AGENDA ON COMPETITION POLICY
JOEL I. KLEIN
The Royal Institute of International Affairs
November 18, 1996
Next month the World Trade Organization will hold its Ministerial Conference in Singapore. The European Union, among others, has proposed the formation of a working group to begin development of a trade and competition agenda at the WTO. These proposals generally start from the view that, first, there are significant market access problems whose solutions lie in the application of competition law, and second, the negotiation in the WTO of multilateral rules on the application of competition law would help to alleviate these problems.
The United States Government has not ruled out the possibility of a role for the WTO in this area, but has urged a cautious approach. The Antitrust Division of the Department of Justice has endorsed this position and, in my comments today, I would like to make the case for caution. To do so, however, it is first necessary to explain what is already going on in international antitrust enforcement, how those efforts relate to traditional trade-liberalization concerns, and what would be lost if a misguided WTO agenda ended up derailing, or even detracting from, these ongoing efforts.
The events that have sparked an intense interest at the intersection of trade and competition policy are well known and need not be belabored here. Two factors have converged to bring this matter to the fore: first, an increasingly globalized economy, spurred largely by technological advances, has meant that markets throughout the world are economically available even to previously domestic businesses; and second, the successive reductions of government-imposed barriers to trade (resulting from the various GATT rounds) has meant that entry into foreign markets is not just economically feasible but practically feasible as well. Taken together, these developments have led to something of an explosion in worldwide trade -- in 1995, there was $4.9 trillion in merchandise and $1.2 trillion in services traded beyond national borders; and in the U.S., for example, currently a quarter of our GDP is comprised of export and import trade, which is double the figure for 1945.
This process of globalization, which will surely accelerate in the years ahead, plainly has important consequences in terms of conventional trade concerns in that the elimination of governmental restraints has brought into focus private business practices that inhibit market access. But globalization has even more significance in terms of international competition issues. For discussion purposes, these issues can fairly be divided into three categories:
An international antitrust agenda should focus on all three areas, whereas a sensible trade program need only focus on the third.
With that recognition in mind, let me next discuss 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. First, the easiest of the three: transnational merger review. In my experience -- based on several specific instances where our agency has worked with our counterparts in other countries, such as the recent Scott Paper/Kimberly Clark merger -- this area presents fewer and less urgent problems than the other two. As a practical matter, transnational mergers are not being inhibited by the necessity for multiple reviews and, at least on an informal level, the various enforcement agencies are already engaged in some cooperation that is likely to increase over time and also likely to lead to more formalized merger agreements in the years to come -- initially on a bilateral, and then, perhaps, on a multilateral, basis. Moreover, merging firms that are subject to multiple reviews can facilitate coordination and cooperation among the various national competition agencies by authorizing them to share otherwise confidential information. These considerations notwithstanding, there is still room for further improvement in this area and I believe it would be profitable to study additional ways to make the multi-agency review process more efficient and less burdensome. BIAC recently submitted some proposals on how to do this to the Competition Law and Policy Committee of the OECD and those proposals are currently being analyzed.
Next, I would like to turn to the two other areas on the agenda of international antitrust enforcement where I believe that new arrangements between governments have the potential to advance things significantly: cartel enforcement and market access cases. Both have important, though by no means identical, effects on international trade. International cartels typically involve arrangements among manufacturers or producers of goods that sell in international markets. And just as happens within domestic markets, the sellers of such goods sometimes conclude that collusion is preferable to competition and they decide to agree either on prices, on volume, or on what markets each will sell in. In the U.S. we recently prosecuted two high-visibility examples of this practice -- one in the $600 million market for lysine (a farm feed-additive product) and the other in the $1.2 billion market for citric acid. The defendants have agreed to pay approximately $100 million in fines in our lysine investigation and the citric matter is still in its early phases but we have already obtained a guilty plea and a $30 million fine. One prominent U.S. corporation, Archer Daniels Midland, 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 ever previously obtained. As high as this fine is -- and the court did impose the fine that we requested -- we would have sought a far larger fine if ADM had not agreed to cooperate fully in our ongoing international investigation in citric acid.
