Jump to ContentJump to Main Navigation
How the Chicago School Overshot the MarkThe Effect of Conservative Economic Analysis on U.S. Antitrust$

Robert Pitofsky

Print publication date: 2008

Print ISBN-13: 9780195372823

Published to Oxford Scholarship Online: January 2009

DOI: 10.1093/acprof:oso/9780195372823.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2018. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 16 January 2019

The Efficiency Paradox

The Efficiency Paradox

(p.77) The Efficiency Paradox
How the Chicago School Overshot the Mark

Eleanor M. Fox

Oxford University Press

Abstract and Keywords

This paper defines various concepts of efficiency and then demonstrates how conservative economic approaches have led to wrong results in several important cases. It asks: What is efficiency? Can antitrust law produce efficiency, and how does it try to do so? It observes that one way antitrust pursues efficiency is by choosing a proxy; notably, either trust in the dynamic of the competition process or trust in (even) the dominant firm. By case examples, it shows the effect of conservative economics in choosing as the proxy trust in the dominant firm. It argues that this phenomenon has produced the Efficiency Paradox: In the name of efficiency, conservative theories of antitrust cut off the most promising paths to efficiency. It is suggested that we can eliminate the Efficiency Paradox by readjusting the pendulum to give more regard to the incentives of mavericks and challengers and less regard to the freedom and autonomy of dominant firms.

Keywords:   efficiency, conservative economics, antitrust, Brooke Group, California Dental Association, Trinko, Leegin, Microsoft


In 1978 Robert Bork wrote his influential book, The Antitrust Paradox. The paradox was that antitrust was meant to unleash competition but, Bork argued, it actually restrained competition. It did so by favoring small business and the underdog. His antidote was to reconstitute antitrust in the service of efficiency.

Thirty years later a chorus of conservative and libertarian policy makers and specialist technicians proclaim the new litany: Antitrust is for efficiency. The perspective has shifted from the notion that antitrust is for competition1 to the notion that antitrust is for efficiency. Many influential supporters of antitrust as efficiency, including jurists, presume that what business does is efficient and what government (antitrust enforcement) does is usually inefficient. Consequently, today, we face the Efficiency Paradox: Modern antitrust (I assume arguendo) is meant to help us reach efficiency. However, by trusting dominant firm strategies2 and leading firm collaborations3 to produce efficiency, modern U.S. antitrust protects monopoly and oligopoly, suppresses innovative challenges, and stifles efficiency.

To set the stage, this essay asks first, what is efficiency? Second, it asks: Can antitrust law produce efficiency, and how does it try to do so? It observes that one way antitrust pursues efficiency is by choosing a proxy; notably, either trust in the dynamic of the competition process or trust in (even) the dominant firm. Third, by case examples, it shows the effect of conservative (p.78) economics in choosing as the proxy trust in the dominant firm. It argues that this phenomenon has produced the Efficiency Paradox: In the name of efficiency, conservative theories of antitrust cut off the most promising paths to efficiency. Fourth, the essay suggests that we can eliminate the Efficiency Paradox by readjusting the pendulum to give more regard to the incentives of mavericks and challengers and less regard to the freedom and autonomy of dominant firms.

I. Mapping Efficiency

Efficiency is often categorized as allocative, productive, or dynamic. Allocative efficiency refers to the allocation of resources to their most valued uses, in view of the choices buyers make given their ability and willingness to pay at least cost for goods and services.

Productive efficiency refers to a firm’s production and distribution at the lowest feasible cost. Given the cost of inputs and the quality desired of outputs, a firm is productively efficient when it produces and distributes a good or service at the lowest cost possible.

Innovative or dynamic efficiency refers to the efficiency benefits achieved through research, development, and innovation, including the diffusion of technology to produce new products and processes. It includes firms’ production of knowledge (of what to make and how to make it), and cross-fertilization among firms to enhance the body of knowledge, all leading to improvements in the state of art.4

Dynamic efficiency gains can easily swamp static efficiency gains; that is, they can swamp the gains that result from pushing price closer to costs.5

Efficiency is sometimes assessed in terms of total welfare. For example, will a particular merger or course of conduct cause the sum of consumer plus producer surplus to shrink?6 Sometimes it is assessed in terms of consumer welfare. For example, will the merger or conduct cause consumer surplus to shrink. Even jurisdictions that define antitrust as a consumer welfare prescription (rather than a total welfare prescription) seldom focus only on whether consumer surplus will shrink. They also—at least sometimes—value consumer choice, both in terms of the variety of goods and services and the autonomy of consumers to choose.7 Moreover, they value producer incentives to invest and to innovate, which will normally inure to consumer benefit.8

In large economies such as the United States, conduct and transactions seldom lessen consumer welfare without also lessening total welfare, so the distinction between consumer and total welfare is often moot. The distinction is most likely to be material in merger analysis, and even then it is usually insignificant because merger parties usually cannot prove that their merger is likely to produce net efficiencies that could not otherwise be achieved.9

(p.79) Efficiency also applies to the form and formulation of rules and standards in view of administrative capabilities—the ease or difficulty of applying the law.10 As applied to conduct that is almost always harmful and almost never beneficial, a rule of per se illegally or quick look (and quick condemnation) is efficient. It saves enforcement resources, gives business greater certainty, and more effectively deters inefficient conduct. If the category is well-drawn, the bright-line rule will not be inefficient in the sense of condemning procompetitive and efficient conduct; but if it is overbroad—a problem produced by the growing per se category in the 1960s, per se illegality will handicap efficiency.

