Remarks before ABA/ Section of Antitrust Law Workshop: Joint Ventures and Strategic Alliances: The New Federal Antitrust Competitor Collaboration Guidelines
A major theme that I hope many of you subscribe to is the following: better late than never.
It is now an oft told tale that Federal Trade Commission staff, at the conclusion of our 1996 hearings on Global and High-Tech Competition (2) inquired of participants what area of antitrust law was most uncertain - especially where the uncertainty led to the business community avoiding transactions that might be legal and efficient. Overwhelmingly, the responses pointed to antitrust law with respect to joint ventures or, more broadly, with respect to horizontal collaboration. As a result, the Federal Trade Commission hosted a series of hearings and roundtable discussions and learned much of value from business and consumer representatives and academics.
The Antitrust Division and the Commission then turned to the challenge of drafting Guidelines that would be useful in clarifying the line between legal and illegal behavior. Susan DeSanti and her colleagues at the Commission, and Doug Melamed and his colleagues at the Antitrust Division worked long and hard on what I believe is the most challenging guideline project of them all. We have now published a draft for comment and look forward to a constructive exchange with a wide range of commentators.
By way of introduction, I offer some brief comments on the role and limits of guideline drafting. In my view, Guidelines should seek to achieve the following:
1. Guidelines should clarify the law and not depart radically from established judicial principles. Drafts that ignore that approach - for example, the Vertical Restraints Guidelines (3) published in the 1980s - are not likely to survive the passage of time.
2. There are circumstances in which the Courts, usually at the margin of the law, have indicated an intention to follow new principles. Guidelines that incorporate likely future concepts - as long as the people doing the drafting are not carried away by their own preferences - offer useful clarification to the private sector.
3. Guideline drafting is an evolving process. The modest but constructive changes in the excellent 1982 Horizontal Merger Guidelines are examples. It is precisely because it is a process that we have put these Guidelines out in draft form and will welcome comments and suggestions.
4. Guidelines ideally have an impact on a number of constituencies. They help the staffs of the enforcement agencies to follow a more uniform approach designed to produce similar outcomes in similar cases. They advise practitioners about the standards that agency staff will employ in investigation and enforcement actions, and that in turn enables staff and private counsel to have a more constructive exchange during investigations. Finally, Guidelines should educate private practitioners who have little or no expertise in antitrust law, alerting them to questions to resolve when advising clients about potentially vulnerable agreements.
As a last point of introduction, let me say a word about tone. The goal in this draft was to achieve balance. We accept that joint ventures are a common and useful device in an economy that is increasingly involved in global competition and that exhibits an expanding high-tech sector. One goal was to diminish the possibility that pro-competitive joint ventures might be abandoned because of uncertainty about the law. On the other hand, history demonstrates that some collaborations occasionally are designed primarily to curtail output and increase price - with few or no countervailing efficiencies. We certainly did not want to imply that the net is down and anything goes if a transaction can be labeled a "joint venture." My own view is that these Guidelines, by clarifying and emphasizing that true joint ventures are rarely challenged under the antitrust laws, tend to be encouraging and permissive.
These draft Guidelines touch upon scores of interesting and important antitrust issues. Nevertheless, it seems to me the heart of the project revolves around two challenging sets of issues:
1. The Guidelines attempt to present a single analytical framework that cuts across many types of agreements (for example, marketing, production, and research and development joint ventures) and cuts across industries.
2. The Guidelines address the line between antitrust's per se rule and its rule of reason, recognizing that sometimes the applicability of per se rules requires considerable inquiry, and sometimes the rule of reason can be applied, as Professor Areeda once wrote, "in the twinkling of an eye." (4) I will address issues that arise in each of those areas this morning.
A. Analytical Framework.
The central approach of the Guidelines is to focus on the competitive effect of a "relevant agreement," and to compare what competition would be like with or without that agreement. That approach has several important consequences.
1. By focusing on agreements, the Guidelines bypass more limited concepts like "entity," "corporation," or "partnership." In a high-tech society, collaborations among competitors often amount to no more than the integration of ideas, or protection of ideas such as patents and trademarks; these are the strategic alliances that businesses turn to in order to accomplish particular functions. The critical point where that analytical approach is adopted is in § 3.2 of the Guidelines which describes circumstances in which integrated economic activity can avoid per se challenges, and notes that such integration typically combines, "by contract or otherwise," significant complementary assets. This approach to integration was urged in some scholarship, and in my view was endorsed in the Supreme Court's decision in BMI, (5) but according to many remained uncertain in the law.
2. By focusing on agreement, the law can single out and isolate a part of an extensive collaboration that may be illegal, without challenging the entire collaboration which in its entirety may have a pro-competitive or neutral effect. Put another way, enforcement can proceed "restraint" by "restraint" - and not necessarily strike down a larger collaboration because some part of the arrangement is over the line into an area of illegality.
