Committee, has played a crucial role in building consensus among OECD members on a wide range of antitrust and competition policy subjects. This Global Forum promises to make important new contributions to mutual education and understanding among an even wider range of antitrust authorities in developed and developing countries alike.
In my remarks this morning, I will briefly review the history of the past decade in international antitrust enforcement. As you know, over this period there has been explosive growth in the number of countries with antitrust laws and agencies. This growth has amplified the importance of cooperation among antitrust agencies in ensuring sound antitrust enforcement in an increasingly global marketplace, and we have made good progress toward such cooperation. I will then examine, now that we have some practical experience in working together, some additional steps we might take to enhance our cooperative efforts in order to combat anticompetitive behavior more effectively and more efficiently.
Our experience with the proposed General Electric/Honeywell merger demonstrates, however, that close cooperation and goodwill between antitrust agencies does not guarantee consistent results in individual cases. A good working relationship cannot overcome significant differences in views about the proper scope of antitrust law in national and world markets.
In the early 21st century, therefore, antitrust agencies should begin to discuss two types of issues, in a detailed and sustained way, in both bilateral and multilateral contexts. First, we need to address certain practical law enforcement issues, especially in the areas of anti-cartel enforcement and merger review. Second, we must begin to address important substantive issues, as the OECD's Competition Law and Policy Committee will do later this week at its roundtable on portfolio effects. We believe that greater substantive and procedural convergence can be promoted by forming a Global Competition Network -- which we in the United States have previously called the Global Competition Initiative -- not to duplicate but to supplement the work of OECD and its Global Forum on Competition. I will close my remarks by setting forth our vision for the new Network.
The Rise of International Antitrust Cooperation
There was a time, not so many years ago, when few countries had antitrust laws and fewer still enforced them. (Indeed -- and this strains the imagination -- there was a time when there were very few international antitrust conferences.) But during the past decade, market principles, deregulation, and respect for competitive forces have been broadly embraced, and many countries have created antitrust laws and agencies that are committed to enforcing them. Over 90 countries currently have antitrust laws of some sort, and roughly 20 more countries are in the process of drafting such laws; all of the nations represented here today are in one or the other of these categories. During the past decade, the Department of Justice and our counterparts abroad have investigated and prosecuted many international cartels, and an increasing number of antitrust agencies review many of the multinational mergers that characterize our global economy. All of these developments support the popular wisdom that increased cooperation between and among antitrust agencies is essential. But as important as cooperation is, it is sometimes quite difficult to achieve.
People have been thinking and talking about international antitrust cooperation for a long time. The Revised OECD Recommendation on antitrust cooperation,(1) which was most recently amended in 1995, is merely the latest in a series of similar OECD Recommendations reaching back to 1967. It has only been in the last decade, however, that the United States federal antitrust agencies -- the Department of Justice and the Federal Trade Commission -- have begun to gain significant practical experience in working with our antitrust colleagues around the world. We have learned that the range and complexity of international antitrust issues requires that we use a variety of cooperation tools.
The simplest and most common of these tools is informal communication between antitrust agencies. While there are important statutory and prudential limits that constrain our ability to share confidential information with colleagues in foreign antitrust agencies, there is a wealth of useful nonconfidential information that can be and is shared.
Technical assistance is another important tool for international cooperation, and the United States has been an enthusiastic provider of such assistance. We sometimes joke that antitrust has been one of the United States' most successful exports. During the past decade, we and the FTC have sent nearly 250 missions to dozens of countries on six continents -- including nearly all of the non-OECD countries represented here today -- on both short-term trips and long-term advisory missions of six months or more. We have hosted hundreds of foreign antitrust officials on visits and internships to the Division and the FTC in order to share what we do and why we do it. And we have participated in many of the valuable conferences organized by the OECD, the WTO, and UNCTAD for antitrust officials from developing and transition countries, including, most recently, last month's workshop in Beijing of the APEC-OECD Cooperative Initiative on Regulatory Reform.
A third important tool is the bilateral antitrust cooperation agreement. We have entered into such agreements with Australia, Brazil, Canada, the European Commission, Germany, Israel, Japan, and Mexico. These agreements have been very useful, both for us and for our partners, and more such agreements are on the way. In international cartel matters, the Antitrust Division also relies on the United States' mutual legal assistance treaties (MLATs), which we now have with nearly 40 countries.(2)
Finally, I would be remiss to stand in this room and fail to mention the wonderful work that the OECD has done in promoting antitrust cooperation. Over the years, OECD members have learned from one another in roundtables on subjects ranging from competition in road transport, to corporate leniency programs in cartel cases, to this week's discussion of portfolio effects. And OECD discussions have led to important successes in convergence, such as the 1998 Council Recommendation on Effective Action Against Hard Core Cartels.
To summarize, ten years ago we thought that antitrust cooperation should work in theory. Now we know it works in fact, and will continue to form a vital part of an antitrust agency's toolkit.
