The Atlantic Divide in Antitrust: An Examination of US and EU Competition Policy 9780226176246

How is it that two broadly similar systems of competition law have reached different results across a number of signific

193 90 2MB

English Pages 304 [319] Year 2015

Report DMCA / Copyright

DOWNLOAD FILE

Polecaj historie

The Atlantic Divide in Antitrust: An Examination of US and EU Competition Policy
 9780226176246

Citation preview

The Atlantic Divide in Antitrust

The Atlantic Divide in Antitrust: An Examination of US and EU Competition Policy Daniel J. Gifford and Robert T. Kudrle The University of Chicago Press :: Chicago and London

Daniel J. Gifford is the Robins, Kaplan, Miller & Ciresi Professor of Law at the University of Minnesota Law School. Robert T. Kudrle is the Orville and Jane Freeman Professor of International Trade and Investment Policy at the Hubert Humphrey School of Public Affairs and the Law School at the University of Minnesota. Both have written extensively on antitrust issues. The University of Chicago Press, Chicago 60637 The University of Chicago Press, Ltd., London © 2015 by The University of Chicago All rights reserved. Published 2015. Printed in the United States of America 23 22 21 20 19 18 17 16 15

1 2 3 4 5

ISBN-13: 978-0-226-17610-9 (cloth) ISBN-13: 978-0-226-17624-6 (e-book) DOI: 10.7208/chicago/9780226176246.001.0001 Library of Congress Cataloging-in-Publication Data Gifford, Daniel J., 1932– author. The Atlantic divide in antitrust : an examination of US and EU competition policy / Daniel Gifford and Robert Kudrle. pages ; cm Includes bibliographical references and index. ISBN 978-0-226-17610-9 (cloth : alk. paper) ISBN 0-226-17610-X (cloth : alk. paper) ISBN 978-0-226-17624-6 (e-book) 1. Antitrust law—United States. 2. Antitrust law—European Union countries. I. Kudrle, Robert T., author. II. Title. k3850.g54 2014 343.2407'21—dc23 2014029548 o This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

Contents 1 2 3 4 5 6 7 8 9 10

American and European Perspectives on Antitrust Welfare, Monopolization, Dominance, and Judicial Review Merger Policy and Efficiencies Price Discrimination Predatory Pricing Exclusive-Supply Contracts Single-Product Loyalty Rebates: Is a Large Gap Narrowing? Bundled Discounts Intellectual Property, the Two Microsoft Decisions, and Antitrust in Dynamic Industries A Summing Up

1

161 197

Notes Bibliography Index

217 277 291

25 39 63 83 101 117 139

1

American and European Perspectives on Antitrust A Comparison and a Puzzle This book examines US antitrust law and EU competition law as they apply to several current and important issues. We also address a puzzle posed by this examination: The United States and the European Union have competition laws that are broadly similar. They are concerned with maintaining competition in the marketplace. The courts and enforcement authorities employ similar language and concepts in their decisions. Yet in a range of significant antitrust issues, the applied law is, or appears to be, different in the two jurisdictions. In the following pages, we will review the role of efficiency in merger evaluations, price discrimination, predatory pricing, exclusive-supply agreements, loyalty, and bundled discounts and rebates. We will also explore the role of antitrust as it relates its essential facilities doctrine to intellectual property in the context of dynamic Schumpeterian competition often found in knowledge-based industries. In these different areas, we will find some similarities, but we will often find that different analyses predominate. Legal Similarities The United States has had an antitrust law since Congress enacted the Sherman Act in 1890.1 Europe has had a compe-

chapter one

2

tition law since the Treaty of Rome in 1957.2 Although the US law is much older, the basic structure of EU competition provisions resembles that of its US counterpart. Section 1 of the Sherman Act is directed against anticompetitive concerted action; Article 101 of the Treaty similarly targets anticompetitive concerted action. Section 2 of the Sherman Act is directed against all acts of monopolization or attempts to monopolize, whether unilateral or concerted, while Article 102 is similarly directed against all abuses of dominant position, whether unilateral or concerted. In addition to the Sherman Act, Congress enacted the Clayton Act in 1914.3 One of the most important provisions of the Clayton Act is Section 7, governing mergers. Section 7 was cast in its present form by an amendment adopted in 1950.4 The European Union (through the Council of Ministers) adopted its analogue to the revised Section 7 in the Merger Regulation of 19895 and its revision in 2004.6 Both the US and European systems have developed categories of behaviors that are deemed inherently unlawful. In the United States, these categories are referred to as per se illegal. In the European Union, this approach is referred to as form-based (as opposed to effects-based) analysis. Examples include horizontal price-fixing agreements and tying arrangements, which have historically been treated as inherently unlawful in both jurisdictions.7 Both the US and European systems employ safeguards designed to prevent condemnation of cooperative arrangements that benefit the public by generating efficiencies. In the United States the so-called Rule of Reason accomplishes this end. The Rule of Reason requires that the government (or private plaintiff) carry the burden of proving that the business arrangement being challenged reduces the supply of goods in the general interbrand market, although this difficult burden can sometimes be mitigated through burden allocations of subordinate issues. In the European Union a safeguard inheres in the structure of Article 101: Article 101(1) sweeps broadly to prohibit all concerted practices that have as their object or effect the prevention, restriction, or distortion of competition, but Article 101(3) then exempts arrangements that contribute to improving the production or distribution of goods or to promoting technical or economic progress and afford consumers a fair share of the resulting benefit. The broad balancing under Article 101 is carried over to Article 102 through judicial decisions. The Merger Regulation carries a similar balancing with different language. In short, EU antitrust is governed by provisions similar to those governing US antitrust. It employs per se rules aimed at behavior deemed inherently anticompetitive, just as US antitrust law does. Like US antitrust, the European law employs the equivalent of a Rule of Reason designed to protect efficiency-enhancing behavior from condemnation. In subsequent

