The current paradigms of antitrust law—price and ef.ciency—do not work well enough. True, they were an immense improvement over their predecessors, and they have served the .eld competently for a generation, producing reasonably accurate results in most circumstances. Accumulated experience has also revealed their shortcomings, however. The price and ef.ciency paradigms are hard to fully understand and are not particularly transparent in their application. Moreover, in a disturbingly large number of circumstances they are unable to handle the important issue of nonprice competition. In this article we suggest replacing the older paradigms with the somewhat broader approach of “consumer choice.”1 The choice framework has several advantages. It takes full account of all the things that are actually important to consumers— price, of course, but also variety, innovation, quality, and other forms of nonprice competition. It is also far more transparent, which is an important administrative virtue even where, as in the great majority of cases, it will reach the same result. And in some important real-world situations it will lead to better substantive outcomes. There are a number of variety-valuing industries and circumstances that can be assessed correctly only by including an effective analysis of nonprice factors. We identify several of those in the article. To illustrate their importance we
* Respectively, Attorney in the Of.ce of Policy and Coordination, Bureau of Competition, Federal Trade Commission; and Venable Professor of Law, University of Baltimore School of Law. The views expressed in this article are solely the authors’ own and are not necessarily those of the Federal Trade Commission or any individual Commissioner. We are grateful for valuable suggestions from Alden Abbott, Terry Calvani, Russell Damtoft, Albert Foer, Paul Halpern, Caswell Hobbs, Elizabeth Jex, Paul Karlsson, William Kovacic, Thomas Krattenmaker, Thomas Leary, Michael Moiseyev, James Mongoven, John Parisi, Suzanne Patrick, Robert Skitol, Mary Lou Steptoe, Randolph Tritell, Oscar Voss, and Erika Wodinsky. We are also grateful for helpful research assistance from Alice Arcieri, Fran Cariaga, Benson Cohen, Sarah Duran, Joseph Pulver, J. Andrew Stevens, Andrea Tony, and Thomas Werthman. Any errors remain our own.
1 We refer to this as the “consumer choice,” or sometimes, for linguistic ease, as simply the “choice” model.
go on to identify eight noteworthy recent cases that would probably have been decided differently under a choice approach. Throughout the article the focus is on the practical issues of day-to-day management, and we show how the choice approach can be made as predictable and administrable as the other paradigms.
The current price and ef.ciency models can deal only awkwardly with nonprice competition. At best, they try to help consumers achieve nonprice objectives indirectly, by folding them into the price analysis in the form of quality adjusted prices, or by assuming that markets that are price competitive will also be competitive for nonprice preferences. That surrogate analysis usually produces reasonable results, but it is not particularly intuitive. In some cases, moreover, it does not work properly. In those cases the choice factors will have to be addressed directly if they are to be considered at all.
Antitrust encounters at least three common situations in which a simple price analysis is inadequate. First, in some markets there is little or no price competition to begin with, as a result of regulation, joint ventures, or third-party insurance payors. There is no good way to assess consumer welfare in those markets without considering the nonprice choice issues. Second, some conduct—such as horizontal agreements to limit advertising—will increase consumers’ search costs or otherwise impair their decision-making ability. This will cause consumers to select products that are less desirable or less well-suited to their particular needs. A complete rule of reason analysis must take account of these adverse effects on suitability and satisfaction as well as the adverse price effects of the conduct. Finally, in some markets the .rms compete not primarilyonpricebutratherthroughindependentproductdevelopment or creativity. These efforts may involve areas, such as high-tech innovation, delivery of new patient-friendly hospital services, or editorial independence in the news media. Effective innovation in these markets may sometimes require more providers than are required to ensure price competition. Thus market concentration principles taken from a price context may not ensure robust competition in the respects most relevant to consumers of these kinds of products. In all three situations, the explicit use of a choice approach to antitrust is likely to lead to enforcement decisions that better re.ect consumer concerns and preferences.
Our proposal for dealing with these issues attempts to combine the virtues of narrowness and breadth—to offer both relatively cautious substantive reform and relatively broad conceptual change.
