What to Do After Catching the Tiger by the Tail
THOMAS O. BARNETT
June 4, 2008
Good evening and thank you for inviting me to Charlottesville to discuss remedies under section 2 of the Sherman Act.(1) While I have said many times that section 2 in general presents some of the most challenging issues in competition law today, the specific subject of remedies is in many ways the most challenging of these issues. As I was considering how best to convey some of those challenges, I decided it would be useful to think about tigers.
Tigers, as you no doubt know, are majestic animals with a combination of strength, speed, and beauty that makes them a match for any lion in my book. They are in some respects similar to large, powerful companies that have come to dominate a particular market or set of markets. I will use my time tonight to explain how this is so by addressing ten interrelated issues.
1. The Benefit of Free-Range Tigers
The first is that free-range tigers are generally good. Free markets are the best means we have discovered to generate wealth in our society. As the Supreme Court memorably put it fifty years ago in Northern Pacific Railway, "The Sherman Act was designed to be a comprehensive charter of economic liberty aimed at preserving free and unfettered competition as the rule of trade" and rests on "the premise that the unrestrained interaction of competitive forces will yield the best allocation of our economic resources, the lowest prices, the highest quality and the greatest material progress."(2)
Thus, we generally want businesses to compete vigorously. But competition is not for the faint of heart. Judge Easterbrook has discussed this destructive aspect of competition many times in his writing on section 2. As he explains with Joseph Schumpeter's famous metaphor, "competition is a gale of creative destruction . . . and it is through the process of weeding out the weakest firms that the economy as a whole receives the greatest boost. Antitrust law and bankruptcy law go hand in hand."(3) Or to use my metaphor, it is in the nature of successful firms, like tigers, to pounce and devour, and to deprive other hunters of their prey. I prefer to watch tigers and successful firms — even dominant firms — from a safe distance and without interfering with their natural activities, confident that any harm they visit on competitors will — in general — redound to society's benefit.
I said "in general" because exceptions exist. Dominant firms also resemble tigers in that powerful animals can inflict great damage that sometimes should be prevented. Economic theory provides a sound basis for believing that firms can injure the competitive process by acquiring or maintaining monopoly power through anticompetitive means. And so we should seek to identify when and how government intervention is justified and will prove beneficial.
2. Think Before Grabbing the Tail
This leads me to my second point. Even though circumstances exist where catching a tiger is warranted, it doesn't mean that you should sign up for the safari without careful preparation. As I describe more fully below, catching a tiger is itself a dangerous endeavor, and good results from catching one are unlikely unless you have thought carefully before grabbing its tail.
Put another way, a bad section 2 remedy risks hurting consumers and competition and thus is worse than no remedy at all. That is why it is important to consider remedies at the outset, before deciding whether a tiger needs catching. Doing so has a number of benefits.
As an initial matter, having found a section 2 violation, you want to craft an effective remedy. It is not easy to craft what is typically a behavioral remedy that will achieve its desired objectives, avoid unintended harm, and be administrable. Thus, the remedy issue warrants careful thought up front.
Furthermore, contemplation of the remedy may reveal that there is no competitive harm in the first place. Judge Posner has noted that "[t]he nature of the remedy sought in an antitrust case is often . . . an important clue to the soundness of the antitrust claim."(4) The classic non-section 2 example is Pueblo Bowl-O-Mat, where plaintiffs claimed that the antitrust laws prohibited a firm from buying and reinvigorating failing bowling alleys and prayed for an award of the "profits that would have been earned had the acquired centers closed."(5) The Supreme Court correctly noted that condemning conduct that increased competition "is inimical to the purposes of [the antitrust] laws"(6)--more competition is not a competitive harm to be remedied. In the section 2 context, one might wish that the Supreme Court had focused on the injunctive relief issued in Aspen Skiing--a compelled joint venture whose ability to enhance competition among ski resorts was not discussed(7)--in assessing whether discontinuing a similar joint venture harmed competition in the first place.(8)
Even in circumstances where competitive harm theoretically could occur, the difficulty of designing a proper remedy may reveal that antitrust litigation cannot effectively remedy that harm. Since the Sherman Act's enactment in 1890, certain kinds of conduct appearing to harm competition have proven themselves beyond the limits of effective antitrust control.