II. How to Achieve Efficiency

How to achieve efficiency is a complex question. In a market society, many factors and arrangements drive toward efficiency, in all of its senses. Eliminating unnecessary regulation produces substantial efficiency gains. A robust corporate take-over market often spurs firms to achieve more productive and dynamic efficiency.11 Intellectual property protection can spur innovation—although too much protection can create perverse counterincentives, frustrating innovative moves by outsiders. Antitrust policy, such as policy to dismantle antidumping duties on low-priced imports, can significantly unleash efficient firm behavior.

It is easy to see how antitrust policy can increase efficiency; but what can antitrust law do?

Antitrust law does not necessarily produce efficiency because it is proscriptive, not prescriptive.12 It can preserve an environment in which firms have the incentive to behave rivalrously and in which upstarts have a clear and open path to wage their challenges. This process or open-market perspective helps to preserve the incentives that produce productive, dynamic, and market efficiency. Since 1980, U.S. courts have retreated from the tradition of protecting the competition process (rivalry) and the openness of markets. They have shifted to a different inquiry: Will the outcome of a particular merger or conduct be inefficient by inducing the aggregate of all producers to reduce the total amount of goods they produce (i.e., will it lower market output)? If so, the merger or conduct is probably illegal, at least unless firm efficiencies outweigh consumer loss or, ex ante, the conduct was an attempt to serve consumers and the market. If not, the merger or conduct is legal. I discuss both perspectives below. I argue that the outcome paradigm is a crabbed perspective that was intended to and does minimize antitrust law. I argue that limiting antitrust to condemning inefficient outcomes and embroidering the analysis with conservative Chicago School economic presumptions (markets are robust; antitrust enforcement normally harms the market) shrinks antitrust law to its smallest possible scope and in doing so harms efficiency in the sense of undermining rivalry (p.80) and forestalling dynamic change, and as a result makes us economically worse off.13

A. The View of Antitrust That It Should and Does Prohibit Only Conduct and Transactions That Will Probably Produce Inefficient Outcomes

By this view, antitrust guards against certain limited (“anticompetitive”) interferences with market efficiency by firm conduct or transactions.

What are these interferences? (1) Dominant firms’ strategies that make no business sense14 except to put costs on rivals or block them from access to needed inputs or markets and that thereby confer on the dominant firm more power to increase price and lessen output are anticompetitive and inefficient;15 (2) Mergers that take competitors off the market under circumstances that will probably lessen output across the whole market, raise prices, and entail no offsetting cost-savings are also anticompetitive and inefficient.16 Permutations can extend the category of anticompetitive interferences,17 but not by much.

Apart from competitor price-fixing and other hard-core cartels, very little private action interferes with both competition and market efficiency.18 Moreover, if one presumes (as does conservative economics) that markets are efficient, the market unencumbered by antitrust produces efficiency.19 Accordingly, antitrust should largely “stay out of the way.” This is akin to the argument of the late-nineteenth-century Social Darwinists such as William Graham Sumner “who had nothing but contempt for the antitrust movement which was merely trying to place artificial obstacles in the way of natural evolution.”20

B. The View of Antitrust That It Should Also Preserve the Competitive Structure (Rivalry) and Openness of the Market

Supporters of this “process” view also want an efficient economy that produces what people want and need and that fosters innovation and growth of robust firms. They, too, know that we cannot engineer efficiency. But they believe that we can enhance efficiency and economic welfare (and other goals as well) by maintaining an environment congenial to mavericks and upstarts;21 an environment that induces firms to be rivalrous, to seek new ways to reduce their own costs, and to vie to meet buyers’ wants.22 A task of antitrust is to prevent this dynamic process from being undermined. Therefore, preserving access by outsiders, preserving contestability of markets, and, at high levels of concentration, valuing diversity, are seen as mechanisms of efficiency. Preventing inefficient outcomes is also an objective, but safeguarding the process is the first-line protector against bad outcomes.

This second perspective was embodied in the U.S. antitrust law for nearly a century, albeit without consciousness that the law was fulfilling an (p.81) efficiency goal. Indeed, in earlier times, the Supreme Court preferred competition (rivalry among sufficient numbers) to the lower costs of dominant firms, if a trade-off had to be made.23 By the late 1970s, the Court reckoned with the law’s overbreadth (it was trading off more firm efficiency for more rivalry) and eventually redefined competition of the sort the law should preserve to be harmonious with efficiency. But, in the 1980s and 1990s and in the new century, a conservative Court24 swung the pendulum from one inefficient position (too much antitrust because it disregarded incentives and efficiencies of dominant firms) to another (too little antitrust because it disregards incentives and efficiencies of firms without power).

