COMPETITION POLICY: ARE THERE MORE
DIFFERENCES THAN WE CARE TO ADMIT?
William J. Kolasky
April 10, 2002
Good morning. It's delightful to be here in Brussels addressing such a large audience so early in the morning during what I understand is something of a holiday week. We've been here since Sunday and have been made to feel very welcome, not least by the glorious sunny weather you've provided. If Brussels were always this sunny, I suspect you would have even more American lawyers rushing over here.
Let me begin by introducing my two colleagues, Debbie Majoras and Ken Heyer. Debbie is our deputy for civil enforcement and, like me, spent her career in private practice before joining the Antitrust Division last year. Ken is our Director of Economics and is the senior career economist at the Division. He has been at the Division for nearly twenty years.
I should also spend a few minutes talking about why the three of us are here in Brussels. It would be impossible to overstate the importance of the economic relationship between the United States and Europe, both to our own economies and to the global economy as a whole. Together, we represent more than 50 percent of global Gross Domestic Product (GDP). Europe represents over two-thirds of total investment flows into and out of the United States; in fact, in 2000, more than 80 percent of all foreign investment into the United States came from Europe.
Because competition policy plays a central role in assuring the efficient functioning of our free market economies, we both have a stake in each of us getting our competition policies right, not just for our own prosperity but for that of the world at large.
On Monday we met with a broad cross section of your private competition bar and business community. We heard a lot of criticisms of the merger review process here in Brussels, but we also received some fair criticisms of our process as well. I suspect the balance of criticisms might well be reversed were senior Commission officials to hold a similar roundtable with our business community and bar.
We spent a great deal of time discussing areas of possible divergence between U.S. and EU law, particularly in the areas of merger review and abuse of dominance. With respect to abuse of dominance, the major areas we discussed were predatory pricing, fidelity rebates, monopoly leveraging, and refusals to deal, especially with respect to intellectual property. We selected these areas because, based on the decided cases, it appeared to us that European law was substantially more restrictive than U.S. law and because in each of these areas both of us have matters currently in litigation or under investigation. As a result, our law and policy in these areas is the subject of lively debate on both sides of the Atlantic. I must report, also, that we found considerable differences of opinion as to whether the differences in these areas are as real in practice as they would appear from the decided cases, many of which are quite old.
Our view is that the debates over these issues that are going on in both our jurisdictions should have a transatlantic dimension as well, so that the best minds on both sides of the Atlantic can work together to come up with the optimal approach to these difficult issues. This is not the say that we will ultimately agree on a single approach. There may be good reasons for us to take slightly different approaches reflecting differences in our economies, our legal systems, and our cultures. But we should make sure that if we take different approaches we understand why we are doing so and we keep as our objective promoting consumer welfare and economic progress.
We met yesterday with our colleagues at the Directorate General for Competition and again discussed a wide range of issues. As you know, we each have under way parallel reviews of the competition issues raised by the growing importance of intellectual property in our increasingly knowledge-based economy. We had a very good discussion of the current status of these reviews and we agreed in principle that we should coordinate our parallel reviews more closely than we have in the past.
In addition, as you know, the Commission is currently reviewing its merger process and is considering a number of potentially significant changes. One of the issues on which the Commission's Green Paper invites discussion is whether the Merger Regulation should be amended to change from the current "create or strengthen a dominant position" (CSD) standard to the "substantially lessen competition" (SLC) standard we apply in the United States. We spent yesterday afternoon, therefore, describing how our merger review process works and how we apply our SLC standard in practice.
One key area on which we focused both days is the role of efficiencies in merger review. As you know, in the United States we pride ourselves in taking an economically-driven, consumer welfare oriented approach to merger review and to antitrust enforcement generally. For us, as Ken Heyer stated very nicely yesterday, "efficiencies are the goal; competition is the process." We value competition, not for its own sake, but because it promotes both allocative and productive efficiency assuring that society's scarce resources are put to their best use and that companies are stimulated to work hard to lower costs and develop new and better products and services, thus raising our standard of living and promoting economic progress and growth.
For that reason, we believe that efficiencies must be an integral part of our competitive effects analysis, just as they are here in Europe under Article 81. To us it is impossible to evaluate fully the likely effect of a proposed merger on consumer welfare without balancing the likely efficiencies against any likely increase in market power. In so doing, we currently apply what some have called a "consumer welfare" standard in which we try to determine what the likely net effect of the transaction will be on price and output.
There is, by no means, unanimous agreement on our side of the Atlantic that this is the right test. In that regard, I would commend to you last Friday's decision by the Canadian Competition Tribunal in the Superior Propane case. In its decision, the Tribunal applies what it calls a "total welfare" standard to approve a merger to monopoly that it found was likely to lead to higher prices on the ground that $28 million in cost savings from the merger swamped a small $6-8 million deadweight loss caused by the likely increase in price. The Tribunal looked at the distributional effects of the likely price increase, in terms of taking money from consumers and giving it to producers, but concluded that any adverse effect on the distribution of wealth was quite small. Taking the view that producers are consumers in their time off, the Tribunal counted only negative effects on low-income consumers.
