Section 2 Remedies:
Presented at the
September 28, 2007
Thank you for the opportunity to speak today on the issue of remedies for unilateral violations of the antitrust laws.(1) When I addressed this audience two years ago, I commenced my remarks by praising the marketplace of ideas as articulated by John Stuart Mill in On Liberty. It seems appropriate to return to that concept today. The marketplace of ideas is one in which people of differing views compete to persuade others of the correctness of their views. This process is inherently beneficial. The "collision of adverse opinions" may reveal truth on both sides and may help both sides achieve a deeper and more vibrant understanding of their own beliefs.
Not only do we learn from the exchange, but the process also is likely to encourage convergence towards more consistent views. As Alexander Schaub, Former Director General for Competition, European Commission, explained in 2001:
Convergence is an organic process that grows out of learning from each other's experience, allowing all of us to retain the best elements. In a globalising world it is important to take an open-minded approach and constantly consider whether one's own rules and practices can be improved.(2)
Thus, I embrace and encourage encounters with those who hold different viewpoints. Indeed, as competition law enforcers, we of all people should recognize the benefits that accrue from a healthy and vigorous marketplace of ideas.
Turning to my specific topic, I will address three subjects today: (i) a brief reminder of the goals that antitrust remedies seek to achieve; (ii)suggested guidance for the formulation and assessment of remedies; and (iii) the application of those principles to Section 2 remedy issues.
I. Goals of Antitrust Remedies
There are four goals that antitrust remedies seek to attain:
An important caveat: Not all of these goals apply in every antitrust action. For example, enjoining a proposed anticompetitive merger prior to consummation likely does not require any further remedy to restore competitive conditions, compensate victims, or deter future proposed mergers. Similarly, in federal antitrust actions in the U.S., the agencies generally focus on injunctive relief. Victims can, of course, seek compensation through their own, separate actions.
I also observe that, as with the standards for determining whether there is a violation of Section 2 in the first instance, these goals relate to the process of competition. We seek to ensure conditions under which companies can compete and succeed or fail without unreasonable restraint. Thus, for example, in mergers we might require a divestiture, but we do not require that the defendant acquiring company limit its competition after the divestiture is accomplished. Indeed, to do so would itself be anticompetitive and, thus, antithetical to the antitrust laws. Similarly, in a Section 2 case, we do not seek to prohibit the defendant company from competing after a remedy has been put in place.
II. Guidelines for Implementation
These goals are relatively easy to state in the abstract, but much more challenging to apply in practice. To help implement these principles, I offer several guidelines for their application. I make no attempt to be comprehensive, but offer them as concepts for your consideration.
"[H]elp, or at least to do no harm."(3) Where we find a violation of Section 2, we seek an effective remedy. We need to consider, however, whether each remedy under consideration is likely to do more good than harm to consumer welfare. In short, if the antitrust laws are intended to advance consumer welfare, then any remedy should, at the least, not harm such welfare. For reasons discussed below, this guideline is not always easy to follow.
There should be a close nexus between the remedy and the proven violation. The remedy should be based on the violation that is proven by the plaintiff (or, in the case of a settlement, acknowledged by the defendant). While a remedy might not be limited to conduct that is identical to the conduct that constituted the violation, it should be limited to conduct that is closely related. The finding of a violation is not an unrestricted license for the plaintiff or court to restructure the industry.
The remedy should seek to re-establish the opportunity for competition or, put another way, an opportunity for the market to work. A tempting error is to think that a remedy is only successful it if leads to a reduction in the defendant's share of the market. Changing market shares is not in and of itself a proper goal of competition policy, even where a violation of antitrust law has been found. Requiring all firms to fight for share in the marketplace by seeking to better satisfy the demands of consumers is the best way to protect and enhance the welfare of consumers.
As an example, if the violation was the unlawful creation of a monopoly, the remedy might include the dismantling of the monopoly to restore the competitive environment that would have existed without the violation. Once that dismantling has occurred, however, the defendant should be free indeed, encouraged to compete aggressively to regain market share through lawful means. Otherwise, consumers lose the benefit of competition.
The remedy should use market competition to the greatest extent possible to achieve its ends and should minimize regulatory restraints, such as market share caps, price regulations, or other behavioral restrictions. The extensively discussed problems with behavioral remedies need not be repeated in detail here. Suffice it to say that agencies and courts lack the resources and expertise to run businesses in an efficient manner.