This swing of the pendulum was possible because efficiency and how to reach it are complex concepts. There is no one thing called “efficiency.” Conduct, transactions, and markets have efficiency and inefficiency properties at the same time, and the relative dimensions of each property are affected by assumptions regarding how well markets work. How one applies a goal of efficiency, therefore, depends on what one values and stresses, as well as hunches as to what will produce the most efficiency—in all of its senses.

III. Conservative Economics

While the Supreme Court today purports to apply the antitrust laws in the name of efficiency,25 in fact it can be demonstrated that efficiency is not the guide to the resolution of the Supreme Court cases,26 which, in contemporary United States, are almost invariably decided for defendants.

In this section I take examples from four cases of the United States Supreme Court, and one case from the European Union. I ask: Did efficiency drive or determine this outcome? If not efficiency, what did?

A. The United States

I look at four United States Supreme Court cases: Brooke Group,27 California Dental Association,28 Trinko,29 and Leegin.30

1. Brooke Group

In Brooke Group,31 Liggett pioneered a generic (unbranded) cigarette, which threatened to make great inroads into the market shares of the branded cigarette oligopoly. Brown & Williamson (B&W), the smallest member of the oligopoly, fearing that generics would cannibalize its market share, introduced a no-frills fighting brand cigarette and a strategy of below-cost pricing. It expressed willingness to lose $48.7 million to discipline Liggett; it did lose almost $15 million by below-average variable-cost pricing over an 18-month period, and it won the war by forcing Liggett to raise its (p.82) own prices, closing the lion’s share of the price gap between branded and unbranded cigarettes and stemming the challenge of generics. B&W’s strategy was successful; it killed the generic segment of the market, just as it had set out to do.

The Supreme Court was called upon to adopt and apply a rule or standard that would govern predatory pricing. It was properly concerned about formulating a rule that would chill sustainable low pricing, but it was not particularly concerned by the prospect that an overly pro-defendant rule might cause the demise of new products that would threaten the established oligopoly. The Court announced, as it had done before,32 and contrary to fact: “there is consensus among commentators that predatory pricing schemes are rarely tried, and even more rarely successful.… ”33 The Court concluded that there was a need for a noninterventionist rule in price predation cases, and it formulated such a rule. It found that B&W had not violated the law because, the Court thought, B&W was unlikely to recoup its losses by raising prices in the future, and if that was the case the below-cost strategy not only benefited consumers in stage 1 (the low price) but threatened no harm to consumers in stage 2 (postpredation). The Court overturned the jury verdict for the maverick plaintiff. The Brooke Group rule—requiring below-cost pricing and probable recoupment by monopoly pricing—is the U.S. rule on price predation.

The Court’s presumption about the rarity of predatory pricing that works to exploit consumers was based on “theory” only, as adumbrated by conservative economists.34 Scholarship establishes,35 to the contrary, that selective price predation is a recurring phenomenon; it is used effectively to eliminate young rivals and to deter potential entry into noncompetitive markets.36 Recoupment (the payback) can come in many forms, including preserving the predator’s market power that would otherwise have been lost—as in Brooke Group itself.

Is the rule announced in Brooke Group efficient? It certainly has efficiency aspects, although the Court was overly bold in ignoring B&W’s own estimation that its predation would be worth it, and in ignoring the reality that B&W’s predation had killed off the challenge to the tobacco oligopoly from generic cigarettes. Would a ruling for Liggett have been efficient? It, too, would have had clear efficiency properties. A plaintiff’s victory along lines argued by its Supreme Court advocate Phillip Areeda37 would have encouraged competitive challenges to entrenched oligopolies. What broke the tie in this dramatic contest (in which Robert Bork argued for the defendant while Phillip Areeda argued for the plaintiff)? Conservative economics, which consistently privileged theory over facts.

2. California Dental Association

The California Dental Association was a professional association composed of most of the dentists in California.38 The dentists agreed to bylaws (p.83) that greatly restricted themselves from advertising their dental services. For example, the dentists agreed that they could not advertise simply: “10% discount for students,” “reasonable prices,” or “gentle care.” The Federal Trade Commission examined the bylaws. It found the restrictions on price advertising illegal per se, and it found the restrictions on quality and other advertising illegal after a quick (but still significant) look at the details of the market. The appellate court substantially affirmed the FTC’s decision. The Supreme Court reversed. Whereas the FTC had deemed the dentists’ rules harmful to consumers, who might, for example, want to locate a low-priced dentist, the Supreme Court took a different tack. First, The Supreme Court defined “anticompetitive” more narrowly—and statically—than the FTC had done. The advertising restrictions could be anticompetitive, the Court said, only if they caused California dentists to reduce the quantity of dental services provided in California.39 (This is the outcome perspective described above.) Indeed, the Court ruminated, the dentists’ bylaw restrictions might be procompetitive. They may have been used by the profession to prevent deceptive advertising. If they did prevent deception, the bylaws would give people more trust in dentists, and the increased trust might lead to an increase in demand and supply of dental services.