When we talk about the need to integrate efficiencies into the merger review process, one of the arguments we encounter from critics on both sides of the Atlantic is that studies show that more than 60% of all mergers fail. That being the case, they say, it must be that efficiencies claims will often be exaggerated and nearly impossible to evaluate.
In thinking about what lessons we should draw from these studies, one has to ask why these mergers failed. The failure rate in these studies is determined, not by whether the merger was good or bad for consumers, but by whether it increased shareholder value. While it may be true that a merger that fails to increase share prices may have failed to deliver on any promised efficiencies, it presumably also did not create any market power. In addition, one cannot exclude the possibility that the merger did produce efficiencies, but that any increased profits were quickly competed away by the firm's rivals matching those efficiencies.
I would certainly agree that it is very hard to evaluate the efficiency claims parties advance, but I would suggest that it is no more difficult than predicting the likely market power effects. One analytical tool that may help in undertaking this difficult task is to use a decision theoretic framework to try to predict the welfare effects of the range of possible outcomes, taking into account the probability of each outcome, in order to come up with the expected value of the merger to society.
A second objection we often encounter, at least in some quarters in Europe, is that there should be an on/off switch for efficiencies when it comes to dominant firms. While efficiencies are good in the hands of ordinary firms, these critics say, we do not want dominant firms to become more efficient because that will make them even more dominant. In its most extreme form, as was arguably the case in the GE/Honeywell merger, this view can lead to the conclusion that we should not allow mergers that enable dominant firms to become more efficient because they may use those efficiencies to drive rivals from the market, after which they will be able to raise price and harm consumers.
There are several problems with this superficially appealing story. First, under this theory, the short-term benefits of the merger are necessarily more certain that the potential long term harms. If the efficiencies are not realized, there is no benefit, but there also is no harm. If the efficiencies are realized, the harms may still not be. Rivals may find ways to respond and, even if they don't, customers may behave strategically so as to preserve competition. Finally, even if the rivals exit, the prices charged by a more efficient monopolist may be lower than the prices charged by small group of less efficient competitors.
To evaluate this rival exit story of consumer injury, we would need to know the probability of exit and would have to quantify both the likely benefits of the merger during the competitive round before the rivals exit and the likely harm following their exit, discounting both to present value. In making these calculations, the likely duration of both the competitive and monopoly rounds and the ability of customers to shift purchases from one to the other will be important factors, as will the likelihood of entry (or re-entry) should prices rise.
In the U.S., we believe these types of calculations will generally exceed our limited predictive abilities. We therefore will generally not intervene in these circumstances. We prefer to let markets sort it out. To those who think otherwise, we would ask a simple empirical question: Show me the money. Show us a merger that led to lower prices but that then resulted in driving rivals from the market and creating a monopoly.
In discussing efficiencies, we've also heard people try to distinguish between mergers that produce cost savings and those that simply give the merged firm the incentive and ability to reduce prices, either because of greater financial power or strategic behavior, but with no claim of predation. In all of these stories, pre-merger prices are necessarily above marginal cost. Our view is that anything that serves to bring price down closer to marginal cost enhances allocative efficiency and thereby benefits consumers. We don't buy the notion, therefore, that some efficiencies are better than others.
This brings me to a question we're sometimes asked as to why we have been so vocal in identifying and discussing these issues. One might even argue that if we believe what we say, our more liberal antitrust policy should give the U.S. a competitive advantage that we shouldn't be so eager to give away.
Our answer relates to our growing concern that companies, both European and American, are increasingly trying to use the antitrust agencies and courts the way they used to use regulatory agencies to impose shackles on their more efficient competitors. This is what economists call rent-seeking behavior. We are particularly concerned about the potential for forum shopping that would exist were there to be a major divergence between the U.S. and Europe. I can see American companies queuing up for the transatlantic shuttle, as they did in GE/Honeywell. That might be good for the airlines, but it won't be good for consumers.
The good news is that to date, and GE/Honeywell aside, we have generally reached consistent results on the mergers we have reviewed in common, despite having different substantive standards (or at least different verbal formulations of our standards) and different merger review processes. Indeed, in reviewing recent Commission merger decisions, we find that the market share levels at which the Commission challenges mergers are very similar to the levels at which we generally file challenges. The only consistent difference we find is that the Commission tends to define geographic markets more narrowly than we do, which is not surprising given the different degrees of geographic integration in our respective economies.
In closing, I want to repeat how important we believe it is that we continue to work hard to establish even stronger working relations between our competition agencies and that we also open up a broader transatlantic dialogue over the kinds of issues I've discussed today. We want the best minds on both sides of the Atlantic to work together to develop the best possible competition policies for both Europe and the United States. Our trip here and the warm reception we have received reflect our mutual commitment to do just that.