To begin with, the proposal accepts that the price and ef.ciency models have brought some much-needed discipline and rigor into antitrust analysis, and it advocates new consideration for choice in only a limited number of cases on the margin. The choice model is anchored in current practice in at least .ve different ways. First, in over 95 percent of cases either the relevant choice is still going to be based on price, or price competition will ensure effective nonprice competition. In these circumstances enforcement will simply continue along familiar lines. Second, even where the antitrust analysis should focus on nonprice effects, we propose only a more explicit and rigorous consideration of those factors than before, not a fundamental break with the past. Third, our approach would not condemn practices that result in only trivial reductions in the range of options. A reduction from ten to nine providers would not normally be an antitrust concern, even though there has been, in principle, some loss of variety.2 Fourth, the choice approach will not condemn practices that limit options through ordinary market competition. It asks only whether a particular business practice has resulted in some unreasonable and signi.cant limitation on consumer choice, unmediated by a marketplace test. And .fth, a choice approach is not a return to the “social and political values” paradigm of the 1960s and 1970s, which proved standardless and unduly hostile to business.3
A consumer choice theory based on these principles can operate in as disciplined and predictable a way as any other model of the antitrust laws. At .rst glance, the choice approach may seem to have less scienti.c objectivity and rigor than the ef.ciency or price models. Those older models, however, are based not only on science, but also on long experience and seasoned judgment. In this article we will demonstrate that a choice model, carefully developed through case-by-case analysis and supplemented by retrospective case studies and experimental economics, can build the same kind of empirical foundation for itself. Such a foundation will identify the relevant standards and thresholds, which can then be expressed and applied in administrable and predictable ways. For example, the enforcement agencies might announce that they will apply theHer.ndahl(HHI)4.guresintheHorizontalMergerGuidelinesmore
2 This is true by analogy to the price model, which does not condemn a practice likely to result in only a trivial increase in price.
3 The social-political paradigm rested on an underlying suspicion of or even hostility toward big business, and this animus is not present in an approach that merely tries to factor consumers’ nonprice desires into the analysis.
4 The Her.ndahl-Hirschman Index (HHI) is a measure of market concentration used in the antitrust agencies’ merger guidelines, calculated by summing the squares of the individual market shares of all the participants. See U.S. Dep’t of Justice & Federal Trade Comm’n, Horizontal Merger Guidelines (1992, revised 1997), 4 Trade Reg. Rep. (CCH) ¶ 13,104 at n.17, available at http://www.ftc.gov/bc/docs/horizmer.htm.
strictly in particular markets where choice is likely to be important, or choice might be identi.ed as an explicit additional factor for a rule of reason analysis.
Although making only moderate changes in practice, our proposed choicemodelisalsobroadlyandessentiallynewinprinciple.Itrepresents nothing less than a new paradigm of the antitrust laws, one that will be helpful throughout the antitrust .eld. The consumer choice approach is fundamentally superior to the price and ef.ciency paradigms because it asks the right question. It recognizes that consumers do not just want competitive prices—they want options.5 Framing the issue in this way starts the analysis on the right foot, presents the questions in a desirably transparent way, ensures that important long-term factors like innovation receive their full due,6 and helps to guard against circumstances in which enforcers inadvertently neglect important choice factors that are simply hard to translate into terms of price.7 Competitive prices will then become just one of the choices that are relevant to consumers—the controlling choice and the focus of analysis in the vast majority of cases, to be sure, but conceptually still a subset of choice.8
Most important, use of the new paradigm should result in better substantive outcomes in some important situations. A key section of the article reviews eight recent cases that would probably have come out differently under our proposed approach. Consider, for example, a merger that ef.ciently combines the last two defense contractors making air-to-air missiles—a product using cutting-edge technology. If the merger were accompanied by circumstances guaranteeing lower prices,
5 Consumers want books that re.ect their interests, not just cheap books; and they want pharmaceuticals that will cure their illness, not just cheap pharmaceuticals.
6 Innovation is the key to having future choices. The choice formulation thus brings to antitrust analysis an increased emphasis on two related elements: the short-term importance of nonprice options and the long-term importance of innovation. This can work either for or against an enforcement action. Sometimes the greater emphasis on innovation might make an innovation defense more likely to prevail, so that the new model could result in certain cases not being brought. In any event, we would see a somewhat different menu of cases.
7 This may be a more frequent shortcoming than is commonly realized. The price or ef.ciency approach typically de.nes “price” as price that is adjusted, somehow, for quality, variety, and innovation. See infra note 28. It is not clear how often these adjustments are actually made in reality.