For instance, the difficulty of providing an appropriate antitrust remedy was central to the Trinko Court's finding that there was no section 2 liability, notwithstanding the Court's acknowledgment of the potential "benefits of . . . intervention"(9)--benefits that another part of the government (the Federal Communications Commission) was already tasked with securing. The Court approvingly cited Professor Areeda for the proposition that:
"No court should impose a duty to deal that it cannot explain or adequately and reasonably supervise. The problem should be deemed irremedia[ble] by antitrust law when compulsory access requires the court to assume the day-to-day controls characteristic of a regulatory agency."(10)
Concluding that "[a]n antitrust court is unlikely to be an effective day-to-day enforcer of [the] detailed sharing obligations" requested, the Court found that section 2 had not been violated.(11)
The influence of remedy on liability standards occurs in other section 2 areas as well. In Brooke Group, for example, the Court acknowledged that above-cost pricing could theoretically have an "exclusionary effect."(12) But the Court eschewed imposing liability for it, basing that decision in part on the practical inability "of a judicial tribunal to control" it.(13)
Because price competition is so close to the heart of the competitive process that our antitrust laws seek to foster, I am not sanguine about section 2's prospects for appropriately remedying other forms of price competition posing theoretical harm to consumers absent clear, cost-based liability standards. In particular, there may be areas of bundled discounting and single-product loyalty discounts where any theoretical harm from above-cost discounting is beyond the practical ability of courts to remedy through antitrust litigation, much as above-cost pricing is in the context of predatory pricing. In this regard, I recommend to you the Antitrust Division and Federal Trade Commission's discussion of those discounting practices in a paper that we recently submitted to the OECD.(14)
In sum, it is critical to think hard about what you are going to do with the tiger before you grab its tail. If you cannot do something constructive, you should consider not grabbing it in the first place. And in any event, it is not the best time to determine what to do with the tiger while holding on to its tail.
3. Protect the Natural Process, Not Individual Animals
The third point relates to the modern view of ecology, where one is advised to care not about individual animals but about the health of species and ecosystems as a whole. Economics and antitrust take a similar approach: when designing a section 2 remedy, it is important always to keep in mind that the goal of catching the tiger is to protect competition, not competitors. That phrase comes, of course, from the Supreme Court's Brown Shoe decision, which concerned a merger challenged under section 7 of the Clayton Act.(15) But it is particularly apt in the section 2 context as well, where it is essential to distinguish between a remedy's impact on a section 2 violator's competitors and the remedy's impact on competition as a whole.
Because a section 2 violation hurts competitors, they are often a focus of section 2 remedial efforts. But competitor well-being, in itself, is not the purpose of our antitrust laws. The Darwinian process of natural selection described by Judge Easterbrook and Professor Schumpeter cannot drive growth and innovation unless tigers and other denizens of the jungle are forced to survive the crucible of competition. Good ecologists don't get emotionally involved with particular animals; good antitrust enforcers don't get too attached to particular competitors. It is the competitive ecosystem that matters.(16)
4. Preserve the Tiger's Spirit
My fourth point is that section 2 remedies should not crush a tiger's spirit; they should teach, not tame. Among other things, this means that equitable remedies should not interfere with the defendant's innovation incentives going forward. Today, it's widely understood that innovation is the greatest contributor to increasing welfare.(17) Future innovation is not less beneficial when it comes from a firm that has violated section 2 in the past. Simply put, a remedy hurts consumers when it deters firms that violated the law in the past from engaging in lawful, efficient, innovative conduct in the future.
Just as importantly, section 2 remedies also should not diminish the innovation incentives of firms competing with a section 2 violator. In Trinko, the Supreme Court accurately observed that forced-sharing obligations "may lessen the incentive" for rivals to invest in "economically beneficial facilities."(18) That observation does not merely apply in the refusal-to-deal context--no section 2 remedy should chill the incentives of industry participants to innovate.