In the merger context, the issue is not as complicated because we deal with a restructuring that is being proposed by the parties to the transaction and that typically has not yet taken place. Our antitrust assessment directly addresses the question of whether the proposed restructuring will harm competition and, if so, the typical remedy is to prohibit the restructuring in whole or in part. The government can then step back to let the market work without on-going regulatory interference.
In the unilateral conduct context, however, we address conduct. That conduct may be related to the structure of the market, but is nonetheless separate from it. An attempted monopolization case or a monopoly maintenance case is a good illustration: the condemned conduct may or may not have altered the structure of the market that would have existed without the violation. Indeed, the size and structure of the firm might reflect the most efficient way to serve customers in the market. Accordingly, as the D.C. Circuit explained in its 2001 Microsoft decision, "structural relief, which is 'designed to eliminate the monopoly altogether . . . require[s] a clearer indication of a significant causal connection between the conduct and creation or maintenance of market power.' Absent such causation, the antitrust defendant's unlawful behavior should be remedied by an 'injunction against continuation of that conduct.'"(4) Thus, Section2 remedies often present the necessary, but difficult, challenge of behavioral relief.
One can easily point to examples of once large and powerful companies that either found themselves outpaced by the market or had to dramatically remake themselves to survive, such as General Motors, IBM, U.S. Steel, or Sony. Recognizing that we cannot predict how markets will change, we place time limits on our decrees. Otherwise, we can end up with situations in which a decree has become obsolete or, worse, has become an obstacle to competition. Consider a specific example of how the passage of time can alter the impact of a remedy. A 1941 Final Judgment entered in United States v. Allied Chemical and Dye Corporation(5) prohibited the defendant from gaining more than 35% of the sales of domestic ammonium sulfate, a type of fertilizer. The decree, which also prohibited the defendant from competing in the western United States, was still in effect earlier this year. The defendant moved to terminate the decree because it no longer had market power, and the decree was restricting the introduction of related new fertilizer products. The Department of Justice agreed and joined the motion to terminate the decree, which the court terminated last month.
The Department of Justice generally limits decrees to a ten-year term, but at times has adopted a shorter duration where warranted by the dynamic nature of the market.
Just as it is important to have clear, objective criteria for determining whether conduct violates Section 2, it is important to set forth clear, objective requirements for a defendant subject to a remedy. Otherwise, one risks deterring beneficial competitive activity. Further, if failure to comply with a remedy is to be penalized, basic notions of fairness and due process call for the defendant to be on clear notice of what is required.
No institution that enforces the antitrust laws is omniscient or perfect. Among other things, antitrust enforcement agencies and courts lack perfect information about pertinent facts, including the impact of particular conduct on consumer welfare. We have discussed this issue extensively in the context of determining whether a particular action violates the antitrust laws. We face the risk of condemning conduct that is not harmful to competition (a so-called Type I error, a false positive) and the risk of failing to condemn conduct that does harm competition (a so-called Type II error, a false negative).
A similar concern applies to fashioning a Section 2 remedy. In fashioning remedies, we should take into account to the best extent possible both the probability and the magnitude of possible errors.(6)
As observed above, agencies and courts have limited resources and are not expert business managers. Thus, remedies that require government entities to make business decisions or that require extensive monitoring or other government activity should be avoided whenever possible.
We have long recognized that markets are dynamic and that a particular action based upon a static model of the world can have unforeseen and undesired consequences in the actual, dynamic world. Thus, we need to recognize the incentives created by imposing a duty on a defendant to provide competitors access to its assets. Such a remedy can undermine the incentive of those other competitors to develop their own assets as well as undermine the incentive for the defendant competitor to develop the assets in the first instance. If, for example, you compel access to the single bridge across the Missouri river, you might improve competitive options in the short term but harm competition in the longer term by ending up with only one bridge as opposed to two or three.
As another example, the D.C. Circuit explained in its 2001 decision on the DOJ's Microsoft case that "courts are properly very skeptical about claims that competition has been harmed by a dominant firm's product design changes ....In a competitive market, firms routinely innovate in the hope of appealing to consumers, sometimes in the process making their products incompatible with those of rivals; the imposition of liability when a monopolist does the same thing will inevitably deter a certain amount of innovation."(7)
In the United States, we do not provide for civil fines in connection with unilateral conduct violations. The defendant is, of course, subject to treble damages in private actions. I nonetheless offer a few considerations regarding the use of civil fines in unilateral conduct cases.
First, there can be a benefit to imposing a fine as a remedy when it avoids the need to impose a behavioral injunction that might harm competition or innovation.