8 This would parallel the structure on the consumer protection side of the FTC Act, where the largest single group of cases, involving deception, are conceptually a subset of the broader authority over “unfair practices.” See Int’l Harvester Co., 104 F.T.C. 949, 1060 (1984).
it might be acceptable under conventional analysis, but choice analysis would call attention to the bene.ts of maintaining a variety of competitive approaches in a high-tech product like this one, where the best path to future improvements is inherently impossible to predict.9 Or consider a hospital merger that does not threaten price competition but that brings all the hospitals in a market under the control of institutions that are unwilling, on religious grounds, to provide certain reproductive services, such as tubal ligations. Conventional price analysis might well permit this merger, but a choice analysis would highlight the importance of protecting the particular services that are at risk.10 Or consider a set of strong incentives to engage in exclusive dealing in a pharmaceutical product. Conventional analysis might have seen the incentives as involving only benign or even bene.cial price discounts, but choice analysis would ask whether the excluded alternatives would have been therapeutically better for some patients or if they might have been a useful starting point for future innovation.11
Approaching such cases in choice terms helps call attention to the relevant kinds of market failures. The industries in which variety and choice are most important tend also to be industries that are especially susceptible to information-related market failures. Among producers, it is sometimes hard to know what kinds of creative products a novelty-craving public or a rapidly evolving technology will demand. Among consumers, it is hard to know what kinds of innovations their suppliers could have offered but did not, and so it is hard for them to demand corrections from the marketplace. For both these reasons it may be important to have additional independent centers of innovation in such markets.
In addition to its primary substantive advantages, the consumer choice model also confers a number of practical administrative and managerial bene.ts. These include its ability to communicate antitrust policy in broadly acceptable terms to other governments in both the developed and the developing worlds, and to explain that policy in intuitive terms to important nonspecialist audiences, such as juries, Congress, and the
9 Cf. Press Release, U.S. Dep’t of Justice, Justice Department Requires Raytheon to Sell Key Electronics Businesses in Order to Go Forward with Its Hughes Aircraft Deal (Oct. 2, 1997) [hereinafter Raytheon Press Release], available at http://www.usdoj.gov/opa/pr/ 1997/October97/415at.html, discussed infra note 172.
10 Cf. Dominican Santa Cruz Hosp., 118 F.T.C. 382 (1994), discussed infra Part IV.C.
11 Cf. J.B.D.L. Corp. v. Wyeth-Ayerst Labs., No. 01-00704, 2005 U.S. Dist. LEXIS 11676
(S. D. Ohio 2005), appeal docketed, No. 05-3988 (6th Cir. Aug. 5, 2005), discussed infra Part IV.E.
general public. The model may also help to highlight possible synergies between antitrust and consumer protection theories in Federal Trade Commission(FTC)activities.Anditmayhelptorationalizetheallocation of cases between the FTC and the Department of Justice (DOJ).
Because the consumer choice model can lead to better analysis and results, and has no signi.cant drawbacks, it deserves to be—and it is in fact—emerging as the new paradigm of antitrust.
We elaborate this thesis in the remainder of this article, which is divided into six principal sections. Part I introduces the idea of consumer choice by de.ning the term, contrasting it with other plausible approaches to antitrust, and showing how it is not merely consistent with the decided body of antitrust case law, but actually the best way of explaining it. Part II starts the process of operationalizing these concepts, by reviewing the economics literature that sheds light on the optimal level of consumer choice. Part III discusses the particular industries and business circumstances for which an ef.ciency or price model is inadequate. Part IV shows that the discussion in the previous section has practical consequences, by identifying eight signi.cant recent cases that were handled one way under a price or ef.ciency theory, but probably would have come out differently under a choice theory. Part V shows that a consumer choice approach can be made at least as administrable and predictable as any other. Part VI identi.es some of the further administrative and managerial advantages of the choice approach. Finally, a brief conclusion explains why the shift to a new paradigm would have many precedents and, far from disrupting day-to-day administration of antitrust law, is instead the next logical step in its evolution.