5. Matching the Remedy to the Tiger's Transgression
In teaching the tiger, it's also important to consider the appropriate scope of remedy for a particular transgression. A finding that section 2 has been violated does not open the door to wholesale restructuring of a market. Instead, the remedy needs to be tied closely to the anticompetitive conduct occasioning it. That means that remedies need to be sufficient but not overbroad, proportional to the offense. Implementing a remedy that is too broad runs the risk of distorting markets, impairing competition, and prohibiting perfectly legal and efficient conduct.
The D.C. Circuit's Microsoft decision, where the theory on which liability was based was that Microsoft had illegally maintained a monopoly, illustrates the issue.(19) In the liability portion of its opinion, the D.C. Circuit rejected Microsoft's arguments that the United States had failed to show a sufficient causal connection between the challenged conduct and Microsoft's maintenance of a dominant market position.(20) The court also stated, however, that those arguments "over causation have more purchase in connection with the appropriate remedy."(21) And it was concern over the lack of a robust connection between the challenged conduct and the continued existence of monopoly power that led the D.C. Circuit to observe further that "[a]bsent some measure of confidence that there has been an actual loss to competition that needs to be restored, wisdom counsels against adopting radical structural relief."(22)
Put another way, unless you have established that the tiger should never have existed in the first instance, you have not established a basis for shooting it.
6. The Evolving Jungle
My sixth issue is the need to balance specificity and flexibility when designing equitable remedies. An ideal remedial decree allows the defendant readily to understand its obligations and the supervising court to determine quickly whether its terms are being met. This consideration calls for decrees to be specific and detailed. But such specificity may reduce the efficacy of a decree and even lead to unintended harm as markets evolve over time.
The importance of flexibility in any particular case is obviously tied to the duration of the decree. Long decrees risk becoming obsolete, with the unintended effect of potentially stifling a firm's ability to compete efficiently. The Division, for example, recently consented to termination of an old decree that appeared to be preventing the defendant from introducing a beneficial new product.(23) In recent decades, the Division's policy on decree terms has avoided the perpetual decrees that were sometimes sought in the past. Nevertheless, the length of the decree remains an important consideration when balancing specificity and flexibility.
7. Many Ways to Skin a Cat
The seventh issue I would like to raise concerns the many ways to skin a cat--or, in this case, the range of ways to teach a tiger. Equitable remedies in section 2 cases run along a spectrum. Prohibitory remedies typically seek to prohibit the conduct that was found unlawful, whether it be a contractual provision like an exclusivity clause or some other specific form of conduct found to be exclusionary. Where necessary or appropriate, prohibitory remedies may also entail fencing-in provisions prohibiting not only the specific conduct found to be unlawful but also other, related conduct.
When they are based on clear and objective criteria, prohibitory remedies confined to stopping and preventing the recurrence of the unlawful conduct are likely to be the most desirable in the section 2 context. They are relatively unlikely to be overbroad, create unnecessary inefficiency, or lead to high administrative costs because directing and determining compliance is more likely to be relatively straightforward. An example of this sort of targeted prohibitory remedy is the prohibition against per-processor licensing obtained through the Division's mid-1990s action against Microsoft.(24) That is not to say that administrative costs are never an issue with prohibition remedies. One need look no further than the litigation surrounding the line-of-business restrictions in the AT&T decree.
Prohibitory remedies are one of the two prominent kinds of conduct remedies in the section 2 context. Remedies imposing affirmative obligations are the other. When extensive changes have occurred in the affected market, it may be difficult or even impossible to reestablish an opportunity for competition simply by barring the specific exclusionary practices or other, similar conduct. In those situations, a conduct remedy requiring affirmative steps may be necessary to reestablish the opportunity for competition.
While affirmative-obligation remedies potentially can be effective, they can, among other things, raise significant administrability concerns. As the Supreme Court noted in Trinko, they force courts into "a role for which they are ill suited."(25) At the design stage, an access remedy typically requires specifying the nature of access, prices, and other terms of dealing. Specifying these matters in an efficient manner may prove difficult. Furthermore, access remedies can require extensive continuing oversight and adjustment to changing market conditions. In some circumstances, they may require decisions of a type traditionally vested in regulatory agencies with resources or attributes better suited for such determinations.
