Prepared Remarks before the Transatlantic Business Dialogue Principals Meeting
The following is a purely personal comment on the current controversy over merger enforcement in the United States and in Europe. It is particularly important to emphasize here that I speak only for myself and not for my agency or anyone else in it.
Today, there are over 80 countries in the world with antitrust laws and about 45 that specifically provide for some kind of merger review.(1) Most of these laws have been enacted in very recent years, many in countries with no history of competition law and vastly different economic systems. The practical effect of these laws is still unknown.
It seems ironic that the looming impact of these developments has, up to now, been largely ignored by the business community and that attention has been focused instead on one or two isolated examples of differences between the United States and the European Union - areas that have so many common interests and such an extensive history of fruitful cooperation.
I will not repeat the significant examples of cooperation and convergence today because it has already been done recently by one of my former colleagues(2) and because I want to move quickly to the current controversy over the General Electric/Honeywell matter.(3) It is sufficient to say that the competition authorities in the United States and the European Union cooperate in an efficient way, that they agree far more often than they disagree, and that the areas of disagreement have progressively narrowed. In fact, there may be as much agreement between the United States and the European Union in competition matters as there is between the federal and state governments within the United States.
II. The General Electric/Honeywell Controversy
This case was reviewed on the United States side by the Antitrust Division of the Department of Justice, so I have no firsthand knowledge of the facts. All I know is what I have read in the trade press, in the public statements of various officials, and in a preliminary and redacted version of the European Commission's opinion, which was made available to me. I simply do not know how I would have voted on the General Electric/Honeywell transaction, had it been reviewed in my agency, but I will assume for purposes of this discussion that I would have voted to clear the deal. This does not mean that I would also conclude the European Commission acted unreasonably when it decided the matter the way that it did.
First, there is no need to linger on the argument that it is somehow wrong for the European Union to block a transaction involving two United States based companies. It is the locus of the effects that counts, not the nationality of the parties, and we in the United States have not hesitated to review mergers that involve offshore entities but affect United States commerce.(4) In this connection, of course, we also need to be sensitive to the fact that potential competitive effects may be different in different areas of the world. I will return to these factual differences later, in a slightly different context.
Second, we need to remember that the European Commission is bound by its governing statutes and regulations just as we in the United States are bound by ours. In the European Union, the governing regulation refers to "a dominant position,"(5) and that naturally leads the European authorities to consider the impact of single firm dominance to a greater degree than we do in this country. In contrast, our merger law refers to an effect that will "lessen competition or tend to create a monopoly." The "competition" standard has a potentially broader reach, and that has naturally led to an historic emphasis on multi-firm coordinated effects, rather than single-firm effects, in this country. It is noteworthy that we are moving together, as the European
Union begins to focus on oligopoly concerns and the United States increasingly focuses on so-called unilateral effects.(6) But, it is also not surprising that some differences remain.
Third, the history of merger regulation is quite different in the United States than it is in Europe. Meaningful merger regulation in the United States dates back to 1950, when the Clayton Act was amended to cover asset acquisitions;(7) merger regulation by the European Commission began in 1990. The 1950 expansion of the merger law in the United States was strongly influenced by Congressional concern about "aggregate concentration" of industries across the entire economy - concerns that we now know were ill founded and now believe to be economically irrelevant, anyway.
United States merger regulation in the two decades following the 1950 expansion had a strongly populist tinge. Big was bad. Vertical and even conglomerate mergers were bad because they might "foreclose" competitors, even in a relatively trivial way, and efficiency was also bad because it might confer a "competitive advantage" over competitors. The courts broadly supported agency enforcement actions after the 1950 amendment. One Supreme Court justice remarked that "[t]he sole consistency that I can find is that in litigation under Section 7, the Government always wins."(8)
Beginning about 25 years ago, there was a revolution in antitrust thought in the United States. Empirical evidence showed that the threat to competition from conglomerates had been exaggerated, and other assumptions were gradually modified or reversed. I think there is a broad consensus in the United States today that efficiency is good and that there is nothing wrong with the pursuit of "competitive advantage" - that is what competition is all about and consumers benefit in terms of price, quality and innovation. We agree that antitrust should protect "competition" not "competitors." But the battle was hard fought and fears of populist revival have never disappeared.
