In May 1996, the FTC released a staff report ambitiously titled "Anticipating the 21st Century: Competition Policy in the New High-Tech Global Marketplace." This conference, six months after, is designed to address the twin questions of whether the analysis and recommendations of the report made sense and whether the report is likely to make a difference.(2) My view is that the report is a constructive contribution to debate over United States' competition policy on the eve of the 21st Century. Among its most important elements:
- Changing nature of competition. I sometimes think the most important contribution of the report will turn out to be Chapter 1. It accumulates in one place and documents that globalization of competition and the increase in intensity of high-tech competition are changing in important respects the ways in which the world does its business.
- Efficiencies. The report accelerated and focused debate on the treatment of efficiencies in merger enforcement and more generally in antitrust policy. Since that portion of the report has attracted most attention and is likely to have the most significant immediate impact, I will devote most of this paper to the issues it raised.
- Other issues. The report touched upon a wide range of other antitrust issues, mainly to conclude either that the enforcement status quo is adequate (failing firm defense, definition of relevant geographic market), or to serve as a preview of coming attractions, signaling areas that deserve or require future attention (definition of "innovation markets," the intersection of antitrust and intellectual property, competitive effects of networks and standards).
One important conclusion of the report is between the lines. There was virtually unanimous agreement that the traditional core of antitrust - hostility to cartels, need to control abuses of monopoly power - has served the country well and requires no drastic change. Rather, the focus of the report was on aspects of antitrust that may require adjustment in light of changing competitive conditions.
1. The State of Play: Competition at the Close of the Century.
There was a consensus among the many witnesses at the hearings that the most important changes occurring in the commercial world involve the rapid internationalization of trade, and the increased importance of rivalry at the innovation stage of competition.
The progress of globalization is easier to measure. Exports from the United States grew from 5.5% of the gross domestic product in 1970 to 12% in 1994; imports grew from 7% in l970 to l4% in l994. In non-adjusted dollar volume, these categories increased ten-fold in the last 24 years. As a result, cross-border trade accounted for 25% of commercial activity in the United States by 1994. In the rest of the world, cross border trade increased 80% from 1980 to 1993, far exceeding the increase in gross domestic product in most industrialized countries.
Some of the reasons are not difficult to identify. As a result of successive GATT rounds, NAFTA and other regional treaties, worldwide tariffs fell from an average of 40% 20 years ago to 6% today, and will decline even further when Uruguay Round GATT provisions are fully implemented. Among other factors, the cost of rail transportation fell 30% from 1981 to 1991, the cost of truck transportation fell 23% from 1980 to 1994 and airline transportation has fallen an average of 3% per year for 30 years. Aside from direct exports and imports, foreign direct investment worldwide as a percentage of gross domestic product, has increased four fold in the last 20 years.
As a result of these changes, we are beginning to see fundamental adjustments in corporate organization. More advanced companies have begun to restructure in a way that ignores borders and even regions, and focuses worldwide on specific product lines. Other firms find that the challenges of global trade require cooperation and collaboration, with increasing resort to joint ventures that appear to be more flexible, less durable, and more focused upon specific projects. Manufacturing and tracking of conventional goods is becoming more flexible, allowing for greater variety of product produced at lower cost to satisfy worldwide demands.
Measuring the impact of technological rivalry is more difficult. Its importance seems most pronounced on a sectoral basis - for example in aerospace, biotechnology, pharmaceuticals, computers and semiconductors. In those segments, there appears to be increased rivalry in research and development (to the point that the competitive game is often won or lost at the R&D level of a product life cycle), shorter time span for products in the market, and a sharp surge in technology transfers. In many product markets, competition appears to turn essentially on the level of investment, efficiency and ultimate success in the innovation phase of competition.