Second, if the fines will be significant, the need for clear, objective standards for defining a violation becomes even more important. In addition to the basic issue of fair notice to a company about what conduct might create liability for fines, uncertainty over the standards can deter beneficial competitive activity.
Third, some argue that fines might be a useful component of deterrence where private damages may be unlikely, such as where plaintiffs are unlikely to sue or where quantifying damages is unusually difficult.
Fourth, any fine should be proportionate to the violation, to avoid deterring pro-competitive activities.
Fifth, a fine should be based on harm in the relevant jurisdiction.
III. Application of these Principles
In the antitrust world, every case is fact-specific. Nevertheless, I offer several general observations about the application of these principles to forging a remedy in specific categories of unilateral conduct violations.
First, some conduct is relatively easy to remedy through an injunction that does not require extensive monitoring or expertise by the agency or court. For example, contractual tying can violate the antitrust laws and, if so, can be prohibited by decree. Such ties are relatively easy to identify and therefore to prohibit. Similarly, egregious actions to hamper a competitor, such as destroying a competitor's factory, can be identified and prohibited through a decree.
Second, some conduct falling into the same category is inherently difficult to identify and prohibit through an easily administered decree. For example, ties that are based on product design present extraordinary challenges. Predatory pricing is difficult to remedy. The Supreme Court has yet to define the appropriate measure of cost, and any uncertainty on these elements can discourage beneficial price discounting.(8)
The D.C. Circuit provided a specific illustration of problems that can arise in the context of alleged ties related to product design:
When IBM introduced [disk drives for computers] in 1956, it sold an integrated product that contained magnetic disks and disk heads that read and wrote data onto disks. Consumers of the drive demanded two functions to store data and to access it all at once. In the first few years consumers' demand for storage increased rapidly, outpacing the evolution of magnetic disk technology. To satisfy that demand IBM made it possible for consumers to remove the magnetic disks from drives, even though that meant consumers would not have access to data on disks removed from the drive. This componentization enabled makers of computer peripherals to sell consumers removable disks. Over time, however, the technology of magnetic disks caught up with demand for capacity, so that consumers needed few removable disks to store all their data. At this point IBM reintegrated disks into their drives, enabling consumers to once again have immediate access to all their data without a sacrifice in capacity. A manufacturer of removable disks sued.(9)
Third, particularly with regard to alleged tying relating to integration of components into a product design, the issue of price regulation can be difficult to avoid. Thus, for example, if a defendant is ordered to offer two versions of a product, one with and one without the component at issue, should the defendant be required to charge more for the version with the component and, if so, how much more? Courts and agencies are ill suited to regulating such pricing decisions.
Fourth, administrability and incentive issues are key reasons why we seek to avoid imposing liability for a unilateral refusal to deal or imposing a remedy that includes an affirmative obligation to deal with another competitor. Inevitably, the court or agency implementing such a remedy would need to regulate the terms of that dealing, and that regulation inherently would engender inefficiencies.
The Division's Microsoft case provides a useful illustration of many of these principles. The Antitrust Division proved that Microsoft violated Section 2 of the Sherman Act. The Division proved that Microsoft had a monopoly position in the IBM-based PC operating system market. The company had an extremely high share of the market that was protected by network effects and an applications entry barrier. Next, the Division proved that Microsoft unlawfully maintained that monopoly position through a range of exclusionary conduct. The gist of the violation was that Microsoft used exclusionary tactics with OEMs, software developers, ISPs, and others to squash a nascent threat to Microsoft's operating system monopoly from middleware products such as Internet browsers.
Middleware products are software products that can potentially expose common APIs across operating systems while providing a platform for other software applications. Thus, applications written to those middleware APIs could operate on any operating system with which the middleware was compatible. This process could break down the applications barrier and thereby increase the chance that Microsoft would lose its monopoly power.
Importantly, the Division never alleged that Microsoft had unlawfully acquired monopoly power in the operating system market. Nor did the government prove that the nascent middleware threat would have, in fact, eliminated Microsoft's operating system monopoly power or reduced its market share. Indeed, the Division's arguments concerning barriers to entry, which supported the finding of market power, also raised questions about the likelihood that the nascent middleware threats to Microsoft's operating system position would, in fact, have had such an impact.