The point that I want to emphasize here is that these kinds of cartel cases can have an impact on trade by taking enormous amounts of money out of the pockets of consumers around the world, but 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. But consumers are hurt. For example, at the U.S. sentencing hearing in the lysine case I just mentioned, Ajinomoto, a Japanese participant in the conspiracy, asserted that almost three-quarters of its sales took place in Japan, where there is every reason to believe that Japanese consumers also bore the brunt of this illegal cartel. For that matter, it is reasonable to assume that consumers throughout the world who purchased these products all suffered by having to pay artificially inflated prices resulting from the cartels' practices.
The third kind of case -- market access cases involving private business restraints -- presents 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. To begin with, this type of case involves a situation where consumers in a domestic market are 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 point of view, this is no different from a situation where a new domestic entrant is prevented from being an effective competitor by dint of private restraints. In either case, domestic consumers are harmed because competition has been diminished. From the trade perspective, however, there is a separate injury when foreign competition is kept out, namely, 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 -- one such effort being the proposals to introduce competition policy at the WTO. In an increasingly globalized economy, stimulated in part by the consistent reduction of government-imposed restraints on market access, it is hardly surprising that trade officials would have bumped up hard against private market restraints that impede foreign access. Some of these restraints may have existed even when government-imposed trade restraints were more numerous and others may have arisen in response to the ongoing removal of government restraints. In searching for a solution to this as well as to the cartel problem, the U.S. Department of Justice has focused not on the WTO but largely on bilateral agreements and working relationships instead. There are reasons for our current approach -- both in terms of how we deal with each aspect of the problem and how those aspects relate to each other -- that I would now like to discuss.
Let us look first at the issue of international cartels. The problem for antitrust enforcers is that the procedures 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 a very fact-intensive exercise that almost invariably places a high evidentiary burden on enforcers. And when competition authorities cannot get access to the evidence needed to prosecute a violation, the world's consumers and businesses ultimately bear the cost.
Unfortunately, we have directly run up against this problem. For example, a couple of years back in the GE/DeBeers case, we 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, and much of the evidence was located overseas and consequently beyond the Justice Department's reach, although we did seek and received 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., beyond the reach of U.S. prosecutors.
In addition to the problem of not being able to obtain the necessary evidence located abroad, our efforts to cooperate effectively with other antitrust authorities can be stymied by the absence of arrangements that allow the sharing of our own evidence with those authorities. In criminal investigations, for example, our rules of criminal procedure are very strict in protecting the secrecy of grand jury proceedings. Fortunately, we now have powerful new legislation, which I will describe in a moment, 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 that I described earlier, 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, but our ability under current law to share such information with countries who are not parties to cooperation agreements is very limited.
So, how can we best promote the international cooperation and information sharing necessary for these high priority investigations? The Clinton Administration and the U.S. Congress, both recognizing the critical importance of such cooperation, recently gave us explicit authority to negotiate bilateral antitrust cooperation agreements in the International Antitrust Enforcement Assistance Act of 1994 (the "IAEAA"). 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 IAEAA agreements with our trading partners is a top priority for the Department of Justice.
In further pursuit of this agenda, the United States also recently proposed an initiative in the OECD'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. There are several reasons why I think this proposal will be adopted. To start with, identifying the type of egregious anticompetitive conduct that constitutes a hard-core cartel is a relatively straightforward matter. There is little or no economic debate in most instances about the competitive effects of this conduct. Second, 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. We are talking about evidence of flagrant wrongdoing needed for law enforcement purposes: 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 prospective business plans of the kind that may be needed in connection with mergers and other economically complex inquiries.
We recognize nevertheless that getting nations 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. And there are always important cultural and sovereignty issues that must be resolved when such agreements are contemplated. Still, it is my personal belief that such differences ultimately will not stand in the way of cooperation aimed at eliminating cartels.
In support of my optimism, aside from my general nature, I would point to encouraging evidence from other models of law enforcement cooperation. 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 (MLATs), which provide for comprehensive reciprocal assistance between the United States and foreign governments in criminal matters. 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 almost 20 such agreements (including an important one with the Serious Fraud Office), 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 MLAT 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. I think these matters are extremely important, and as I assume they are familiar to many of you, I will only briefly describe them here. In our Fax Paper cases, the Department has charged six Japanese firms, 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 have agreed to plead guilty and pay fines totaling nearly $10.5 million; contested proceedings against the four other defendants are pending. The Canadians have charged, and obtained guilty pleas from, some of the same firms, and I understand that their investigation is continuing.