When Cooperation Is Not Enough: the Lessons of GE/Honeywell
A decade of sustained cooperation has yielded a fair amount of substantive convergence among antitrust agencies around the world with respect to both cartels and mergers. Despite different verbal formulations in our various antitrust laws, agencies have tended to reach similar conclusions on matters when we become fully engaged with one another on the analysis and are working from a common set of facts. Indeed, there have been days when we thought (or hoped) that such cooperation itself would eventually minimize or resolve even the most serious areas of antitrust divergence. More recently, however, we have come to understand that cooperation alone will not resolve some significant areas of difference among antitrust regimes that must be addressed if we are to maintain the integrity of antitrust on a global stage.
After reviewing the recent proposed $42 billion merger of General Electric and Honeywell, the Justice Department cleared the merger, while requiring divestiture to address competitive concerns in two markets. But the European Commission, analyzing identical product and geographic markets and having access to the same facts we did, blocked the transaction in its entirety.
The U.S. and EU agencies reached inconsistent decisions despite a tremendous amount of coordination over several months, made possible by the parties' waiver of their confidentiality rights. In fact, I do not believe that we could have worked together more closely. Our staffs talked on the phone frequently and had extensive meetings in Washington and Brussels; the EC staff had access to our economic expert; and we had extensive substantive discussions at the very highest policy levels about the evidence and the theories the two agencies were pursuing. The glaringly inconsistent decisions, then, were not the product of a failure of cooperation or a lack of effort by either agency to ascertain the other agency's point of view.
The differences between the Justice Department and the EC flowed from an apparent substantive difference, perhaps a fundamental one, between the two agencies on the proper scope of antitrust law enforcement. We concluded that the merged firm would have offered improved products at more attractive prices than either firm could have offered on its own, and that the merged firm's competitors would then have had a great incentive to improve their own product offerings. This, to us, is the very essence of competition, and no principle is more central to U.S. law than that antitrust protects competition, not competitors.(3)
In stark contrast, the EC focused on how the merger would affect European and U.S. competitors, essentially concluding that the very efficiencies and lower prices the transaction would produce would be anticompetitive because they might ultimately drive some of those competitors from the market or reduce their market shares to a point where they could not longer compete effectively. In other words, the EC determined that the fact that customers would be "induced" to purchase more attractive and lower-priced GE/Honeywell products, rather than those of its competitors, was a bad thing of a sort that its antitrust law ought to prohibit.
This difference in approach to analyzing mergers under U.S. and EU antitrust law is significant. Under U.S. law, we believe that the purpose of the antitrust laws "is not to protect business from the working of the market; it is to protect the public from failure of the market."(4)
Indeed, the competitive process is largely about encouraging the more efficient to grow at the expense of the less efficient. This process inures greatly to the benefit of consumers. Firms are rewarded for cutting costs, lowering prices, and in the process displacing their rivals. Such competition sometimes means that inefficient rivals are driven out of business, but even if they are, consumers are better off overall. Our experience, however, is that business rivals rarely go quietly into the night. Instead, they typically respond by lowering their own costs and prices, competing harder to survive.
In our view, the so-called "portfolio effects" or "range effects" analysis as it has recently been employed is neither soundly grounded in economic theory nor supported by empirical evidence, but rather, is antithetical to the goals of sound antitrust enforcement. We fear that it will result in some procompetitive mergers being blocked, and others never being attempted, to the detriment of consumers in many countries. It will dissuade merging parties from talking candidly to antitrust agencies about the efficiencies they expect to realize, out of fear that such efficiencies -- even when they would clearly benefit consumers -- would be viewed negatively.
What are we going to do to address this difference of view about what goals antitrust laws should serve? Several things. First, of course, antitrust agencies should continue to recognize the many areas in which we do agree, and cooperate even more in those areas. Second, the U.S. agencies and the EC have agreed to take up this issue in our bilateral mergers working group, where we will educate one another about our respective views and, working at both staff and senior policy levels, try to reach some common ground. Finally, we believe that the "portfolio effects" theory is an excellent candidate for broader public discussion, including in appropriate multilateral fora; indeed, it will be discussed at OECD on Friday, at a roundtable offered by the Competition Law and Policy Committee. At some point, it might also be taken up in the Global Competition Network, a subject to which I now turn.
Towards a Global Competition Network
Strong cooperative relationships between and among antitrust agencies necessarily will continue to be an integral part of vigorous law enforcement efforts. But they are not a panacea for every issue that arises out of the globalization of antitrust. As I have already indicated, antitrust enforcement is no longer the concern of a handful of highly developed countries, nor even of the thirty OECD member countries. In order to achieve truly global convergence on important enforcement issues, multilateral efforts must supplement bilateral ties.
Perhaps because I have spent much of my professional life either investigating or defending mergers and acquisitions, I believe that the need for a new multilateral exercise to deal with unaddressed issues is most apparent in the merger area. Well over 60 jurisdictions around the world already have premerger notification regimes, and this number is likely to increase. Very few of these agencies review only local matters, even though their focus necessarily is on a merger's impact in their national economies. Rather, because both markets and firms are becoming increasingly global, antitrust agencies increasingly are finding that they are reviewing mergers that are also being reviewed by five, ten, or twenty other agencies around the world.