The Division's experience with affirmative obligations in our Microsoft case well illustrates the issue. I give great credit to those at the Division for a job that was exceptionally well done in crafting the remedy. Nonetheless, the decree has been time and resource intensive for the Division, the parties, and the court.
Access remedies also raise efficiency and innovation concerns. By forcing a firm to share the benefits of its investments and relieving its rivals of the incentive to develop comparable assets of their own, access remedies can reduce the competitive vitality of an industry.(26)
Thus, although affirmative-obligation remedies can play an important role in remedying section 2 violations, they must carefully balance the short-run benefits they bring to consumers with the considerable costs and inefficiencies they may engender. These concerns suggest that affirmative-obligation remedies are most likely to be suitable when ongoing regulation can be minimized and market forces quickly can be restored.
The last section 2 equitable remedy that I will discuss is the one that the Supreme Court has recognized as the "most drastic."(27) I'm talking, of course, about structural remedies, which require a violator to divest certain assets or even dissolve. Structural remedies can restore competitive conditions by reducing the defendant's market power, thereby providing existing or potential rivals the opportunity to compete effectively, or by constructing entirely new competitors.
In the merger context, the Antitrust Division generally prefers structural remedies over conduct remedies because, in the words of our Policy Guide to Merger Remedies, structural remedies are "relatively clean and certain, and generally avoid costly government entanglement in the market."(28) Because the parties to a merger either have not yet or have only recently merged, clear demarcations between entities and units generally still exist, facilitating a structural solution.
These advantages are generally absent in the section 2 context, especially where the firm in question has not grown through acquisitions. That greatly heightens the risk that a section 2 structural remedy will create market inefficiencies. And just as the problems of dividing a company into parts present challenges for a court, the separate entities created by divestiture may face challenges post-breakup due to a lack of personnel, organization, or information necessary to compete. Those challenges may be particularly likely in technologically dynamic markets.
Moreover, what structural remedies may gain in reduction of long-term administrative burdens, they may lose in imposition of short-run costs. Divestiture presents acute, up-front administrability challenges, forcing a court unfamiliar with a market's details to decide specialized questions involving, for example, an intellectual property portfolio or assessing the roles of different employees or groups of employees.
Thus, while structural remedies are an important part of the government's remedial arsenal, they should be used sparingly in the section 2 context and only where the violation has a clear, significant causal connection to the acquisition or maintenance of monopoly power, and even then only after determining that alternative remedies would not satisfactorily achieve the proper goals of a section 2 remedy.
To sum up, clear prohibitions based on objective standards are the easiest way to teach and, if properly crafted, least inhibiting to the tiger. Trying to make the tiger affirmatively sit or jump through a hoop is much harder, and drawing and quartering the tiger is rarely justified.
8. Care and Feeding of the Tiger
My eighth and related point tonight is that the care and feeding of the tiger after it is caught should be carefully considered. For reasons I have just mentioned, when formulating remedies, one should take into account the costs of food, monitoring the tiger, maintenance of any barriers, vet services, and the like. The effectiveness of a remedial solution depends in large part on how easily it can be administered or enforced. It is important to remember that a remedy's costs are paid by businesses and consumers, not just the government. Judicial oversight risks lowering affected businesses' efficiency and diminishing their innovation. Complex and relatively burdensome decrees may be warranted in some circumstances, such as those presented in the Division's Microsoft case; at the extreme, however, a remedy may be so difficult or expensive to administer that it is unlikely to provide significant net benefits to consumers.
9. Making the Tiger Pay
My ninth point concerns making the tiger pay for its illegal conduct. I only have time briefly to mention two areas that, in my view, warrant further consideration and that are therefore particularly appropriate to raise in this academic setting. The first concerns the trebling of antitrust damages and, in particular, whether it is appropriate in the context of section 2, where offending conduct is, typically, not concealed and where it is often difficult to distinguish lawful from unlawful conduct. The second concerns the ability of an enforcement agency to collect civil fines for section 2 violations. The Division has not sought authority to impose civil fines and we have some reservations about such authority, but it is certainly a topic worthy of discussion.