Several points are worth noting about the fax paper investigation. The Canadians first brought the case to the attention of U.S. authorities, who 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 MLAT, sharing documents, and jointly interviewing witnesses. Cooperative analysis of documentation, and the sharing of a data base created by the Canadians, were both possible without violating any domestic confidentiality rules.
I would like to quickly mention two other examples of cooperative antitrust enforcement under our MLAT with Canada. In the Plastic Dinnerware case, the Federal Bureau of Investigation and the Royal Canadian Mounted Police simultaneously executed search warrants on both sides of the border, which ultimately led to Department of Justice 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. As it turned out, the seized documents revealed that the conspiracy did not affect the Canadian market, so no Canadian investigation ensued. Lastly, the Department and the Canadians conducted parallel investigations into anticompetitive behavior in the ductile pipe industry. Although we concluded that we did not have sufficient evidence to prosecute under U.S. law, the Canadian authorities assembled a different body of evidence of violations of Canadian law that led to a guilty plea and then-record criminal fine in Canada, from a Canadian subsidiary of a U.S. firm.
In short, we must extend the type of cooperation that has been so successful under our agreement with Canada to our other trading partners. The first step of our international antitrust agenda, the promotion of meaningful international cooperation in the prosecution of global cartels, is one which all responsible antitrust authorities should endorse, and I hope we can move quickly on this front.
This brings me to our second principal international antitrust initiative -- dealing more effectively with market access problems that result from foreign anticompetitive conduct. In the Department's experience, 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 protectionistic influences. This is absolutely the situation in the U.S. and I can point to specific cases where we opened our markets to foreign competition by challenging private restraints that protected domestic competitors. From the perspective of competition policy, this makes sense since the more competition there is, the 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. But with respect to those countries that do have such a tradition, we believe that bilateral, or what are generally referred to as "positive comity," agreements are the best way to ensure effective enforcement. Under such agreements, 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 and that authority conducts the investigation and then reports back to, and consults with, the initial country as to the nature of its investigation, its findings, and any remedy it is considering. The referring country can accept these conclusions, or seek to modify them, or it can subsequently conduct its own investigation and take actions that it thinks are appropriate.
This positive comity approach is not a new concept, having its roots in a 1967 Council Recommendation of the OECD on Co-operation between Member Countries on Restrictive Business Practices Affecting International Trade. The term "positive comity" itself was coined and given emphasis in the 1991 cooperation agreement between the U.S. and the European Union, and positive comity is also a prominent feature of the 1995 agreement between the U.S. and Canada. We are currently working to refine the positive comity procedures in our agreement with the EU, so as to specify more clearly the situations that would presumptively call for referrals and to flesh out the report-back and consultation mechanisms that would come into play once a referral has been made.
This approach has several things to recommend it. First, as I have mentioned, competition authorities tend to have the greatest stake in taking such complaints seriously -- even if they do involve foreign access. Second, such a process makes it most likely that the kind of evidence necessary to properly decide these kinds of cases can be obtained. The limits of jurisdictional reach for all countries means that, in the absence of a treaty, the ability to gather evidence on another's territory is extremely limited. This thwarts effective investigations and, I would candidly acknowledge, 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. And 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 tensions by providing a sensible, systematic approach to fact-gathering, reporting, and bilateral consultation among competition authorities.
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 recognizes that the country whose market is most immediately affected has the principal responsibility for enforcement. I would note in this regard that the first recommendation of the business sector's Transatlantic Business Dialog issues group on competition policy was to "pursue strengthened transatlantic positive comity procedures under the bilateral agreement," in consultation with the business community and other affected interests.
Still, there are problems in this approach that should be frankly acknowledged. Many competition authorities currently lack the independence, if not the will, to do a proper job. And until that situation changes, positive comity referrals are obviously not a satisfactory answer to the problem of market access that is blocked by private anticompetitive restraints. Moreover, different countries have different substantive law in the competition area, so there is at least some risk of disparate enforcement. While true, this problem is likely to be reduced, though not eliminated, through the kind of cooperative relationship that I have been discussing. And, I might add, the differences -- at least when they are discussed by people in the competition enforcement field -- are not so drastic as some might expect. The basic organizing concern for consumer welfare adds a measure of consistency in the competition area that is likely to grow over time.