american and european perspectives on antitrust

3

chapters, we will examine how the similar legal structures in the United States and Europe often generate substantial differences in practice. We direct our attention in this chapter to a broad overview of the current similarities and differences between the competition policies 8 of the two jurisdictions and to the factors that have shaped these policies and their administration. These similarities and differences can be best understood in historical context, by comparing competition policy developments on both sides of the Atlantic in terms of the interactions of three sets of considerations across time: ideology, institutions, and interests.9 Many of these issues will be revisited later when individual competition policy issues are examined. Ideology and Doctrine The relative absence of the residue of feudalism left the American Republic largely free of both reactionary and revolutionary impulses.10 The convulsive history of Europe since the French revolution has generated a far broader array of basic postures toward government and its role in the economy, both within states and among them. This fundamental difference must condition all serious considerations of state action, including competition policy. As the following discussion will illustrate, much of this difference finds roots in several distinctive characteristics of European thought and experience that have almost no counterpart in the United States: 1. The role of cartels, 2. The role of the national state in the economy, 3. The role of political forces skeptical of market competition or capitalism more broadly, 4. The role of ordoliberalism as a political philosophy, and, in the period since 1957, 5. The use of competition policy to mesh national markets with each other.

The United States. The United States stands sharply apart from Europe at

the level of ideological complexity. Many writers have noted the relative narrowness of America’s ideological foundation. In the middle of the twentieth century, Arthur Schlesinger saw the American political environment as similar to Emerson’s description of it a hundred years earlier: a broadly prevailing consensus marked by two strands of liberalism, one more optimistic about the efficacy of state action and one more skeptical.11 And that situation has continued.

chapter one

4

All accounts of the development of competition policy stress the singularity of the US Sherman Act of 1890,12 which responded to a confluence of developments in post– Civil War America. Increasing economies of scale and scope in production were enabled and increasingly strengthened by improvements in transportation and communication. This contributed to a drop in commodity prices as part of generally falling general price levels that disadvantaged debtors. These circumstances created irresistible political pressure for government action to curb the power of the great manufacturing and railroad enterprises that came to dominate American commerce in the late nineteenth century and that seemed to be benefiting at the expense of smaller firms and independent agricultural producers.13 The erratic enforcement of the Sherman Act— and the subsequent passage of the Clayton Act and Federal Trade Commission Act14 in 1914 that were intended to clarify, enforce, and strengthen it— has been related well and often.15 The Sherman Act attempted to protect small market participants from inappropriate behavior by larger private actors, but to do so through a unique and circumscribed form of public intervention that was consistent with the almost universally prevailing belief in limited government. This view (allowing limited government intervention through the courts in affairs that are otherwise subject to market forces) has been broadly accepted and has endured with minor exceptions over the entire period since. At the time of its enactment the kind of intervention authorized by the Sherman Act had little parallel, domestic or foreign. Its initial uniqueness is often largely ignored because it has become so familiar in the United States and, very recently, almost everywhere.16 The Sherman Act particularized the idea of selective intervention in private markets by applying two concepts from the common law: “restraint of trade” and “monopolization.” These concepts were used to craft tersely worded legislation that relied for enforcement on administrative and private initiative in bringing cases before the regular courts.17 The consequent judicial decisions establish the rules for competition policy. After enacting the Sherman Act and following it up with the Clayton and Federal Trade Commission Acts, Congress was largely content to allow the courts to develop the applicable legal standards, intervening only occasionally. Yet the judicially set rules for competition policy leave much unsaid. As David Evans has recently pointed out, “unlike most sports, the rules for the game of competition say little about how the game should be played, only how the game should not be played. The rules identify certain foul practices, though with varying degrees of particularity.”18 The period from 1890 to 1933 saw considerable variation in enforcement: some of the most heavy-handed remedies in US history can be found