2. Analysis of Efficiencies Claims.
The question of how to treat efficiencies claims in defense of antitrust transactions has troubled antitrust enforcement for many years, particularly where merger analysis is concerned. As you know, the Supreme Court increasingly has paid attention to issues of efficiency in analyzing alleged anticompetitive effects of horizontal behavior(3) and in adopting a more lenient approach to non price vertical restraints,(4) but at least according to still valid Supreme Court merger cases, efficiencies are irrelevant when asserted as a defense to a claim of illegality.(5) Even in those areas of law where efficiency claims are relevant, treatment has been relatively superficial, with virtually no discussion of which asserted efficiencies are entitled to substantial weight, and how they ought to be assessed and taken into account. At least since publication of the 1982 Merger Guidelines,(6) the government has indicated a willingness to consider specific efficiencies as a mitigating factor in defense of mergers, but has expressed some skepticism about claims of efficiencies and therefore has insisted that they be demonstrated by "clear and convincing evidence."(7) The government today still may argue that efficiencies are irrelevant to merger analysis when asserted in court.(8)
The reason for hostility to efficiency claims is that efficiencies have been thought to be exceptionally difficult to measure, and it has also been thought difficult to trade off efficiencies against the disadvantages of reduced competition. Those concerns have not prevented the Supreme Court from paying more attention to efficiencies in other areas of antitrust, but the general position appears to continue to prevail with respect to mergers.
The American position of hostility to efficiency claims in merger analysis is odd since there is nothing in the legislative history of the Sherman or Clayton Acts to justify exclusion or even special skepticism toward efficiency claims, and because antitrust is supposed to encourage efficiency in order to serve consumer welfare. The American position has become odder still as American firms find themselves increasingly locked in commercial combat with companies around the world, often located in countries where barriers to mergers are extremely low or non existent, or where efficiency claims are generously viewed.(9)
A. Findings of the FTC Staff Report
The essential findings of the FTC Staff Report represent departures from the conventional wisdom in several respects:
(i) Virtually all witnesses - scholars, present and former enforcement officials, defense lawyers - asserted that efficiency claims were not more difficult to measure than many other issues in antitrust enforcement, including minimum efficient scale and barriers to entry. It was recognized that evidence of efficiencies, unlike evidence of other relevant factors, is almost entirely in the hands of the merging parties but, according to the witnesses and the staff report, that was no reason to exclude the issue entirely.
(ii) In some market settings efficiencies are likely to have a positive effect on competition by deterring the exercise of market power. For example, if firms ranking four and five in a market lower unit cost substantially by merging, that could easily and promptly disrupt pre-existing cartel-like behavior. Assuming beneficial effects on competition, the report urged that such efficiencies be taken into account not only as a matter of prosecutorial discretion but be admissible in court.
(iii) Even if efficiencies are not likely to have a procompetitive effect, they may be sufficiently substantial to justify the exercise of prosecutorial discretion. That could arise where the anticompetitive effects are marginal and there is at least some reason to think that the efficiencies over the long run would have a pro-competitive ripple effect on other players in the industry.
(iv) Clarification of law with respect to efficiencies and a more generous acceptance in merger analysis are all the more important in an increasingly global market. Finding the most efficient ways to do research and development, manufacturing, distribution and sales is crucial to maintaining and increasing the competitiveness of U.S. firms in that marketplace, and competition must take into account firms'needs to pursue business conduct that they believe will foster efficiency.
B. Effect on the Antitrust Equation: Facts vs. Assumptions
I agree with the various conclusions of the staff report, summarized above, but would go further with respect to the implications of treatment of efficiencies in antitrust analysis. In the l970s and l980s, antitrust enforcement was eased across the board, and particularly with respect to mergers, on a theory that most transactions are efficient (or else why would they have happened) and that therefore it could be assumed that easier antitrust standards will allow many efficient transactions to be completed. One result was that the merger guidelines, especially in the late 1980s, were not enforced as written.