Was California Dental an efficient decision? Or were the dentists’ bylaws inefficient by suppressing information consumers wanted and chilling price discounting?40 Both the FTC and the Association had efficiency arguments. What broke the tie and induced the Court to hold that the FTC had to prove output limitation? Conservative economics, combined with a conservative—but widely held—view that professionals try to operate in the public interest and should be given wide range to regulate themselves.41

3. Trinko

AT&T, the old Ma Bell, was broken into one long distance telephone service company and seven regional “Baby Bells,” which were at the time legal monopolies.42 Bell Atlantic was a baby bell. It was the incumbent telecom service provider in the Northeast. Among other things, it inherited the local loop, access to which all local service providers need. When technological developments made competition in the local markets feasible, Congress passed the 1996 Telecoms Act, inviting multiple service providers into the formerly monopolized regional and local markets, and it required the incumbents to give the entrants full and nondiscriminatory access to the elements of the local loop on a cost-plus-reasonable-profit basis.

Bell Atlantic was not happy about the new competition and the cost-plus cap and it decided to impair its new competitors’ access to the local loop as a means to forestall their rivalry and prevent its customers from being siphoned off by the rivals. The discriminated-against rivals complained to the Federal Communications Commission. The FCC found that (p.84) Bell Atlantic had violated the 1996 Act and ordered remedies. A customer of one of the buffeted rivals sued Bell Atlantic in antitrust, claiming that the conduct of Bell Atlantic was also an antitrust violation.43 Bell Atlantic moved to dismiss the complaint, and the issue—was there an antitrust remedy for Bell Atlantic’s defaults—eventually went to the Supreme Court.

The Supreme Court ruled that the complaint had to be dismissed; the plaintiffs could not get to trial. The Court characterized Bell Atlantic’s strategy as a simple refusal to deal. It then asked: Did Bell Atlantic’s conduct come within the general principle that a seller—even a monopolist—has the right to refuse to deal; or did its conduct fall within a “narrow exception” from that rule? So framing the question was nine-tenths of a defendant’s victory. (The Court might have asked: Is the monopolist’s use of its power over a needed input to prevent its competitors from competing on the merits for the monopolist’s customers a violation of the Sherman Act?) The Court declared that freedom to deal is a first principle of monopoly law; that compelling a firm to share what it owns and may well have created is (1) a very serious infringement of the right to property and is likely to chill a firm’s innovation, and (2) is likely to drive competitors into cartels. Note that Bell Atlantic inherited the local loop, it did not invent it; and that cartelization was not a possible scenario in the case; the competitors needed the input to compete, not cartelize. Stressing that duties to deal are exceptional, the Court suggested that such duties should be ruled out in the absence of a prior voluntary course of dealing followed by a refusal to continue dealing in order to get higher monopoly profits in the future;44 or possibly, a duty to deal might be found in the case of denial of access to an essential facility where no access was being provided and no regulatory agency had a right to order access.

In the course of so deciding, Justice Scalia, writing for the Court, said that monopoly power is “good” because monopoly pricing “attracts ‘business acumen’ in the first place; it induces risk taking that produces innovation and economic growth.”45 The Justice called on courts to avoid false positives (erroneously condemning procompetitive conduct), and, separately to find no antitrust violation where effective relief would require considerable court or agency supervision.

Was Trinko efficient? The principles it recites certainly had efficiency properties—for duties to share what one has created may induce less investment to create. A judgment more sympathetic to the abused rivals and more concerned by Bell Atlantic’s perverse incentive to degrade the rivals’ access to the essential input over which it had sole control would also have had efficiency properties.

But what appears to have motivated Justice Scalia’s remarkable and unprecedented formulation of pro-dominant-firm antitrust law principles in Trinko? 46 Conservative economics.

(p.85) 4. Leegin

Leegin designed and produced belts under the brand name Brighton.47 It decided to sell only to select retail stores and to maintain one price, hoping not to “confuse” consumers with constant sales. It established a policy of not selling to retailers who sold Brighton products below its suggested prices. Kay’s Closet (PSKS), a women’s apparel store in Lewisville, Texas, pledged to adhere to Leegin’s new policy but later marked down its Brighton line by 20 percent in order to compete with nearby retailers. Leegin demanded that PSKS stop discounting, PSKS refused, Leegin cut it off, and PSKS sued, invoking the nearly century-old precedent that resale price maintenance is illegal per se (Dr. Miles).48 PSKS won a jury verdict that, as trebled, amounted to nearly $4 million. The case made its way to the Supreme Court, which confronted the question: Should resale price maintenance agreements remain illegal per se? The Court held no, reversing Dr. Miles by a vote of five to four.