american and european perspectives on antitrust

5

here (the dismemberment of Standard Oil; the restriction of International Harvester to one dealership per town), but for the most part enforcement was lax even after the passage of the Clayton and Federal Trade Commission Acts in 1914. The former attempted to specify better what actions by individual firms, particularly price discrimination and mergers, were to be antitrust violations, while the latter, in section 5, established a specialized agency that would combat “unfair methods of competition” that remained unspecified in the other legislation. Ever since the Federal Trade Commission Act was passed, there has been continuing dispute about what this means. Does it refer to the same kind of competition issues as the other laws only perhaps earlier (in their “incipiency”) or more broadly construed (lower thresholds for action), or is it essentially redundant absent fresh agendas such as that provided in 1938 when the Wheeler-Lea amendment added “unfair or deceptive acts in commerce” to the Federal Trade Commission’s (FTC) agenda, seemingly empowering it as a proto-consumer-protection agency. This period saw the development of the broad parameters of antitrust enforcement, including the Supreme Court’s adoption of the Rule of Reason as the basic analytical approach to antitrust in its twin decisions in the Standard Oil and American Tobacco cases; only activities that “unreasonably” restrain trade were to be illegal, but price-fixing agreements were deemed per se illegal.19 Congress’s enactment of the Clayton and Federal Trade Commission Acts three years later was at least in part a reaction to the Court’s embrace of the Rule of Reason in 1911. Although Justice Louis Brandeis set forth the classic parameters of a Rule of Reason case in 1918,20 generally during the 1917– 18 period antitrust enforcement was restricted in the interests of war preparation. In 1933 the Franklin Roosevelt administration also initially pursued a policy of suppressing antitrust enforcement. Challenged by the Great Depression, the president and the Congress experimented with a form of corporatism under the National Industrial Recovery Act (NIRA).21 Under the NIRA, industries were required to organize themselves cooperatively under presidentially approved “codes of fair competition,” thus temporarily abandoning competition as a social goal, leaving the Sherman and Clayton Acts in a kind of political limbo. This hiatus in antitrust enforcement ended abruptly in 1935 when the Supreme Court invalidated the NIRA as an unlawful delegation of legislative power to the president. The next period, 1935– 74, began with the New Deal’s abandonment of sympathy for firm cooperation to maintain prices and output and a renewed determination to assure competition. This return to an active antitrust policy began with the appointments of Robert H. Jackson and Thurman Arnold as heads of the Antitrust Division, both of whom aggressively

chapter one

6

enforced the antitrust laws. But Congress also passed the Robinson-Patman Act in 1936,22 legislation that targeted price discrimination by suppliers that, in Congress’s view, disadvantaged small retailers vis-à-vis their larger rivals, such as chain stores. The act effectively replaced protections accorded small retailers under the NIRA. In 1938 Congress established the Temporary National Economic Committee that produced important studies on the state of competition and set the stage for the postwar amendment of the Clayton Act’s merger provisions. This renewed focus on antitrust was short-lived, as rearmament for World War II resulted in another temporary suspension of antitrust enforcement. Antitrust revived with the end of the war. The Second Circuit Court of Appeals issued a final decision in the Alcoa case,23 setting forth a proposed framework for construing the “monopolization” clause of Section 2. The Supreme Court added other foundational monopolization cases in Griffith24 and Grinnell.25 The failure of the government’s case against Columbia Steel 26 gave the final impetus to Congress to enact the Clayton Act’s revised antimerger provisions in 1950.27 During the postwar period, the Court set forth a per se rule against tying agreements28 that became more severe throughout the 1950s and 1960s. The Court gave an expansive construction to the original language of the Clayton Act dealing with stock acquisitions.29 The Court greatly expanded antitrust law during the 1960s. Construing the amended version of the Clayton Act’s merger provisions, the Court extended the scope of their prohibitions until they virtually barred all horizontal mergers by companies of any significant size. In Von’s Grocery,30 the Court barred the merger of two grocery chains that collectively possessed only 7.5 percent of the relevant market, and in its Pabst 31 decision later the same year, the Court’s opinion indicated that the Clayton Act would bar a merger of companies collectively holding an even smaller market share. In Pabst, the Court also dispensed with the need for the government to prove any relevant market. In 1967 the Court condemned the underpricing of a locally dominant frozen pie producer by out-of-state suppliers as predatory pricing and unlawful price discrimination, even though in fact intense price competition among the producers reduced concentration in the local market, reduced consumer prices, and did not produce losses for the local firm.32 In that year, the Court also condemned as per se illegal a bicycle manufacturer’s imposition of territorial and customer limits on its distributors.33 The third period, 1974 to the present, cast antitrust law as an engine to generate economic welfare through the promotion of efficiency. In prior periods, antitrust law was understood as concerned with the furtherance of several— and not always consistent— goals: efficiency in resource use,

american and european perspectives on antitrust

7

economic growth and innovation, stability in output and employment, equitable income distribution, fair conduct, and limiting the size of large business units.34 At least from 1974 onward, however, antitrust law began to coalesce around the single goal of fostering efficiency. Richard Posner and Robert Bork, among others, argued effectively that this single goal was the logical understanding of a law enacted to preserve marketplace competition.35 After extensive academic journal debate, a consensus developed around efficiency as the exclusive goal of the antitrust laws with the corollary that antitrust rules should be grounded in microeconomic analysis. The new approach can be traced directly to the contestation of the previously prevailing structure, conduct, performance (S-C-P) paradigm of analysis associated with Harvard University 36 by its intellectual antagonists centered at the University of Chicago who attacked both the logic and the empirical basis underlying the S-C-P approach. While there has been some rethinking of the usually very simple Chicago reasoning about competition policy, the approach that now dominates the academy and the courts remains utterly different from that prevailing prior to 1974. The Supreme Court unambiguously embraced this understanding in 1977 in its GTE Sylvania decision37 and began implementing it by overruling or limiting an array of precedents from earlier periods.38 In 1979 the Court, in its BMI decision,39 refused to apply the per se rule against horizontal price-fixing to a copyright society. In 1984, in Jefferson Parish,40 the Court restricted the scope of the per se rule against tying arrangements by reinvigorating a requirement of market power. In 1985 it limited the per se rule against concerted refusals to deal in Northwest Wholesale,41 and, in State Oil Co. v. Khan in 1997, it eliminated the per se rule that condemned vertical maximum-price-fixing agreements.42 Finally, in 2007 the Court overturned a ninety-six-year precedent by ending the per se rule outlawing vertical minimum price agreements in its Leegin decision.43 This new focus on efficiency as the core of the antitrust laws has come to be known as the “antitrust revolution.” Even in the wake of the “revolution,” American antitrust law has only gradually accepted efficiency as a justification for an otherwise prohibited merger. And that acceptance was at first theoretical, manifested only in the antitrust enforcement guidelines of the Department of Justice (and later of the Department and of the Federal Trade Commission). The current Horizontal Merger Guidelines, originated in 1982 in the Reagan administration, were subsequently revised three times in their entirety, and the provisions dealing with efficiencies were separately revised during the Clinton administration.44 Even on this level, the recognition of an efficiency justification emerged slowly, with each iteration conceding more to efficiency. The courts