Efficiencies are important but I believe they ought to be a matter of evidence rather than assumption. Also, most transactions, including mergers, are neutral from a competitive point of view. If the parties to a transaction are given a full opportunity to come forward with evidence of efficiencies and they cannot do so, then it seems to me the enforcement agencies can be more comfortable challenging transactions that are otherwise a threat to the competitive process - comfortable in the sense that the challenge is unlikely to do harm. That is not an indirect call for a return to the era of Brown Shoe(10) and Von's,(11) where extremely small horizontal and vertical mergers in markets with low barriers to entry were blocked, but rather a suggestion in the merger area that the guidelines can be enforced as written more confidently after the parties have a full opportunity to demonstrate efficiency claims. The implicit assumption of all this, shared by most antitrust enforcers today, is that the l982 Guidelines and their successors strike a reasonable balance between protecting markets against undue concentration on the one hand and not unnecessarily burdening transactions that are procompetitive or neutral on the other.
C. Issues that Remain.
A determination to take efficiency claims more fully into account hardly solves all problems. Partly because American law has so infrequently addressed in detail questions of relative weight and priority in the efficiency area, many issues remain unsettled.
A DOJ-FTC Joint Task Force has been organized since the FTC staff report was published, and has been meeting more or less on a weekly basis to address the details of incorporating efficiencies as a mitigating factor in merger enforcement. I believe they are making excellent progress but the fact is that exceptionally difficult issues remain and the correct answer to many of them is far from clear. I would like to offer some initial thoughts about some of the issues that will require consideration and analysis:
(i) Competitive Effect.
As noted earlier, there are some market settings in which efficiencies directly diminish the likelihood of the exercise of market power; the FTC staff report urged that these be taken into account, and even be allowed to be asserted in court. When efficiencies have a direct procompetitive effect (for example, leading to a likely reduction in price), the issue of whether efficiencies can only be taken into account in those markets where they are likely to be passed on to consumers is avoided. By definition "competitive effect" efficiencies lead to consumer benefits.
The concept of efficiencies, that are directly procompetitive because of the setting in which they occur, developed by the FTC staff in response to the hearing record, is a useful one. Nevertheless, I believe it will be rare that one can reliably predict these procompetitive effects and therefore, if efficiencies are more fully to play a role in merger analysis, there needs to be a reconsideration, clarification and ( in my view) an expansion of circumstances in which they affect a merger enforcement decision as a matter of prosecutorial discretion.
(ii) Which efficiencies and how much do they matter?
I believe the most difficult challenge facing the Joint Task Force will be to examine different kinds of efficiencies - e.g. scale, scope, distribution, R&D, capital savings and managerial - and indicate at least in a broad sense which are most likely to influence an agency decision. The standards that determine priority and weight could be the following: which efficiencies are most likely to have a procompetitive effect, which are subject to reliable proof, and which are merger specific in the sense that they could not almost as easily be achieved through a less restrictive alternative like a temporary joint venture or a contract arrangement.
In my view, the Areeda-Turner Treatise (as is so often the case in antitrust analysis) offers the most thoughtful guideline to evaluating different types of efficiency(12). It establishes a range from efficiencies of size and scope, which are most significant and susceptible of proof, to capital cost and managerial savings which should be far less influential because they can so frequently be obtained by means short of merger such as unilateral action (hiring a capable manager), contract, or joint venture. I would not reject at the outset consideration of any form of efficiency, but I believe the guidelines would be of more use, assuming the Task Force proposes retreatment of efficiency claims, if they incorporate some indication of how efficiencies will be evaluated by the enforcement agencies.
(iii) A Sliding Scale.
Virtually all witnesses and most scholarship addressing the question of efficiencies argue that the greater the likely increase in market power, the greater the magnitude of efficiencies that will be needed to overcome that likely anticompetitive effect. That leaves open the question of whether mergers to monopoly or near monopoly can ever be justified by claimed efficiencies. I assume that efficiency claims in those situations would be very rarely successful. I do not go so far as to say "never" because of some concern with the distressed industry situation. There may be markets where there is great overcapacity and only sufficient demand to support a single efficient enterprise. Some would argue that hospital mergers occasionally satisfy that standard. Without addressing the merits of such claims, I would only say that parties asserting efficiencies as a defense to merger to monopoly are likely to face an uphill battle.
(iv) Nature of Proof.
I agree with the view of so many witnesses in the hearings that most efficiencies (including particularly efficiencies of scale and scope) are no more difficult to examine as a matter of evidence than other difficult issues in antitrust analysis, and therefore should be treated as just another issue of fact.