The Court, by Justice Kennedy, said that per se rules are a disfavored category and are reserved for types of conduct that have manifestly anticompetitive effects; conduct that would always or almost always tend to restrict competition and decrease output. RPM does not fit this category. It can encourage retailers to invest in services and promotion by eliminating free riders, who would let other resellers provide the service and then undercut their prices; and it may give consumers more options to chose among low-price, low-service brands and high-price, high-service brands. It can encourage market entry of new firms that may choose to use RPM to induce investment and establish a reputation. The Court acknowledged that RPM can have anticompetitive effects, but, it said, these can be identified in rule of reason inquiry.

The Court rejected the argument that per se rules can properly serve a function of administrative convenience. It worried that per se rules in general might increase the total cost of the antitrust system by prohibiting procompetitive conduct, and increase litigation costs by promoting frivolous lawsuits.

Justice Breyer, dissenting, had a different view. He was reluctant to overturn the century-old Dr. Miles’s rule without any evidence showing that it had produced harmful effects on consumers; and there was no such evidence in the record. He cited a study showing that prices rose 19 percent to 27 percent when RPM was allowed. “The law assumes that … a marketplace, free of private restrictions, will tend to bring about the lower prices, better products, and more efficient production processes that consumers typically desire,” he said.49 Sometimes the probable anticompetitive consequences of a practice are so serious and the potential justifications are so few or so difficult to prove that a per se rule is justified, and these characteristics might justify the per se prohibition of RPM. As Breyer noted, the Court majority assumed that free riding that chills provision of services consumers find (p.86) useful is a significant problem. Breyer refused to accept this assumption. He cited respected economists who are skeptical that harmful free riding actually occurs. Moreover, he said, relegating RPM cases to a rule of reason category can make consumer-harming RPM cases too difficult to prove, and may tempt producers and dealers to adopt anticompetitive RPM that will go untouched by the law.

Did Leegin have efficiency properties? Yes. RPM can be used to enhance interbrand competition. But is it so used? And should we treat sympathetically a practice that always raises prices, as Breyer asked? Breyer wanted evidence.

Did efficiency drive the outcome in Leegin? No; it was conservative economics-based theory rather than fact.

Virtually every other contemporary U.S. Supreme Court antitrust case but one50 reflects the same ambiguities regarding efficiency. The majority opinions in all of these cases have applied some combination of the narrow output paradigm and a heavy thumb on the scale in favor of the autonomy of the dominant or leading firms.

B. The European Union

The European Union values openness, access, rivalry, and the competitive structure of markets as mechanisms to produce economic welfare, competitiveness, innovation, and market integration.51 The Microsoft case illustrates this perspective, which is an alternative to the outcome/output paradigm.52

In this section I concentrate on one of the two sets of conduct condemned by the European Commission and court as violating the abuse of dominance prohibition of the EC Treaty of Rome:53 Microsoft’s strategic withholding of information necessary for its rivals’ workgroup server software to interoperate with Microsoft’s operating system and other software.

Microsoft supplies more than 95 percent of personal computer operating systems (OSs). It has a “superdominant” position on this market, given its near monopoly and the constant reinforcement of its power as a result of network effects.

Novell and Sun Microsystems pioneered workgroup server software. Workgroup servers are computer servers for relatively small business establishments that deliver common file and print services and administration services to the group network. To do their job, workgroup servers must interoperate with Microsoft’s PC operating system.

Before Microsoft developed its own workgroup server software, Microsoft gave full interoperability information to the providers of workgroup server software. But it developed its own competing product and then withheld information, handicapping the competitors and causing them to underperform merely because they could no longer “speak” clearly to the operating system. The question before the European Court was whether (p.87) Microsoft’s refusal to provide the necessary information constituted an abuse of a dominant position within Article 82(b) of the Treaty of Rome, which prohibits dominant firms from abusing their dominance and in particular “limiting production, markets or technical development to the prejudice of consumers.”

The European Union, like the United States, generally allows firms, even dominant firms, to refuse to deal, since freedom of firms to choose their customers is normally good for the market and good for consumers. However, EU law—more than U.S. law—has crafted duties to deal. These duties are tightly circumscribed when they entail licensing intellectual property.

In the European Microsoft case, as noted, the European Commission charged Microsoft with withholding necessary interoperability information. Somewhat belatedly, Microsoft asserted that disclosing the information would entail disclosing intellectual property. The Commission disputed the claim that the protocols for which it sought disclosure were protected by intellectual property. It pointed out that Microsoft’s assertion of IP rights was an afterthought, and it argued that it was a pretext. Nonetheless, on Microsoft’s appeal from an adverse Commission decision to the European Court of First Instance, the Commission decided to frame its case under the demanding standards designed for refusal to license intellectual property.

The European Court upheld the Commission and found that Microsoft had indeed violated article 82. In doing so, the Court took a process-and-access approach to consumer welfare and efficiency.