10. Who Else Is Hunting for the Tiger?
My final point tonight is that it is worth asking whether anyone else is trying to catch the tiger. Congress has established an overall antitrust regime in which private plaintiffs as well as others have the right to bring section 2 claims in addition to the federal government.(29) Thus, in addressing questions such as whether civil fines should be available for government section 2 cases, it is relevant to consider the complete picture. As one example, I do not have precise figures, but public reports indicate that Microsoft has paid billions of dollars to resolve private litigation involving claims that followed on from the Division's case.
Thank you for the opportunity to speak tonight.
1. I thank Joseph Matelis for his help in preparing these remarks.
2. N. Pac. Ry. v. United States, 356 U.S. 1, 4 (1958).
3. Frank H. Easterbrook, When Is It Worthwhile to Use Courts to Search for Exclusionary Conduct?, 2003 Colum. Bus. L. Rev. 345, 34546; see also Frank H. Easterbrook, The Limits of Antitrust, 63 Tex. L. Rev. 1, 5 (1984). See generally Joseph A. Schumpeter, Capitalism, Socialism and Democracy 87 (Harper Perennial 1976) (1942).
4. Brunswick Corp. v. Riegel Textile Corp., 752 F.2d 261, 267 (7th Cir. 1984).
5. Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc., 429 U.S. 477, 490 (1977).
6. Id. at 488.
7. See Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 U.S. 585, 598 n.23 (1985).
8. See generally Dennis W. Carlton, A General Analysis of Exclusionary Conduct and Refusal to Deal--Why Aspen and Kodak Are Misguided, 68 Antitrust L.J. 659, 662 (2001) (maintaining that "the only outcome to expect from court intervention" in situations like Aspen Skiing "is inefficiency").
9. Verizon Commc'ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 414 (2004).
10. Id. at 415 (quoting Phillip Areeda, Essential Facilities: An Epithet in Need of Limiting Principles, 58 Antitrust L.J. 841, 853 (1990)).
12. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 223 (1993).
15. See Brown Shoe Co. v. United States, 370 U.S. 294, 320 (1962).
16. This point is crucial to any meaningful evaluation of the effectiveness of a remedy. The effectiveness of a remedy does not properly depend only on the health and well-being of individual competitors.
17. I discussed this issue in greater detail in previous remarks. See, e.g., Thomas O. Barnett, Assistant Attorney Gen., U.S. Dep't of Justice, Maximizing Welfare Through Technological Innovation (Oct. 31, 2007), available at http://www.usdoj.gov/atr/public/speeches/227291.pdf.
18. Verizon Commc'ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 408 (2004).
19. United States v. Microsoft Corp., 253 F.3d 34, 10304 (D.C. Cir. 2001) (en banc) (per curiam).
20. Id. at 80.
21. Id. When the D.C. Circuit revisited this issue when reviewing the non-structural relief later obtained by the Division, its assessment was "Well done!" Massachusetts v. Microsoft Corp., 373 F.3d 1199, 1210 (D.C. Cir. 2004) (en banc).
22. Microsoft Corp., supra n.19, at 80.
23. Order Terminating Final Judgment, United States v. Allied Chem. & Dye Corp., No. 41-320 (S.D.N.Y. Aug. 8, 2007).
24. See United States v. Microsoft Corp., 56 F.3d 1448, 1452 (D.C. Cir. 1995).
25. Verizon Commc'ns Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 408 (2004).
26. See generally id.
27. United States v. E. I. du Pont de Nemours & Co., 366 U.S. 316, 326 (1961).
28. See Antitrust Div., U.S. Dep't of Justice, Antitrust Division Policy Guide to Merger Remedies 7 (2004), available at http://www.usdoj.gov/atr/public/guidelines/205108.pdf.
29. See generally William F. Baxter, Separation of Powers, Prosecutorial Discretion, and the "Common Law" Nature of Antitrust Law, 60 Tex. L. Rev. 661, 690 (1982).