chapter one

8

have since begun to acknowledge the legitimacy of efficiencies as a merger defense or justification, but they remain wary of actually applying the defense in their decisions. Developments in Europe. The Sherman Act was very much a product of its immediate environment. This truth is highlighted by the completely different experience of other capitalist countries. In most countries the freedom of contracting recognized by the modern state was interpreted to privilege market participants to act as they chose for an additional half century and more. Only much later would policies be put into place to balance that freedom against the losses of others. Although public authority in most states could interfere with firm activity where it seemed to threaten the general welfare through “abuse,” there was no widely accepted agreement about the constituents of abuse. Most specifically, until after the Second World War a prime manifestation of economic freedom was the formation of cartels, which governments typically did not oppose. Indeed, officials were often attracted by the ability of the cartels to stabilize prices in times of economic flux.45 The differing policy histories of various European countries are at least partially explained by the division of the European continent into separate political units of limited geographic scope. In the United States the development of scale economies in production and falling of transportation costs fostered the development of large enterprises marketing to large demands unimpeded by trade barriers. Europe, of course, experienced the benefit of the same technologies, although from the late nineteenth century, a growing protectionist movement in many European nations impeded trade expansion.46 Even so, world trade overall continued to grow throughout the pre– World War I period. During the interwar period large-scale production by European industries was impeded by protectionist policies that were sometimes exacerbated by currency instability. Again, scale economies in production were sometimes sought through monopoly policies at home and marginal-cost pricing abroad. These policies often rewarded producers at the expense of consumers in home markets, and tariffs or other trade barriers often impeded marginal-cost sales abroad.47 Both factors impeded the full exploitation of scale economies. The post– World War II economic history of Germany is particularly noteworthy because it played the largest role in developing postwar European economic policy. Cartels featured prominently in Nazi economic policies. Government action fostered intraindustry collusion that served as an element of state control of the national economy. Unsurprisingly, German

american and european perspectives on antitrust

9

postwar policy developed as a sharp reaction to the Nazis, and, although the occupying forces had insisted on decartelization laws and the introduction of some competition law before the end of occupation, the model chosen was largely indigenous. Ludwig Erhard, who served first as minister of economics and then as Christian Democratic chancellor, elaborated a “social market economy” that embraced market capitalism and a restricted role for state ownership but in an explicit context of a highly developed welfare state and strong trade unions. Erhard’s competition-based “economic constitution” was an integral component of a larger conception of the state and society that differed sharply from prevailing views across the Atlantic. Its apparent success in reviving the German economy was such that the rival Social Democratic Party pursued quite similar policies when it came to power under Willy Brandt in 1969.48 Erhard’s thinking was strongly influenced by a vision of society incubated subversively during the Nazi period called ordoliberalism (Ordoliberalismus). The doctrine was developed by Walter Eucken and his colleagues in what became known as the Freiburg School.49 As it was elaborated after the war, ordoliberalism seems best considered a general theory of political economy rather than a view of economic policy, and still less of competition policy alone.50 It continues to command considerable attention as the major strand of wholly European thought on competition policy, and, to an extent that varied widely by country, it suggested an early postwar alternative approach to the ad hoc administrative intervention in firm behavior that prevailed in most of Europe at the time. Ordoliberalism assumes not only that oligopoly leads to cartelization and monopoly, but also that concentrated economic power presages concentrated political power and ultimately the breakdown of liberal society.51 Therefore all deviations from price taking 52 in product markets threaten the liberal state to some degree. Ordoliberals championed small- and mediumsized businesses not only from a concern for the diffusion of economic power but also on grounds of social justice.53 Such concerns often trumped a concern for economic efficiency.54 The German competition law, finally adopted in 1967 after protracted legislative consideration, combined elements of ordoliberal thinking, some previous German practice, and some features of American antitrust.55 Elsewhere in Western Europe, the main political groupings were, as in Germany, typically the Christian-oriented moderate right with a historically ambivalent attitude toward both capitalism and competition opposed by a left bloc sympathetic to state control and ownership of industry and