The Court cited “the public interest in maintaining effective competition on the market.… ”54 It found that the Commission had established that the rivals’ products were innovative; that Microsoft, by strategic withholding of protocols, had killed off specific innovations by the rivals that consumers liked;55 that users rated the rivals’ workgroup server software more highly than they rated Microsoft’s on all qualities (e.g., reliability, availability, security) except interoperability; and that Microsoft’s “manifest and increasing lead over its competitors [was explainable] … not so much by the merits of its products as by its interoperability advantage.”56

The Court endorsed the Commission’s findings that “the refusal at issue had limited technical development to the prejudice of consumers,” and its judgment that “Microsoft’s arguments regarding its incentives to innovate did not outweigh” the exceptional circumstances that gave rise to the duty to disclose.57

Microsoft found no shelter in arguments (so well tailored to the ears of conservatively economic U.S. jurists)58 that it had “made significant investments in designing its protocols and the commercial success which its products have achieved represents the just reward”;59 that disclosure of its protocols would undermine its incentive to invent;60 and that an obligation to share the fruits of its investment with others would mean less investment in R&D.61

(p.88) Was the European Court’s decision efficient? It has significant efficiency properties. Computer users benefit from interoperability and from rivalry on the merits to supply the best applications. And they profit from rivals’ enhanced incentives to innovate better products.

Would a decision for Microsoft have had efficiency properties? Yes. A firm’s exclusive right to its system and particularly to its intellectual property conduces to productive efficiency and to innovation by the dominant firm.

Did “efficiency” decide the case? No. What made the difference? A perspective that applies nonconservative economics.

IV. The Future of “Efficiency” in Antitrust: Solving the Efficiency Paradox

For nearly 100 years, U.S. antitrust law stood against power. U.S. antitrust law was for competition, not centrally for efficiency, although efficiency was an expected by-product. The contemporary antitrust community posits that antitrust law is for efficiency and that the efficiency goal should drive the outcome of antitrust cases and limit the scope of antitrust.

In this chapter, I have assumed that antitrust law is for efficiency. I have demonstrated that efficiency is a multifaceted concept; that efficiency does not and usually cannot determine the outcome of cases;62 and that enforcers and judges know little about how to “reach” efficiency.

I have shown that contemporary U.S. cases are commonly (although not always) determined by a conservative perspective that has created the Efficiency Paradox. The Efficiency Paradox is that, in the name of efficiency, economically conservative U.S. antitrust law protects inefficient conduct by dominant and leading firms and thus protects inefficiency. Antitrust enforcers and jurists can topple the Efficiency Paradox. They can do so by recognizing that the output/outcome paradigm is just one means to identify anticompetitive conduct and transactions; by appreciating that conservative economic presumptions are commonly misaligned with the reality of markets; and by adjusting the pendulum to put more trust in open markets and dynamic rivalry and less trust in the autonomy of dominant firms.


(1.) U.S. antitrust law was adopted to contain private power; to give free and fair access to market actors without power; and to try to assure “that consumers get to determine who wins the competitive race.” See Irwin Stelzer, “Coping with Market Power in the Modern Era,” White Paper, Hudson Institute (Spring 2007), 12, available at www.hudson.org/files/publications/StelzerWhitePaperMarch07.pdf; Eleanor M. Fox, The Modernization of Antitrust—A New Equilibrium, 66 Cornell L. Rev. 1140 (1981).

(2.) See Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398 (2004).

(3.) See Texaco v. Dagher, 547 U.S. 1 (2006).

(4.) See Wolfgang Kerber & Nicole J. Sam, Competition as a Test of Hypotheses: Simulation of Knowledge-generating Market Processes, 4 J. ARTIFICIAL SOCIETIES & SOC. SIMULATION n0.3, http://www.soc.surrey.ac.uk/JASSS/4/3/2.html (2001).

(5.) See Joseph F. Brodley, The Economic Goals of Antitrust: Efficiency, Consumer Welfare, and Technological Progress, 62 N.Y.U. L. REV. 1020 (1987); F. M. Scherer, Antitrust, Efficiency, and Progress, 62 N.Y.U. L. REV. 998 (1987).

(6.) Losses to producers squeezed out of business by exclusionary practices are typically disregarded, whether they are squeezed out by competition itself or by anticompetitive conduct.

(7.) See FTC v. Indiana Federation of Dentists, 476 U.S. 444 (1986). See also Thomas B. Leary, The Significance of Variety in Antitrust Analysis, 68 ANTITRUST L.J. 1007 (2001); Neil W. Averitt & Robert H. Lande, Using the “Consumer Choice” Approach to Antitrust Law, 74 ANTITRUST L.J. 175 (2007).

(8.) See Trinko, 540 U.S. 398; United States v. Microsoft Corp., 253 F.3d 34, 49–50, 65 (D.C. Cir. 2001), cert. denied, 534 U.S. 952 (2001).

(9.) See, e.g., United States v. Oracle Corp. (Oracle/PeopleSoft), 331 F. Supp. 2d 1098 (N.D. Cal. 2004).

(10.) See FTC v. Superior Court Trial Lawyers Ass’n, 493 U.S. 411 (1990); Leegin Creative Leather Products, Inc. v. PSKS, Inc., 127 S. Ct. 2705 (2007), Justice Breyer dissenting.