This history may help to explain why there is such a strong reaction in this country to any rhetoric that may signal a revival of a philosophy that we have fought hard to change. Europe does not share this history and this same rhetoric may not have the same overtones there. For example, reference to immediate adverse effects on competitors does not necessarily mean that the European Commission is more concerned about "competitors" than about "competition."
In fact, we in the United States continue to worry about "competitors," as well - not because injury to any particular company is necessarily cause for concern, but because this kind of injury may portend something ominous for the future of competition. (In this country, we might be careful to cite injury to a particular competitor as an illustration of the broader effect that we fear.) Let me express this in a different way. Likely injury to competitors does not prove likely injury to the competitive process, but it does not disprove it either. Antitrust authorities in this country no longer assume - if they ever really did - that competitor complaints necessarily indicate a merger will be efficient and pro-competitive. This may be true in a static view of the world, but we are also concerned about long-term consequences. Competitor complaints (appropriately discounted for self interest) may convey some helpful insights about dynamic competition.
I am not suggesting that the differences over General Electric/Honeywell are purely rhetorical; I am merely suggesting that we in the United States should not overreact when people with a different history express themselves in a different way.(9) The important thing is the bottomline assessment of the facts.
Fourth, it is simply wrong to suggest that United States merger regulation is informed by empirical economic knowledge and European Union merger regulation is not. It is true that we no longer apply the so-called "portfolio" theory of single-firm dominance when reviewing mergers in this country because research here indicates that the possible harm to competition is likely to be much more speculative than the benefits to consumers, absent horizontal overlaps or vertical relationships.
I am not so sure, however, that we can identify markets well enough to differentiate between purely conglomerate mergers and those with problematic horizontal or vertical effects.(10) Economic learning derived from empirical research in some industries is not necessarily transferable to other industries. We also are beginning to have a better appreciation of strategic behavior, which may enable firms that dominate adjacent market niches, or sell complimentary products, to engage in subtle strategies - short of outright foreclosure - that could ultimately weaken competition. I will acknowledge that I personally worried a lot about these possibilities in the recent AOL/Time Warner matter,(11) even though I ultimately concluded that my concerns were unwarranted, in light of the far-reaching settlement.
Moreover, all of us in the United States antitrust community need to acknowledge that there are lingering elements in our own Merger Guidelines that are not grounded in economic science. Market concentration has fortunately been progressively de-emphasized as we move from the 1968 Guidelines through those of 1982, 1984 and 1992 (the last word on the subject). The 1992 Guidelines, however, still embody a "presumption" that transactions with a postmerger HHI greater than 1800, and a delta greater than 100 points,(12) are problematic. This suggests, for example, that it makes a difference whether a market has six significant competitors or five, and, to my knowledge there is no economic learning whatever to support this presumption. Our European friends, thus far, have not gone down this road.
We also need to recognize that all merger regulation is inherently uncertain. With all our talk of markets and concentration and offsetting efficiencies and the like, we are ultimately attempting to weigh somewhat uncertain present effects against even more uncertain future effects. We simply do not have an economically rigorous way to do this. What we do is apply certain presumptions or rules of thumb, in order to make the law predictable and less subjective.
We do the same thing in other areas of antitrust. The law of predation - cases like Microsoft - raise similar issues and we have presumptions and rules of thumb in that area, as well. I have aggressively supported up-front tests, like the so-called "Areeda-Turner" cost-based standard for price predation,(13) in order to make predation principles workable, but would never claim that these standards necessarily draw an accurate balance between present and future effects. There always will be some elements of unpredictability and subjectivity, and reasonable people can differ.
I do not intend to denigrate economic research and analysis in any way, with these comments. Economists have made immense contributions to the sensible evolution of antitrust law generally, and merger law in particular. I personally rely heavily on their advice. But, unfortunately, the difficult cases cannot be resolved in a purely objective way. A computer is no substitute for the individual judgement of a CEO, and those of us with responsibility for antitrust enforcement have to exercise individual judgement as well.