chapter one

10

frequently dominated by organized labor. A doctrine of sharply limited government of the kind Republicans and most Democrats in the United States usually espoused was typically championed only by minor parties. In the early postwar years, most European states viewed their national firms pragmatically as instruments for reconstruction and growth. Governments (other than Germany) were suggestible but had few ideas that would be recognized as competition policy. Ordoliberal viewpoints became influential in Europe because they were associated with Germany’s economic success and because they largely filled a vacuum— they were unopposed by other well-developed European views about the same policy issues. Most influential Europeans simply assumed that the US situation was so different from the one they faced as to nullify any direct transatlantic learning.56 Abuse of Dominance versus Monopolization. Ordoliberalism can be contrasted with academic views across the Atlantic in the development of specific competition policy doctrines. For example, the 1930s saw a flowering of US thinking about imperfect competition and policies to deal with it. Such scholars as Edward Mason and John Maurice Clark, who coined the term workable competition, accepted the inevitability of market power and attempted to conceptualize how intervention should curb its excesses. The early US structural approach in economics saw antitrust as preventing market concentration beyond the level necessary for efficiency.57 In sharp contrast, ordoliberalism rejected such trade-offs as slippery slopes. When the Treaty of Rome58 was devised as the founding document of what would ultimately become the European Union, the Germans insisted that it include provisions to promote competition, and the remarkable rebounding of the German economy over the previous decade helped generate deference both to that insistence and to German preferences about the content of the law. Articles 101 and 102 of the Treaty on the Functioning of the European Union (TFEU) (in earlier numerations, Articles 85 and 86 and Articles 81 and 82 of the governing treaty) bear strong superficial resemblance to Sections 1 and 2 of the Sherman Act.59 Section 1 and Article 101 appear to be quite similar, but while the parallel between Section 2 and Article 102 cannot be missed, the difference is equally striking. Instead of what appears to be the condemnation of attempts to increase market power in the US law, the European document directs special attention to all “abusive” behavior of a class of “dominant” firms (which in German national law may be ones with a market share as low at 30 percent, although the usual threshold at the EU level is generally 40 to 50 percent).60 Many writers61 trace the disjunctive way that “dominant” firms are treated in Article 102 directly to ordoliberalism, and the article closely par-

american and european perspectives on antitrust

11

allels German domestic legislation. Because all market power is to be regretted, such firms should act as much as possible as if they do not possess it. While all exercise of market power in Europe cannot be curbed by either selfrestraint or public intervention, such exercise is prima facie suspect in the ordoliberal tradition, and therefore “abuse of a dominant position” in Europe encompasses much more than attempts to monopolize in the United States. While the charging of monopoly prices, or the earning of monopoly profits has usually not been regarded to violate US antitrust laws, such power can bring official intervention in Europe. This European posture had broad resonance when the first law was developed and continues to hold substantial appeal. Beyond the assumptions of ordoliberalism, the approach conforms to a broad tradition of greater European than American concern for the rivals and commercial customers of relatively large firms. Moreover, that concern is magnified by a European reflex against disruptive change and in favor of “fairness” toward market participants. Giorgio Monti has stressed the latter feature of European competition policy, explicitly identifying it with “distributive justice.”62 Others have stressed threats to the “economic freedom” of all victims of powerful firms.63 The welfare of competitors has played virtually no role in US antitrust for many years, but that has not always been the case. In fact, at the time of the Treaty of Rome in 1957, the purposes underlying American antitrust law were quite unclear. While it was obvious that antitrust law was about preserving competition, there was no general agreement on the ultimate meaning of competition or just what goals its preservation were supposed to be serving. Just thirteen years before the Rome Treaty, Judge Learned Hand had written in his Alcoa opinion that the ideal organization of industry under the Sherman Act involved small producers, regardless of the economic costs of such an arrangement,64 and that the Sherman Act incorporated various noneconomic goals, such as the fostering of initiative, self-reliance, and other social and moral goals.65 Nevertheless, Judge Hand’s Alcoa opinion was ultimately ambiguous. He also embraced efficiency as a justification for behavior that would otherwise be condemned as monopolization,66 and although this position appeared to contradict his model of the ideal industrial system, it can be read as consistent with the kind of trade-off between concentration and efficiency that was developing in the economic literature on workable competition.67 Thus as the Europeans launched competition policy as an element of their unification project, the foundations of modern US monopolization law were still in a state of development.68 One year earlier, the Court had

chapter one

12

ordered the DuPont Company to divest its holdings in General Motors,69 a decision that some contemporary observers explained as furthering an underlying political purpose of antitrust: that of limiting private power in the interests of democracy70— a goal entirely consistent with ordoliberalism. The Court’s 1962 decision in Brown Shoe71 seemed to reaffirm Judge Hand’s assertion that the ideal organization of industry involved small firms, and that efficiency considerations should be subordinated to that organizational goal. During that same decade, its subsequent merger decisions exhibited hostility to horizontal expansion, even those that did not significantly affect market concentration.72 These decisions could be read as American support for some of the goals of ordoliberalism. Europeans could also relate to much American opinion that regarded economic fairness and the protection of small business as embraced within antitrust’s broad purposes. The prohibition of price discrimination, which often disadvantaged small business and had been banned on those grounds in the Robinson-Patman Act of 1936, was widely seen as a core US antitrust principle. Writing in 1959, Harvard professors Donald Turner and Carl Kaysen accepted the many purposes incorporated in US antitrust law. They argued, however, that efficiency was the dominant purpose and thereby attempted to reconcile various legal and social concerns with those highlighted by economic analysis.73 As noted, the conduct-structure-performance paradigm was then widely accepted as approximating the purely economic dimension of the law. Under this paradigm, increases in concentration are suspect because they contribute to recognition of mutual dependence by market participants and tacit collusion; they therefore need explicit (static) efficiency justification. Consistent with this view, the Kaysen and Turner text boldly advocated horizontal divestiture as a remedy for excess market power on economic grounds alone. This policy direction gained only modest political and legal traction, but a concern for preventing greater concentration found wide adherence, a position that again echoed ordoliberal predispositions. Yet despite the ordoliberal antipathy to concentration, Europe did not adopt its Merger Regulation until 1989.74 When the Treaty of Rome was adopted in 1957, the antitrust policies adopted bore strong resemblance to the US policies of that period. Those US policies exhibited a number of resemblances to the ordoliberal vision in their hostility to concentration and in Judge Hand’s articulation of the ideal organization of industry. During the same period US antitrust policy also incorporated “fairness” concerns, including protecting small business firms from larger and often more efficient rivals. This fairness strand underlay the Robinson-Patman Act that was being vigorously enforced by the FTC throughout the 1950s and 1960s. These latter concerns still pervade EU