(11.) Even so, particular takeovers are often inefficient. They can increase firms’ costs by creating cultural incompatibilities and decreasing flexibility and adaptability. A notorious example is Time Warner/AOL. See Rob Walker, Creating Synergy Out of Thin Air, N.Y. Times, July 28, 2002, C13.

(12.) In rare cases, antitrust law can be prescriptive. It can impose duties to open markets, as opposed to mandates not to close them by anticompetitive acts. Affirmative duties are disfavored in the United States. See Trinko, 540 U.S. 398; see Makin Delrahim, “Forcing Firms to Share the Sandbox: Compulsory Licensing of Intellectual Property Rights and Antitrust,” paper presented at the British Institute of International and Comparative Law, London, England, May 10, 2004, available at http://www.usdoj.gov/atr/public/speeches/203627.htm.

(13.) See Stelzer, supra note 1 at 14.

(14.) “No business sense” is one of several tests commonly suggested today as the screen and the standard for anticompetitive conduct by dominant firms. It is similar to the test suggested by Judge Bork, when he said in Rothery Storage & Van Co. v. Atlas Van Lines, Inc., 792 F.2d 210 (D.C. Cir. 1986): if the conduct or agreement is not designed “to restrict industry output, then [it] must be designed to make the conduct of their business more effective.” Id. at 221. Other proposed tests include whether the dominant firm sacrificed profits at stage one to make monopoly profits at stage two, and whether defendant’s conduct would destroy an equally efficient competitor. Balancing anticompetitive harms against procompetitive (efficiency) benefits would yield more enforcement, but is disfavored by the current enforcers.

(15.) See Schor v. Abbott Labs., 457 F.3d 608 (7th Cir. 2006), cert. denied, 127 S. Ct. 1257 (2007), Easterbrook, J.: “And if a manufacturer cannot make itself better off by injuring consumers through lower output and higher prices, there is no role for antitrust law to play.” Id. at 612.

(16.) Again, this formulation produces the narrowest scope of illegality. It is possible, but less likely, that vertical or potential-competition mergers will have this proscribed effect.

(17.) See, for critical role of presumptions, Andrew J. Gavil, Exclusionary Distribution Strategies by Dominant Firms: Striking a Better Balance, 72 ANTITRUST L.J. 3 (2004).

(18.) This paper is about antitrust law other than cartel law for the following reasons: The strong law against hard-core cartels (meaning: price, output, or market division agreements among competitors that are designed only to get rid of the competition among the parties and have no credible claim of being, for example, a synergistic joint venture) has been a staple of antitrust enforcement for a hundred years and is supported by all perspectives on antitrust. Liberals or pluralists support the law because cartels exploit their customers or suppliers and because they paradigmatically offend the principle of market governance by competition, not powerful firms. Libertarians or conservatives might support the law because cartels are inefficient and output limiting; and, since cartelists have no excuse that they are responding to and serving the market, the costs of error from the prohibition are virtually nonexistent. In the matter of cartels, liberals and (p.99) libertarians meet. The law is supported by economics and socio-political concerns. Only industrial policy advocates are likely to take exception.

(19.) Note that the outcome perspective requires that private action, to be caught by the law, must (probably) decrease output. This means that most tying and other uses of leverage by dominant firms, even if unjustified, will not be caught by antitrust. Leveraging is “only” a use of power, not an increase of power. See Trinko, 540 U.S. 398, at n. 4. For the contrasting EU approach to distortion of competition, see Microsoft Corp v. Commission, Case T-201/04, Sept. 17, 2007.

(20.) Hans B. Thorelli, THE FEDERAL ANTITRUST POLICY: ORIGINATION OF AN AMERICAN TRADITION 566 (1955). Note that Sumner’s argument was not that markets would always work well but that, however they worked, no law could stop them. As for the professional economists, “insofar as is known,” Congress considered one antitrust bill after another without calling on their advice. The legislators of the time distrusted experts. However, if Congress had sought the advice of the economists, it could not have expected support for the Sherman Act. Id. at 120–21.

(21.) Irwin Stelzer argues that process and access are all the more important in high tech industries, “ ’lest high-tech’ be converted to ‘my-tech’ by dominant firms” and powerful incumbents slow down or exclude “incursions of technologically superior challengers.” Stelzer, supra note 1 at 11, 14.

As to the importance of mavericks, see Jonathan B. Baker, Mavericks, Mergers, and Exclusion: Proving Coordinated Competitive Effects Under the Antitrust Laws, 77 N.Y.U. L. REV. 135 (2002).

(22.) The process may be regarded as iterative. It provides a learning and feedback mechanism. Firms observe, learn, compete, innovate, and adjust. See Kerber, supra note 4. See, for a description of antitrust rules and standards based on a process/open market approach, E. Fox, Abuse of Dominance and Monopolization: How to Protect Competition without Protecting Competitors, in Claus-Dieter Ehlermann & Isabela Atanasiu, eds., EUROPEAN COMPETITION LAW ANNUAL: WHAT IS AN ABUSE OF A DOMINANT POSITION? (Hart, 2006).

(23.) See Brown Shoe Co. v. United States, 370 U.S. 294 (1962).