The district court initially imposed a structural remedy after receiving only paper submissions and without any evidentiary hearings on relief. The court of appeals rejected the structural remedy based on such a record. On remand, the district court adopted a behavioral remedy that had been negotiated by the Division and a group of plaintiff states. The district court (and later the court of appeals) expressly rejected the call by other, non-settling plaintiff states for more extensive relief, as unwarranted by the violations that had been proven at trial and upheld by the appellate court.
As we near the end of the term of the final judgments for most of their provisions, we have taken stock of the impact of the decree on the proven violation. Looking to the principles discussed above, the remedy was well crafted and has been successful.
The Microsoft case also illustrates the important distinction discussed above regarding the success of a Section 2 remedy. The purpose of the remedy is to protect market opportunities, not to guarantee the success or failure of individual competitors. There is no doubt that Microsoft retains an extraordinarily high share of the operating system market. The Division did not, however, prove that Microsoft had gained that market share unlawfully or that Microsoft would have lost a significant portion of that share in the absence of the violations. To the contrary, Microsoft remains free to (indeed, is encouraged to) compete vigorously on the merits of its products.
Antitrust enforcement is, of course, global today, and we are engaged in extensive dialogues about the best way to enforce competition laws in a range of contexts, including unilateral conduct. Among other things, the ICN has a Unilateral Conduct Working Group that is stimulating discussion on these issues. In short, I'm happy to report that the marketplace of ideas is alive and well in this realm.
I will make several brief observations in this regard:
I believe that there is a general consensus in the international enforcement community that competition laws should be enforced to protect competition, not competitors. Indeed, just this week Commissioner Kroes of the European Commission reiterated her strong commitment to that principle. As she said, "U.S. and EU antitrust laws agree on most things, not least the objective of benefiting consumers."(11)
On a similar front, there seems to be consensus that we should prohibit unilateral conduct only where it is demonstrated through rigorous economic analysis to harm competition and thereby to harm consumer welfare.
As I discussed above, these principles are relatively easy to state in the abstract. Their application to a particular set of facts is often more challenging, and it is in this application that differences are most likely to arise. Useful topics for discussion include whether to presume that certain conduct has anticompetitive effects; whether and, if so, when a refusal to license an intellectual property right can violate the law; the degree to which protecting intellectual property rights is presumed to benefit innovation; and an assessment of the limitations of the institutions that administer remedies.
As John Stuart Mill explained long ago, we can only benefit from open discussion and debate over all issues surrounding the application of competition laws to unilateral conduct. Where we find that we agree, our discussions will nonetheless improve our understanding of the issues. Where we find that we do not agree, we will gain a better understanding of why differences exist and, in all likelihood, will move closer to each other in the process.
1. In the U.S., unilateral conduct generally would be evaluated under Section 2 of the Sherman Act. In Europe, it would generally be evaluated under Article 82.
2. Alexander Schaub, Director General, Directorate-General for Competition, European Comm'n, Continued Focus on Reform: Recent Developments in EC Competition Policy, address at the Fordham Corporate Law Institute 28th Annual Conference on International Antitrust Law and Policy (New York, Oct. 25, 2001), http://ec.europa.eu/comm/competition/speeches/text/sp2001_031_en.pdf.
3. Hippocrates, Epidemics, Bk. I, Sect. V.
4. United States v. Microsoft Corp., 253 F.3d 34, 106 (D.C. Cir. 2001) (internal citations omitted).
5. 42 F. Supp. 425 (S.D.N.Y. 1941).
6. See Ken Heyer, A World of Uncertainty: Economics and the Globalization of Antitrust, 72 Antitrust L.J. 375 (2005).
7. U.S. v. Microsoft, 253 F.3d at 65.
8. See generally Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993) (establishing a two-part, objective test for predatory pricing). The U.S. Supreme Court recently reaffirmed these concerns in the related context of a predatory bidding claim. See Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co., Inc., 127 S. Ct. 1069, 1078 (2007) ("Given the multitude of procompetitive ends served by higher bidding for inputs, the risk of chilling procompetitive behavior with too lax a liability standard is as serious here as it was in Brooke Group").
9. U.S. v. Microsoft, 253 F.3d at 94 (internal citations omitted).
10. Available at http://www.microsoft.com/about/corporatecitizenship/citizenship/businesspractices/windowsprinciples.mspx.
11. Neelie Kroes, Why Microsoft Was Wrong [Op-Ed], Wall Street Journal Europe, Sept. 26, 2007 [original text published separately by the European Commission sub nom "Antitrust in the EU and US our common objectives," http://ec.europa.eu/commission_barroso/kroes/antitrust_eu_us.pdf].