american and european perspectives on antitrust

13

competition policy. Article 102(c) forbids price discrimination generating secondary-line effects, policies remarkably similar to those of the RobinsonPatman Act. And protection of rivals from aggressive competition of dominant firms has been a major use of Article 102. Protection of rivals has also been evidenced in the administration of the Merger Regulation. The Overriding Goal of Economic Integration in Europe. The principal objective

of the Treaty of Rome, the breaking down of national barriers, loomed large in the first important European antitrust decision. The European Commission (EC) ruled in 1966 that a German manufacturer (Grundig) could not lawfully confer absolute territorial protection on its French distributor (Consten).75 That is, Grundig could not contractually obligate itself to bar its distributors, located in other nations, from selling in France, in competition with Consten. When the case was appealed to the Court of Justice, the German government argued that absolute territorial protection would actually enhance competition because that would give Consten a greater incentive to promote Grundig-branded products as they competed against other brands. The European Court of Justice (ECJ) rejected this argument and affirmed the Commission. Nevertheless, this “Chicago”-style argument gained traction a decade later in the United States and still later in Europe. The French distributor’s assigned territory in the Grundig-Consten case was all of France, and this appeared to impose market restraints that replicated state political boundaries. These boundaries challenged the prime economic objective of the new common market: the stimulation of trade between and among the member states. Grundig set the stage for the community treatment of vertical restraints for many years and was not remarkable by US standards of the time. More broadly, although there were substantial differences between US and EC antitrust laws for the first fifteen years after the Treaty of Rome, they were minor by comparison with the huge gap that began opening in the later 1970s. During the period from the beginning of the European Economic Community (EEC) until the reforms of 2004, the Directorate-General for Competition (DG Comp)76 devoted much of its attention to vetting vertical agreements and disapproving those that involved exclusive dealing or allocating exclusive territories. It moved cautiously on horizontal agreements because a frontal attack on cartels was still regarded as inappropriate in many parts of Europe. Nevertheless, all horizontal as well as vertical agreements had to be reported to Brussels, and this provided DG Comp with information about the need for policy development. In contrast to the delay on merger control until 1989, the Commission moved on dominant firm “abuse” beginning in the 1960s. Nevertheless, both the somewhat hesitant action on abuse

chapter one

14

and the delay on merger control are consistent with an overriding concern with increasing Europe’s productive efficiency.77 Pinar Akman has examined a large body of travaux propriétaire and other documents. She argues strongly that horizontal competitors were not the concern of the drafters of Article 102 but only their customers, who were to be protected from exploitative abuse. Nevertheless, both the final language and the ultimate legal enforcement have largely been interpreted as reflecting the opposite concern, a development that might itself suggest the influence of ordoliberal thought.78 Limited Change on Vertical Arrangements in the EU. The antitrust revolution

resulted in a sharp change in the attitude taken by US courts toward most vertical restraints. Although control of such restraints was once an important part of US antitrust law, most such restraints are now treated as benign. Europe changed much more gradually. The European Commission released a green paper on vertical restraints in 1997,79 clearly influenced by the changes in the United States. That paper appeared to presage a move toward a purely economic approach based on consumer welfare. This was followed in 1999 by a revised vertical-restraints regulation80 that moved in that direction but fell short of the green paper’s approach. In 2010 the Commission issued a new vertical restraints regulation largely following the provisions of its 1999 regulation, broadly exempting vertical agreements from Article 101(1) of the Treaty so long as the market shares of both parties do not exceed 30 percent and so long as the agreements do not contain certain so-called hardcore restrictions, such as minimum resale price restraints.81 The Commission moved to change its approach to horizontal agreements in 2001 with guidelines that paralleled those for vertical restraints. They too suggested the primacy of consumer welfare as viewed by economists, as did the revised Merger Regulation of 2004. This was followed by a parallel attempt to change the approach to unilateral conduct reflected in a working paper in 2005 and a written guidance in 2008.82 Policy change in the European Union is constrained by the Court of Justice, which ultimately oversees the Commission’s actions. The ECJ has taken an approach to competition law that relies on the use of behavioral categories that are treated as innately unlawful. This approach is similar to that of the US courts, which have historically created several categories of behavior deemed per se illegal, although in the recent past, US courts have sharply reduced the scope of such categories and substituted a Rule-of-Reason analysis instead. In recent years the Commission has attempted to substitute effects-based analyses for form-based analyses in its enforcement of the Article 102 prohibitions, but the General Court (called the Court of First Instance from 1989 to 2009; we will use the current term throughout) and