(v) Will Efficiencies be Passed on to Consumers?
I have argued elsewhere that a requirement that claimed efficiencies be taken into account only if it can be demonstrated that they occur in a market setting that insures that the savings will be passed along to consumers is unnecessary.(13) I still believe that it is very difficult at the time of merger to demonstrate, either short term or long term, that efficiencies will be passed along. On the other hand, I appreciate the force of the argument that if it is fairly clear that the only consequence of merger generated efficiencies is that they will put money in the pockets of corporate managers and shareholders, the efficiency defense is less appealing. It could be argued that the transfer of funds to corporate managers and owners has no rightful role at all in antitrust analysis.
This difficult issue is one the Task Force will need to resolve. My own view is that a "pass on to consumers" requirement will constrain the effectiveness of efficiency claims. Nevertheless the more important goal is to clarify circumstances in which efficiency claims will be taken into account and which efficiencies matter most, and a "passing on" qualification could be viewed as a small price to pay for an important step in the right direction.
3. Other Issues addressed in the Report.
In devoting so much of this paper to the efficiencies issue, I do not mean to suggest that other subjects examined in the hearings and discussed in the report are insignificant. On the contrary, many will need to be addressed if antitrust is to adjust to new commercial conditions.
Several important sections of the staff report addressed areas of antitrust analysis that have been the target of criticism, but the report came to the conclusion that the status quo is more than adequate. Specifically the weight of testimony and the hearings report concluded that the demanding standards for satisfying the failing firm proviso in the United States should be retained, and no special provision is needed for distressed industries as long as an efficiencies defense allows firms in distressed industries to merge with somewhat greater freedom. I still believe that by requiring firms virtually to be in bankruptcy rather than clearly on the road to highly probably failure, United States law (unlike the law of many of our trading partners) is too restrictive. In my view, that is a relatively minor issue for another day.
The weight of testimony in the hearings report also concluded that definition of relevant geographic market, including situations where there is a claim that relevant market extends beyond domestic borders, is already sufficiently flexible, both in concept and implementation. My brief experience in a year and half back at the FTC is that the staff is extremely sensitive to import and export competition, and more than willing to give cross border rivalry significant weight where it is fairly established. I agree that the guidelines, including their outstanding treatment of relevant market, need no adjustment.
In the entire hearings, one of the issues that divided witnesses sharply, and is reflected in the pros and cons approach of the report, is whether there can be anticompetitive restraints in something called "an innovation market."
Should antitrust aggressively intervene and review the terms of technology transfers when the product is one, two, even five years away from actual sales? Those opposed to innovation markets argued that we lack a reliable theory of technological competition. Patent acquisitions, licensing and cross licensing can increase the speed of technological development but decrease the amount of technological rivalry going on - but we have no theory to guide us on how much technological competition is a good thing. Those opposing "innovation markets" also pointed out that supposed anticompetitive effects are often speculative, will often occur, if at all, many years after the technology arrangement is introduced, and may have no long term effects because barriers to entry to ideas are usually so low. Opponents argue with considerable force that antitrust should stay its hand during the technology phase and intervene only if technology arrangements lead to anticompetitive effects in the product market.
While the testimony of witnesses was divided, I continue to believe that in special circumstances, anticompetitive effects at the technology stage must be addressed as a matter of enforcement. Business witnesses were clear that technology competition in many sectors of the economy is becoming the crucial dimension of competition, and anticompetitive effects in the technology market (for example, anticompetitive patent acquisitions) may be difficult and expensive to remedy once a strong market position is established in the product market. There was powerful testimony that "first mover" advantages, often derived at the innovation stage, are more and more difficult to overcome in a high tech economy. The arguments against an "innovation market" should properly counsel great caution before enforcement authorities move in. To me those arguments do not counsel abandonment of enforcement efforts.(14)
Finally, Chapter 9 of the report, discussing competitive problems associated with networks and standards, and particularly the issue of appropriate access where a joint venture network or standards confer significant competitive benefits, is well worth careful review. In the year and a half that I have been back at the Commission, I have been struck by the frequency with which the anticompetitive effects of networks and standards have become critical to antitrust enforcement.(15) Chapter 9 summarizes the pros and cons of various approaches to these issues. The one thing that is certain is that this set of questions will test antitrust enforcement in coming years.