(24.) SeeJeffrey Rosen, Majority of One: Stevens at the Supreme Court, N.Y. Times Mag., Sept. 23, 2007, 50.

(25.) This is not the precise language of the Court, but is readily inferred from the majority opinions. See, e.g., Leegin, 127 S. Ct. 2705.

(26.) See Thomas B. Leary, The Inevitability of Uncertainty, 3 COMPETITION LAW INT’L 27 (2007) (Journal of Antitrust Committee of International Bar Ass’n): Although the Chicago revolution substituted a “single lodestar”—economic welfare of consumers—for diffuse populist objectives, “this did not mean that cases would necessarily be easier to decide or to handicap. Quite the contrary.” Id. at 28.

(27.) Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993).

(28.) California Dental Ass’n v. FTC, 526 U.S. 756 (1999).

(29.) Supra note 2.

(30.) Supra note 11.

(31.) Supra note 26.

(32.) Matsushita Elec. Indus. Co. v. Zenith Radio Corp. (Japanese Electronics), 475 U.S. 574 (1986).

(33.) 509 U.S. at 226.

(34.) See, e.g.,John McGee, Predatory Pricing Revisited, 23 J. L. & ECON. 289, 292–94 (1980).

(35.) See, e.g.,Joseph F. Brodley, Predatory Pricing: Strategic Theory and Legal Policy, 88 GEORGETOWN L.J. 2239 (2000).

(36.) Nonetheless, in a subsequent case charging predatory buying (Weyerhaeuser) in which the dominant defendant overbought saw logs at inflated prices to eliminate its rivals from the market and did eliminate them, the Court reaffirmed its dictum in Brooke Group. It said that price predation almost never happens. The Court declared that predatory buying is the mirror image of predatory selling and that enjoining the high buying price (the first leg of buyer predation) was just as harmful to consumers as enjoining a low selling price, and that therefore the tough standards of proof for predatory selling should apply equally to predatory buying. Weyerhaeuser Company v. Ross-Simmons Hardwood Lumber Co., 127 S.Ct. 1069 (2007).

(37.) A ruling for plaintiff could have been based on the loss of consumer welfare in the particular case, as argued by Liggett’s appellate lawyer Phillip Areeda.

(38.) Supra note 27.

(39.) 526 U.S. at 776–77.

(40.) See Justice Breyer, concurring and dissenting: “[W]hy should I have to spell out the obvious? To restrain truthful advertising about lower prices is likely to restrict competition in respect to price—‘the central nervous system of the economy.’ ” 526 U.S. at 781, 784.

(41.) I stress conservative and not conservative/libertarian. Skepticism regarding professional self-regulation and state professional regulation is one point at which liberal and libertarian philosophies meet. Both FTC Chairman Michael Pertschuk (appointed by President Jimmy Carter) and FTC Chairman Timothy Muris (appointed by President George W. Bush) brought or supported proceedings against doctors and dentists and restrictive eye glass regulations.

(42.) Supra note 2.

(43.) The rival was AT&T. AT&T had already settled its regulatory and antitrust claim against Verizon.

(44.) This is the Court’s rendition of the “Aspen exception.”

(45.) 540 U.S. at 407.

(46.) I distinguish formulation of the antitrust principles from the outcome of the case. The case was a regulated industries case and, under an unusually procompetitive regulatory statute, the Federal Communications Commission had already taken action against the anticompetitive conduct.

(47.) Supra note 11.

(48.) Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911); overruled, Leegin, 127 S. Ct. 2705.

(49.) Leegin, 127 S. Ct. 2705.

(50.) State Oil Co. v. Khan, 522 U.S. 3 (1997) (an antitrust rule of law preventing maximum resale price agreements per se is clearly inefficient).

(51.) See Annual Reports on Competition Policy of the European Commission; e.g., Reports of 2005, 2006.

(52.) Microsoft, supraassumedEleanor M. Fox, What is Harm to Competition?—Exclusionary Practices and Anticompetitive Effect, 70 ANTITRUST L.J. 371 (2002).

(53.) Treaty of Rome establishing the European Community, as last amended at Nice, Official Journal C 325, Dec. 24, 2002.

(54.) Microsoft Corp. v. Commission, Case T-201/04, Court of First Instance, para. 691., Sept. 17, 2007, available at http://curia.europa.eu.

(55.) Id. at para. 654.

(56.) Id. at para. 407.

(57.) Id. at paras. 708, 709. The Court endorsed the Commission findings that Microsoft “did not sufficiently establish that [the required disclosure] would have a significant negative impact on its incentives to innovate.” Para. 697.

(58.) See, e.g., Trinko, supra note 2.

(59.) Microsoft Corp. v. Commission, para. 666.

(60.) Id. at para. 266.

(61.) Id. at para. 670.

(62.) An efficiency standard can, however, weed out cases in which a plaintiff’s victory would protect inefficient competitors at the expense of consumers. So, too, does a refined understanding of WHAT IS HARM TO COMPETITION. See Fox, What is Harm to Competition, supra note 52.