american and european perspectives on antitrust

15

the ECJ have not been receptive to such a shift. The courts have repeatedly combined their approval of the Commission’s effects-based analyses with the statement that the latter was unnecessary because the conduct involved fell into a category where it could be condemned without the showing of any market effects. In so doing, the courts are clearly retarding EU law from following the change made in the United States by the “antitrust revolution.” EU competition policy explicitly addressed mergers in the Merger Regulation of 1989 after a quarter century of debate about whether such regulation was necessary or appropriate.83 The European Union historically appears to have been even more reluctant to accept an efficiency defense than the United States. In fact, in the years before the Merger Regulation was revised in 2004, critics claimed that the European Commission was blocking mergers because they promoted efficiency. Instead of being a defense, efficiency was treated as an offense, a situation that had prevailed earlier in the United States.84 The Commission considered, but ultimately rejected, a proposal to cast its revised Merger Regulation in the language of the Clayton Act that examines mergers to see if they “substantially lessen competition or tend to create a monopoly.” Some believed that this would have signaled the acceptance of an efficiencies defense. Nevertheless, EC officials at the time asserted their acceptance of an efficiencies defense and their belief that European merger law was converging with that of the United States despite the then-recent decision of the Commission and the General Court in the GE/ Honeywell merger case85 appearing to show the opposite. Unilateral firm conduct presents the greatest continuing transatlantic gap in law and policy. The different approaches to the market power of the same firm, Microsoft, illustrate some of them. For example, the EU Microsoft 86 decision was the culmination of case law developments that had imaginatively applied the so-called essential facilities doctrine to intellectual property (principally copyrights), thereby requiring the rights holder to share access to its intellectual property with its competitors for a reasonable compensation. It may be significant in this regard that Walter Eucken, the most important figure of ordoliberalism, opposed intellectual property in the form of patents, which he saw as a “threat to open markets” and a tool for monopoly.87 This view of patents may still have more resonance in Europe than in the United States. The European Commission required Microsoft to make the protocols by which its server software communicates between servers available to its competitors. The Commission also ruled that Microsoft’s integration of its media player software into its operating system constituted an illegal tie. Both GE/ Honeywell and Microsoft can be contrasted with US views. In the

chapter one

16

US Microsoft antitrust case,88 the DC Circuit had created an exception to the per se rules governing tying arrangements that involved the integration of new functionalities into operating systems. On the server software issue, it appears that US courts would be reluctant to deprive an innovator of the economic fruits of its innovation, as the EU decisions did. Moreover, the US Supreme Court, in another antitrust case, has questioned the continuing validity of the essential facilities doctrine.89 Looking more broadly at innovation-intensive industries that are often called the “new economy” reveals great transatlantic differences. These industries, which are frequently typified by the interaction of intellectual property and network effects, may sometimes generate serial monopolists as rivals strive to create new technologies that will undermine the dominant position of the leading firm. Of course, the danger is that one firm may succeed in achieving a long-term monopoly. While antitrust law in the United States has struggled with this issue to balance the fostering of innovation with the tolerance of too much market power, the European Union has refused any special consideration for new economy firms. Another substantial transatlantic difference can be seen in the treatment of price discrimination. Although the Robinson-Patman Act has forbidden certain types of price discrimination since 1936, the US government no longer enforces it, and private plaintiffs are rarely successful. Since the antitrust revolution of the 1970s, with its emphasis on efficiency, most antitrust observers no longer view price discrimination as socially detrimental but as a competitive tool.90 In Europe, however, a prohibition against price discrimination by a dominant firm is incorporated into Article 102(c) of the Treaty, to protect distributor and producer-buyers from disadvantage vis-àvis their rivals, a rationale that is substantially identical to that underlying the Robinson-Patman Act. Moreover, price discrimination also has been treated as within the prohibitory scope of the general language of 102, where it has been used to protect the rivals of a dominant firm from “selective” price-cutting, although a recent decision of the Court of Justice suggests that the Court may be changing its position and moving toward a position on this issue that is closer to the US position.91 Bundled rebates are now being confronted on both sides of the Atlantic, and policy is in flux. The US Third Circuit’s decision in the 3M case drew widespread attention to this matter in 2003.92 That decision, unfavorable to bundled rebates, has been criticized by the Antitrust Modernization Commission and rejected by at least one other circuit.93 The European Commission has taken an aggressive stance against the use of both single-product (or loyalty) and bundled rebates by dominant firms and has developed a conceptual framework for distinguishing between benign and unlawful rebating,

american and european perspectives on antitrust

17

as explained in chapters 7 and 8. The European Commission’s approach appears to have influenced the Federal Trade Commission, which opened a case against Intel shortly after the European Commission rejected that company’s use of loyalty rebates. The FTC case was settled shortly thereafter, with Intel accepting a consent order. Existing European law seems to support the European Commission, but the FTC action appears to be new because much Sherman Act jurisprudence conflicts with many elements of the complaint and the settlement.94 European law condemns a broader swath of conduct on all of the issues just reviewed. This consistently stricter stance across several unrelated areas suggests a profound difference in approach: European antitrust authorities are directing their law at objectives that differ from maintaining competition in the service of efficiency, the only goal now generally recognized in the United States. Moreover, that difference cannot be largely explained by the recognized European objective of preventing private contractual arrangements from creating trade barriers that mimic the political boundaries of the member states. Instead, the pull of maintaining rivalry or serving other objectives beyond directly promoting consumer welfare continues to exercise considerable influence. Institutions Doctrinal differences do not fully explain the differences in transatlantic policy outcomes. Differing institutions and interests also matter greatly. The Institutional Structure of US Competition Policy. The United States stands apart not just from Europe but from nearly all other jurisdictions in the relative insulation of the competition policy system from domestic politics. Since the Clayton Act and the Federal Trade Commission Act before World War I, Congress has acted only rarely to effect substantial change in competition policy,95 which remains firmly grounded in the Sherman Act. The executive can influence some policy by its choice of the assistant attorney general for antitrust and the chairman of the Federal Trade Commission from among the five commissioners and to fill vacancies; the commissioners serve seven-year terms.96 Even though there are well-documented instances of executive— and congressional— interference with various cases,97 the effects are typically marginal. Precedent, professionalism, and judicial review protect the entire system in such a way that one study concluded: “there is little credible evidence that politics drives enforcement.”98 This has left the courts, the bar, the universities, and the enforcement agencies unusual latitude to rethink