Finally, those who think it is improper for antitrust regulators to balance necessarily speculative long-term harm against more credible present benefits might reflect on the fact that we routinely do the reverse when we protect certain kinds of intellectual property. For example, we are willing to grant patent monopolies for 20 years, with relatively certain short-term competitive effects, based on the speculation that this monopoly grant will encourage invention and benefit consumers down the road. I suspect everyone in this audience, including myself, supports a patent regime, but the particular rules that have evolved are not based on empirical economic research either.
My purpose here, obviously, is to put the recent controversy over General Electric/Honeywell in a broader perspective. I personally believe that antitrust authorities in the United States and in the European Union are a lot closer together than the public would believe and I also believe that, with mutual tolerance and goodwill, we will draw closer still.
1. The United Nations Conference on Trade and Development (UNCTAD) lists 82 countries that have competition authorities. See: <www.unctad.org/en/docs/c2clp99d16.en.pdf> The American Bar Association has identified and described 46 international merger notification requirements. See Antitrust, Vol. 15, No. 2, Spring 2001, at p. 16.
2. See Prepared Remarks of Chairman Robert Pitofsky, "EU and U.S. Approaches to International Mergers - Views from the U.S. Federal Trade Commission," before European Commission Merger Control 10th Anniversary Conference, The European Commission Directorate General for Competition, International Bar Association, Metropole Hotel, Brussels, Belgium, 14-15 September 2000, available at: <www.europa.eu.int/comm/competition/mergers/cases/decisions/m2220_en.pdf>.
3. General Electric/Honeywell, Case No. COMP/M.2220, Commission decision of 3 July 2001 (not yet published in Official Journal), available at: <www.europa.eu.int/comm/competition/mergers/cases/decisions/m2220_en.pdf>
4. For example, in Ciba-Geigy/Sandoz, 123 FTC 842 (1997), the FTC took action against a merger of two Swiss pharmaceutical firms that would have affected United States consumers in several relevant product markets. Other mergers of foreign firms in the pharmaceutical industry that raised competitive concerns in the United States and resulted in FTC enforcement actions include Glaxo/Wellcome, Astra/Zeneca, and Glaxo Wellcome/SmithKline Beecham. Examples in other industries include Mahle/Metal Leve, 123 FTC 1431 (1997); Montedison S.p.A., 119 FTC 676 (1995); and Guinness/GrandMetropolitan, 125 FTC 735 (1998). An example of a merger of foreign firms in which the FTC took enforcement action where the parties did not produce the relevant product in the United States was Oerlikon-Buhrle/Leybold, 119 FTC 117 (1995), an acquisition of a German firm by a Swiss firm.
5. Article 2 of the European Union's Merger Control Regulation (Council Regulation (EEC) No. 4064/89 of 21 December 1989 on the control of concentrations between undertakings, OJ L 395, 30 Dec. 1989, corrected in OJ L 257, 21 Sept. 1990, as amended by Council Regulation (EC) No. 1310/97 of 30 June 1997, OJ L 180, 9 July 1997, corrected in OJ L 40, 13 Feb. 1998) provides the legal standard against which the European Commission must appraise mergers: "[a] concentration which creates or strengthens a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it shall be declared incompatible with the common market."
6. See Prepared Remarks of Chairman Robert Pitofsky, supra n. 2.
7. The Cellar-Kefauver Act of Dec. 29, 1950, Pub. L. No. 81-899, 64 Stat. 1225 (codified as amended at 15 U.S.C. § 18 (1994)).
8. United States v. Von's Grocery Co., 384 U.S. 270, 301 (1966)(Stewart, J., dissenting).
9. The fact that English words may also have different overtones to people with different first-languages is another complicating factor that we always need to be aware of.
10. See Thomas B. Leary, "The Significance of Variety in Antitrust Analysis," 68 Antitrust L.J. 1007 (Vol. 68, Issue 3, 2001).
12. U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, 4 Trade Reg. Rep. (CCH) ¶ 13,104 (1992), §1.51.
13. See Prepared Remarks of Commissioner Thomas B. Leary, "The Need for Objective and Predictable Standards in the Law of Predation," before the Steptoe & Johnson and Analysis Group/Economics 2001 Antitrust Conference, Washington, D.C., May 10, 2001, available at: <www.ftc.gov/speeches>.