I. DEFINING THE CONSUMER CHOICEAPPROACH TO ANTITRUST
The concept of “choice” pervades trade regulation law, at both the general and the particular levels. We begin with an introduction to the broad “consumer choice theory” used to explain trade regulation law as a whole—antitrust and consumer protection laws collectively—and then from this we derive a narrower “choice” interpretation of antitrust law in particular. We then explain how this choice-oriented approach to the goals of antitrust is broadly consistent with case law and the existing consensus on antitrust policy.12
12 We have addressed these background issues in a number of earlier theoretical papers. See Neil Averitt & Robert Lande, Consumer Sovereignty: A Uni.ed Theory of Antitrust
A. Nonprice Competition and the General Concept ofConsumer Choice
The proposed choice paradigm of antitrust law arises from a long experience with the operation and interpretation of the Federal Trade Commission Act. In particular, it arises from a general theory of consumer choice that has been developed over the last 20 years as a way of harmonizing the two functions of antitrust and consumer protection, as they have been presented by the two halves of that agency’s statute.
This general consumer choice theory suggests that antitrust and consumer protection laws perform different but complementary tasks.13 Operating together, these two bodies of law ensure that consumers have the two ingredients needed to exercise effective consumer choice14— options, and the ability to choose among them. Antitrust law protects a competitive array of options in the marketplace, undiminished by arti.cial restrictions, such as price .xing or anticompetitive mergers. Consumer protection law then guards against other market failures by ensuring that consumers are able to make a reasonably free and rational selection from among those options, unimpeded by arti.cial constraints,
and Consumer Protection Law, 65 Antitrust L.J. 713 (1997) [hereinafter Consumer Sovereignty]. For a more concise statement see Neil W. Averitt & Robert H. Lande, Consumer Choice: The Practical Reason for Both Antitrust and Consumer Protection Law, 10 Loy. Consumer L. Rev. 44 (1998). For a discussion of choice theory in an antitrust context, see Robert H. Lande, Consumer Choice as the Ultimate Goal of Antitrust,62
U. Pitt. L. Rev. 503 (2001) [hereinafter Choice as Ultimate Goal]. See also Robert H. Lande, Wealth Transfers as the Original and Primary Concern of Antitrust: The Ef.ciency Interpretation Challenged,34 Hastings L.J. 65, 124 (1982) [hereinafter Wealth Transfers] (antitrust violation requires a market failure external to consumers that “restrict or distort consumers’ available options”); Neil W. Averitt, The Meaning of “Unfair Acts or Practices” in Section 5 of the Federal Trade Commission Act, 70 Geo. L.J. 225, 281–82 (1981) [hereinafter Unfair Acts or Practices] (effective marketplace requires two elements: “a reasonable number of options” and ability “to make a free and rational choice from among those options”); Neil Averitt, The Meaning of “Unfair Methods of Competition” in Section 5 of the Federal Trade Commission Act,21 B.C. L. Rev. 227 (1980) [hereinafter Unfair Methods of Competition] (reviewing various theories by which marketplace options can be protected).
13 For a full elaboration of this thesis, see Averitt & Lande, Consumer Sovereignty, supra note 12. Each of these bodies of law is aimed at, and limited to, correcting the consequences of market failures.
14 In our earlier writings we have sometimes referred to this type of consumer choice as “consumer sovereignty.” The two terms have the same meaning. See Int’l Harvester Co., 104 F.T.C. 949, 1061 n.47 (1984) (referring collectively to “consumer choice or consumer sovereignty”).
such as deception or the withholding of material information.15 Together the laws function to protect a free market economy.16
B. The Paradigm of Antitrust Within the Choice Model
Antitrust .ts very comfortably within this framework. Its mission is to protect the array of options in the marketplace.17 The setting of antitrust policy within the more general “choice” framework has several implications for its proper construction, however. It suggests that the role of antitrust should be broadly conceived to protect all the types of options that are signi.cantly important to consumers. An antitrust violation can, therefore, be understood as an activity that unreasonably restricts the totality of price and nonprice choices that would otherwise have been available.
15 This interpretation was not changed by the passage of 15 U.S.C. § 45(n), which requires, among other things, that an unfair consumer practice be one that “is not reasonably avoidable by consumers themselves.” Section 45(n) is an additional screen, not a complete de.nition of unfairness. Consumers normally avoid injury through the exercise of choice in the marketplace. This view underlay the Commission’s 1980 Unfairness Policy Statement, 4 Trade Reg. Rep. (CCH) ¶ 13,203 (1980). Section 45(n) was intended to provide a rational, empirical means of determining when the conduct has impaired consumers’ ability to make choices.