4. Conclusion.
The epilogue to the hearings report places this project in some historical perspective:
"One of the principal responsibilities of government agencies is to ensure that the laws that they enforce are regularly reviewed and occasionally adjusted to take into account changing conditions in the world. Congress created the Federal Trade Commission in 1914 with this in mind and instructed the new body to gather accurate and complete information about industry and the nature of competition. In the years since then, the FTC has held a number of hearings on issues of economic importance, leading to, among other things, Congressional adoption of the Securities Act of 1933, the Federal Communications Act of 1934, the Public Utility Holding Company Act of 1935, and amendments to Section 7 of the Clayton Act."
In reviewing caselaw and scholarship, and providing a forum for a broad range of expert opinion, these hearings and the hearings report are in the proper tradition of an independent agency whose responsibility includes insuring that the laws it enforces continue to make sense. Time will tell what impact, if any, this report will have.
It does seem that the extremely active enforcement of the 1960s (when global competition was insignificant) and the far less active enforcement of the 1980s (when cartel and horizontal merger enforcement dominated antitrust) are not quite the right answers for the 2lst Century. On the other hand, the cogent criticism of some 1960s antitrust theories has led to a more intelligent and balanced competition policy for the country, while discomfort with 1980s underenforcement has led (on a bi-partisan basis) to more aggressive enforcement in both the Bush and Clinton Administrations. A worthy ambition would be that these hearings can merge what was best in the thinking of the l960s and l980s and establish some common ground that is appropriate for the 21st Century.
1. The views expressed are those of the Chairman and do not necessarily reflect the views of the Federal Trade Commission or any other Commissioner or staff.
2. While the report addressed consumer protection as well as competition policy issues, this paper will be limited to the competition portion of the report.
3. See, e.g., Northwest Wholesale Stationers, Inc. v. Pacific Stationery & Printing Co., 472 U.S. 284 (1985); Broadcast Music, Inc. v. CBS, 441 U.S. 1 (1979).
4. See, e.g, Monsanto Co. v. Spray-Rite Serv. Corp., 465 U.S. 752 (1984); Continental T.V., Inc. v. GTE Sylvania Inc., 433 U.S. 36 (1977).
5. See FTC v. Procter & Gamble Co., 386 U.S. 568 (1967); Brown Shoe Co. v. United States 370 U.S. 294, 344 (1962).
6. U.S. Department of Justice, Merger Guidelines (l982)
7. The point was emphasized in the 1984 Merger Guidelines at Section 3.5 (1984).
8. See Pitofsky, Proposals for Revised United States Merger Enforcement in a Global Economy, 81 Geo.L.R. 195, 206-08 (1992).
9. Id. at 213-15.
10. Brown Shoe v. United States, 370 U.S. 294 (1963)
11. United States v. Von's Grocery Co., 384 U.S. 270 (1966)
12. See IV Areeda and Turner, Antitrust Law, pp. 146-96 (1980).
13. See Pitofsky, supra note 8 at 207-08.
14. In fact, the DOJ and FTC have brought innovation cases in recent years - generally based on evidence that there was a threat the merger would reduce innovation competition. See e.g., United States v. General Motors Corp. and ZF Friedrichshafen,Civ. No. 95-530 (D.D.C. 1993); United States v. Pilkington plc., Civ.No. 94-345 (D.Ariz. 1994); Boston Scientific Corp., File No. 951-0002, 5 Trade Reg. Rep. (CCH) 23,774 (D.D.C. 1995); Glaxo plc, File No. 951-0054,5 Trade Reg. Rep. (CCH) 23,784 (1995).
15. See, e.g., Dell Computer Corp., C-3658 (1996); First Data Corp., C-3635 (1996); Rite Aid Corp., File No. 961-0020 (1996) (preliminary injunction authorized, but not filed).