3. The analytical approach of the Guidelines is consistent with the goal of cutting across all types of joint ventures and all segments of the economy. The goal is to offer fundamentally the same analytical approach whether one is talking about R&D in basic metals, marketing arrangements for airlines, or production joint ventures designed to facilitate retailing on the Internet. Some may inquire whether that marks the death knell of other Guidelines that take up joint venture issues in the process of addressing specific issues in specific sectors of the economy. For example, how do these Guidelines relate to Guidelines with respect to the health care providers or intellectual property? The answer is that sector-specific Guidelines remain in effect. We see no reason why we cannot specify relevant efficiencies or describe safe harbors in the health care area in somewhat greater detail than in generic joint venture Guidelines. That happens where the enforcement agencies have a great deal of experience with a particular sector of the economy and therefore can be more specific in the rules of the road. In short, when the agencies know more, they will say more in terms of providing guidance.
B. The Line Between Per Se and Less Rigorous Antitrust Treatment.
1. Distinguishing Types of Analysis.
An immense amount of time in our hearings and in the drafting process was devoted to trying to draw a sensible and understandable line between per se and rule of reason.
The tension between the two types of analysis is first touched upon at page 3 of the Guidelines (Section 1.2) where per se is described, in accord with precedent, as types of agreements "so likely to harm competition and to have no significant pro-competitive benefit that they do not warrant the time and expense required for particularized inquiry into their effects." (6)
The Guidelines (Section 3.2) note that "typically these are agreements not to compete on price or output." Agreements of a type that might be considered illegal per se are nevertheless accorded rule of reason treatment if they are reasonably related to an efficiency-enhancing integration of economic activity and reasonably necessary to achieve its pro-competitive benefits......" (See Section 3.2). As noted earlier, the integration can be "by contract or otherwise."
What does all that language mean? First - competitor collaborations of a type that tend to lessen competition will nevertheless avoid per se treatment if they have an appropriate nexus to an efficiency-enhancing integration of economic activity. The Guidelines describe "integration" as collaborating to perform or causing to be performed one or more business functions such as production, distribution or R&D, and as likely to benefit consumers. Mere coordination of decisions on price, output and the like, without integration, does not qualify.
There is no limit in the Guidelines on the type of efficiencies that can contribute to an "efficiency-enhancing integration," other than that they must benefit or potentially benefit consumers. The illustration of an "efficiency-enhancing integration" as including a combination of "significant capital, technology or other complementary assets" - especially the reference to complementary assets - points to a broad range of possibilities. Moreover, recent government modifications in the Merger Guidelines to spell out more fully which efficiencies can be relied upon in defense of a merger, and the amendments in the Health Care Guidelines to expand cognizable efficiencies beyond those involved in financial integration, suggest a permissive attitude in recent years toward the range of efficiencies that can be taken into account.
Let me clarify with some examples. An agreement among competitors to divide markets - you stay west of the Mississippi I'll stay east of the Mississippi - is not an integration of economic activity; rather, it is what the Courts and scholars commonly described as "a naked restraint." On the other hand, joint marketing by business rivals - for example, where firms combined their sales forces to market a particular product or to cover a particular geographic area which they could not have done separately, or to accomplish marketing in a substantially more efficient way that is likely to benefit consumers - could be integrative. If so, a pricing agreement covering the jointly marketed products will escape per se treatment if the agreement is reasonably related to the integration and reasonably necessary to achieve its pro-competitive effects. That doesn't mean it is legal - only that more extended analysis is required. Note also that an "efficiency-enhancing integration of economic activity" is not limited to circumstances in which the collaboration "creates a new product." If the agreement improves quality or service, reduces price or increases incentives for innovation, all of those qualify as well. In that sense, these Guidelines fully embrace the principles and perhaps go beyond the narrow facts of BMI. (7)
2.Short and Long Versions of the Rule of Reason.
Rule of reason analysis is a flexible inquiry calling for an examination of a variety of factors in sufficient detail to understand accurately an agreement's competitive effect. Sometimes it can be conducted in an abbreviated manner, without losing accuracy. Under these Guidelines, there is room for a quick look role both to exonerate a collaboration or to find a violation. Generally speaking a quick look analysis focuses on the nature of the agreement and its effects, without a detailed inquiry into market structure (for example, barriers to entry) or market share. Let me note, however, that the absence of a detailed market inquiry does not necessarily mean no market inquiry. Market power, like other factors, is examined to the extent necessary to reach an accurate judgment of overall competitive effect.