Having separately outlined our views on international cartel enforcement as well as on positive comity, I would like to touch briefly upon the connection between the two areas. It is apparent to me that the ability among 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, an area where trust among countries -- even among competition authorities -- is not high. Consequently, unless countries can be assured that information is not being used for trade-related purposes -- but only for international cartel enforcement, as to which all countries tend to share an acknowledged common enforcement interest -- overall cooperation will be diminished and cartel enforcement, as well as market access enforcement, will suffer. This would be a serious loss.
Now that I have described our existing approach to international antitrust enforcement, let me turn specifically to the question of a trade and competition agenda for the WTO. As an initial matter, I think that anything we do must not jeopardize or even detract from, the ongoing efforts that I have just described. By and large these efforts involve bilateral undertakings between and among countries that have well-established commitments to, and experience in, competition matters. They are the kind of efforts that are most likely to bear fruit and lead to agreements that are enduring and that can subsequently become a template for other countries.
Moreover, in part because of this concern, but even going beyond it, a hasty effort at the WTO is fraught with risk. First, it will be hard to reach agreement on sound competition rules, which depend so much on the strict application of neutral legal and economic principles, in the WTO. Even among the relatively like-minded member states of the European Union, it took seventeen years to agree on the Merger Control Regulation. A WTO competition policy debate would have to balance many (often diverse) national interests, with the possibility of positions shifting in response to trade-offs in other trade negotiations related to agriculture, services, intellectual property, or any of the myriad fields currently covered by WTO agreements.
Second, and related to the first concern, we must guard against a lowest-common-denominator outcome in the development of competition rules by the WTO. That is, efforts to achieve a "minimum" set of competition principles or to identify common substantive standards could end up legitimating weak and ineffective rules, which certainly would not serve the goals of trade liberalization. As we all know, minimum standards often become the maximum.
Third, although a universal commitment to the adoption and enforcement of competition laws, and cooperation in antitrust enforcement, are worthy goals, I believe that they go beyond core WTO concerns. This is evident when we consider the problem of how the WTO would identify much less devise remedies for, violations of any multilaterally agreed upon competition obligations. By my count, some 65 countries, accounting for 80 percent of the world's GNP, already have competition laws on the books, and I would speculate that most of these laws would likely meet the requirements of any minimum substantive rules the WTO could adopt. Competition policy, moreover, is often very fact-intensive, and to my knowledge no government has proposed turning over to a WTO body the kinds of confidential business information typically required for a proper competition analysis in particular cases. Nor do I expect that such a process is likely to be acceptable on a worldwide basis for many years to come.
This problem of dispute settlement highlights the difference between competition law and other areas covered by the WTO. On the one hand, in the absence of broadly shared views on the precise objectives and supporting analysis applicable under competition laws, the use of dispute resolution with respect to a general requirement that member states adopt and enforce antitrust laws, and also consider requests to investigate from other states, is likely to have little impact on trade liberalization, and could in fact give procedural legitimacy to harmful actions masquerading as competition policy. On the other hand, if dispute settlement were extended to individual decisions taken by domestic competition authorities, this could interfere with national sovereignty concerning prosecutorial discretion and judicial decision-making, and could also involve WTO panels in inappropriate reviews of case-specific, highly confidential business information.
For these and other reasons, the U.S. government remains, as I said at the outset, "cautious" about a WTO agenda on competition policy. For the WTO to study what is going on elsewhere and to analyze the significance of those developments might well make sense. But only if such work is not seen as a precursor to negotiations in the WTO on competition policy.
To conclude, I want again to emphasize that there currently is a busy and important agenda in international antitrust enforcement that is relevant in part to trade issues but whose concerns are substantially broader than such issues. That agenda 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 we wait to see what, if anything, emerges in the way of a WTO work program. And any such program should be based on sound, realistic judgments about what is reasonably feasible and should not, in any respect, jeopardize the other work that is currently ongoing.
A Note Of Caution With Respect To A WTO Agenda On Competition Policy
Monday, November 18, 1996
Updated June 25, 2015