When transactions are reviewed by multiple authorities, the risk of substantive and procedural conflicts can increase dramatically, and effective cooperation among a large number of agencies can be extraordinarily difficult. On the substantive side, the potential for inconsistent outcomes increases substantially. On the procedural side, the burdens, costs, and uncertainties associated with filing in and dealing with a large number of reviewing jurisdictions pose serious concerns for the international business community. Among other things, they may discourage, unduly delay, or at best, constitute a tax on efficient, consumer-friendly transactions. These are difficult issues that may not have easy solutions, and certainly cannot be resolved unilaterally or through bilateral efforts alone.
Many antitrust officials and members of the antitrust bar have come to believe that these enforcement issues can best be resolved, not in any existing organization, but in a new one that is focused exclusively on the procedural and substantive issues directly affecting multijurisdictional antitrust enforcement, a task for which no existing organization has both vocation and mandate. We support the proposed Global Competition Network (GCN) because we believe that it can serve as exactly this type of problem-solving vehicle. I have spent a great deal of time during my first few months at the Antitrust Division thinking about the GCN and consulting with Tim Muris, Mario Monti, Konrad von Finckenstein, and others about how to get the GCN moving. We hope to "launch" the GCN next Spring.
In my view, the GCN should be a venue where senior antitrust officials from developed and developing countries formulate and develop consensus on proposals for procedural and substantive convergence in antitrust enforcement. (I would not exclude any antitrust agency that is prepared to participate meaningfully in the GCN's work.) The GCN's general approach to issues should be as practical and concrete as possible; it should avoid abstract discussions that are unlikely to lead to improvements in the practice of antitrust enforcement. Unlike OECD, WTO, and UNCTAD, the GCN would not be a "bricks-and-mortar" organization with a permanent secretariat, and it will not deal with trade issues, or even non-antitrust issues that could conceivably be included under the rubric of "competition policy." As I've stated previously, it would be all antitrust, all the time.
GCN meetings would provide a structured dialogue by focusing on only two or three projects at a time. As indicated, I believe it would be appropriate to start with some merger process issues, among other things. These projects would be aimed at developing to non-binding general guidelines or "best practices" recommendations. Where the GCN reaches consensus on particular recommendations, it would be left to governments to implement them voluntarily, through unilateral, bilateral, or multilateral arrangements, as appropriate.
The private sector should play an important role in the GCN. However, because the goal of the GCN is to promote convergence of government procedures and enforcement policies, it would not be appropriate for the private sector to be involved in the GCN's decision-making functions. But I would expect that legal and economic antitrust practitioners, academics, and businesspeople will help the GCN to identify projects, participate in GCN information-gathering exercises, and share their views on how GCN projects should proceed and where they should lead. In addition, I hope that international organizations will provide appropriate input.
That brings me to a question relevant to our meeting today: how should the GCN relate to this Global Forum on Competition? In my view, the two exercises are very compatible; they should play quite different roles in the international conversation on antitrust, yet they should be mutually reinforcing.
As have explained, the GCN will focus on narrowly-defined issues that we hope to resolve (or make progress on) in a relatively short time, at least by the usual standards of multilateral exercises. In contrast, I understand that the purpose of the Global Forum on Competition is to bring together a limited number of developed and developing countries to share experiences on broad range of antitrust subjects, such as those we will discuss later today and tomorrow. These subjects are not necessarily controversial, and those that are may not be good candidates for seeking consensus in the near term. On the other hand, it may well be that the dialogue in this Global Forum on Competition will provide valuable input to GCN projects, or develop issues for consideration by the GCN. Similarly, many of the the discussions in the GCN should enrich related discussions here. These two exercises can be partners, not rivals.
The world has come to understand that antitrust enforcement has an important role to play in national and global marketplaces. We should be gratified by that. The antitrust community has come to understand that cooperation between and among antitrust agencies must play an essential role in sound law enforcement. That also is a very good thing.
But we have now recognized something else: that some of the procedural and substantive differences among us do matter, to us as agencies, to the businesses whose conduct we review, and to the consumers we serve. Many of those differences cannot reasonably be expected to disappear solely through strong enforcement cooperation. For that we need new, broadly-based mechanisms devoted exclusively to real problems of antitrust enforcement. We need a Global Competition Network, and we need this Global Forum on Competition. Thank you for your attention.
1. Revised Recommendation of the OECD Council Concerning Cooperation Between Member Countries on Anticompetitive Practices Affecting International Trade, OECD Document No. C(95)130/Final (Sept. 21, 1995).
2. Of course, the U.S. is not alone in our use of bilateral cooperation agreements. To mention two obvious examples, Canada has several antitrust cooperation agreements and an increasing network of MLATs; and the EU and its member states are working both within and outside the Community to improve their cooperative arrangements.
3. See, e.g., Northern Pacific Railway v. United States, 356 U.S. 1, 4 (1958).
4. Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447, 458 (1993).