If anti-competitive effects are clear and predictable from the type of agreement involved, a violation can be found in short order. (8) By the same token, if there is no actual anti-competitive effect and it is not the kind of agreement that traditionally leads to anti-competitive effects, it can result quickly in a finding of no violation. Note that these Guidelines in effect reject for purposes of quick look or for rule of reason treatment the idea of an absolute market power screen. We appreciate that market power is a key factor but in and of itself it should not be enough to condemn or exonerate a transaction - except in the circumstances described in the safe harbor sections of the Guidelines. Market definitions are sometimes too uncertain, and differentiation within markets may be too great, to rely on market share or concentration estimates to the exclusion of all other factors. Not only have the Courts not widely adopted market power screens, but I find no evidence that the law is moving in that direction.
3. Role of less restrictive alternatives.
These Guidelines clarify that less restrictive alternative is part of an analysis of whether a per se or rule of reason analysis is appropriate. Thus, efficiency-enhancing integration is described as one "reasonably necessary to achieve its pro-competitive benefits." There seems no point to engage in an extensive analysis of purported efficiencies if those very efficiencies could be achieved without the restraint at issue. The Guidelines do acknowledge, as I believe the law is beginning to do, that an agreement may be "reasonably necessary" without being essential. Enforcement officials should of course not be in the business of second-guessing reasonable expectations of parties that initiated joint ventures. In that respect the Guidelines indicated in Section 2.4 that assessment should be "sensitive to the reasonable expectations of participants." These Guidelines would defeat their own purpose if they undermine the incentives of joint venturers to initiate risky collaborations because of unnecessary fears that subsequent review, under changed circumstances, would undermine their efforts.
Some might question whether this treatment of less restrictive alternatives is consistent with my opening remark that Guidelines follow and clarify the law but do not introduce significant new interpretations. I believe this approach is not a new interpretation.
A focus on reasonable necessity is usual agency practice, as already reflected in Health Care Statements 8 and 9, and it flows from Supreme Court case law. For example, in Maricopa, (9) the Court concluded that even if a maximum fee schedule were desirable, it could be set by insurers rather than an agreement among the doctors, so there was a practical less restrictive alternative. In BMI, the Court asked the same question but reached a different answer, finding "a bulk license of some type a necessary consequence of the integration necessary to achieve these efficiencies," and then determining that a necessary consequence of an aggregate license was that a price must be established.
I believe, also, that a fair reading of the Supreme Court's decision in Topco (10) is consistent with this approach. In Topco, an association of small and medium size regional super market chains joined forces to purchase, store and distribute grocery products to its members. As a condition of membership, the grocery chains agreed not to sell Topco brands outside an assigned marketing territory, and members were also given the right effectively to veto new members who might offer actual or potential competition. I would describe Topco as an efficiency-enhancing integration, but would also conclude that provisions dividing markets and protecting incumbents from challenge were not reasonably necessary to achieve those efficiencies. Indeed, it seems reasonable to expect that additional members of the Topco collaboration would have made the joint venture more efficient, not less efficient, by aggregating additional purchasers.
I have only scratched the surface in discussion of relevant aspects of these Guidelines. I did choose the particular issues that I think were at the heart of debates over drafting, and raised the most challenging issues. Other speakers during the rest of this workshop will, of course, raise and discuss an array of additional subjects treated in this draft.
1. Chairman of the United States Federal Trade Commission. The views expressed are my own and do not necessarily reflect the views of the Commission or other Commissioners.
2. FTC Staff Report, Competition Policy in the New High-Tech, Global Marketplace (1996).
3. U.S. Dep't of Justice, Vertical Restraints Guidelines, 4 Trade Reg. Rep. (CCH) ¶ 13,105 (1985).
4. Phillip Areeda, The "Rule of Reason" in Antitrust Analysis: General Issues 37-38 (Federal Judicial Center, June 1981).
5. Broadcast Music Inc. v. Columbia Broadcasting System, Inc. 441 U.S. 1 (1979).
6. See, e.g., Northern Pac. Ry v. United States 356 U.S. 1, 5 (1958); NCAA v. Bd. of Regents 468 U.S. 85, 103-04 (1984); Catalano Inc. v. Target Sales Inc. 446 U.S. 643, 647 (1980).
7. In BMI, the Court concluded that a blanket license fee was not per se illegal price-fixing, in part because the product would not have existed in the absence of the joint pricing. But the Court did not conclude that the restraint would have been per se illegal if cooperation had not been "essential" to the creation of a new product. Under the Guidelines, an efficiency-enhancing integration does not result solely from the creation of a new product, nor must the restraint be essential to achieve such integration.
8. FTC v. Indiana Fed'n of Dentists, 476 U.S. 447 (1986).
9. Arizona v. Maricopa County Med. Soc'y, 457 U.S. 332 (1982).
10. United States v. Topco Associates, Inc. 405 U.S. 596 (1972).