chapter one

18

and revise policy continually as evidence has accumulated and conceptual understanding has evolved. The dramatic changes of the 1970s— those incorporating a much greater emphasis on economic effects the Chicago School urges— neither resulted from nor generated any federal legislation. Consistency in merger enforcement is facilitated by the Merger Guidelines, and these guidelines have been followed during both Republican and Democratic administrations.99 Consistency throughout the entire antitrust spectrum is now fostered by the law’s focus on consumer welfare100 as its overarching goal and by the use of explicit economic analysis to justify decisions. Administrations can and do shift policy somewhat by limited personnel changes at the top of the agencies, but these are usually minor changes considered within the entire antitrust system. The reach of this system is much larger in the United States than elsewhere due to uniquely incentivized private litigation: treble damages for violation of the federal antitrust laws.101 In fact, private cases filed annually in US District Courts ranged from 83 to 96 percent of the total from 1975 to 2003,102 an astounding contrast with the situation in Europe, where private litigation began to play a nonnegligible role only after that period ended. This explains another big difference between the United States and the European Union: American courts are very conscious of how their decisions will guide private litigants and assume a responsibility to make that guidance as clear as possible. Despite the stability in formal law and the absence of any marked change in the structure or approach of the agencies, there has nonetheless been one quite substantial US change that can be regarded as institutional: the great increase in the role of economics and economists relative to the law and lawyers in both public and private competition cases. Writing in the early 1970s, Richard Posner noted that the Antitrust Division’s “economists today are handmaidens to the lawyers and rather neglected ones at that.”103 The lawyers called them “statisticians” because their principal role was to compile data at the lawyers’ behest. This picture changed dramatically over just a few years. As the courts began to expect antitrust cases to be defensible in economic terms, economists came to play key roles in developing approaches to litigation from the very outset. This is turn led to a huge increase in outside consultation and in specialized economic consulting firms that served both the government and private clients. Competition Policy Institutions in Europe. The European institutional picture

is more complex and has changed more over time. David Gerber has done pioneering research on the variety and extent of European competition law both before and after the Second World War. He notes several  initiatives

american and european perspectives on antitrust

19

beginning as early as the time the Sherman Act was passed. These were somewhat consumer welfare-oriented competition proposals based on administrative control— most importantly in Austria, Germany, and Norway— but they failed to gain either adoption or effective implementation104 in the turbulent political and economic European landscape of the first decades of the twentieth century. Instead, governments simply intervened intermittently in business to serve a state end or to combat “abuse.” These measures were almost everywhere simply regarded as elements of discretionary economic policy and subject to legal review only on grounds that the actions exceeded authority.105 A key element of the ordoliberal approach to competition policy was enforcement by an independent agency enforcing clearly stated principles, but German big business wanted discretionary administrative intervention instead of clear prohibitions. The innovators won, and the Bundeskartelamt became Europe’s model competition agency. While there was a role for some political intervention and its operation was administrative— the issuance of directives and fines— the new agency was to follow settled law, and its behavior was reviewable by regular courts. There has been much postwar development of national competition laws of other EU states in these directions too, in parallel with the unfolding of competition law at the supranational level. The competition provisions of the Treaty of Rome were not at first taken seriously as enforceable law by some members, but with strong German support they became increasingly understood to be binding. The original investigation and enforcement mechanism of the EU— what became DG Competition— combined elements of the administrative tradition with policy innovation and judicial review. The Commission developed the law (which was then ratified by political authorities), but it was closely overseen by the European Court of Justice and intermediated after 1991 by the General Court.106 Brussels foresaw enormous strain in the aftermath of the collapse of the Soviet Union, the anticipation of many more community members, and a blizzard of potential activity relating to the shift of many economies from state ownership to private enterprise. The ultimate solution, first laid out in a white paper of 1999 and put in into effect in 2004, combined increased central authority with devolution of enforcement. Prior to 2004, while all cooperative exemptions under Article 101(3) had to be cleared at the EU level, purely national competition issues were left to the member states, and there was relatively little interaction between EU and state mechanisms. Subsequently, the member states could grant their own group exemptions under 101(3), but all competition matters with any community dimension would

chapter one

20

be judged using EU law. The European Competition Network of national competition authorities was established largely to insure uniformity. This set of innovations was not controversial in most member states, but it did generate stiff opposition from Germany, which had the most highly developed national laws and enforcement.107 The anticipated onrush of economic activity as the community expanded also increased interest in encouraging the growth of private litigation, which had always been possible but had been discouraged until 2004 by the ambiguous power relations between EU and national law. Following extensive consideration,108 the European Commission proposed a directive in June of 2013109 intended to facilitate private action by clarifying the path to redress following the discovery of wrongdoing by the Commission. Private parties will be able to bring action in national courts using procedures that will allow for consistent treatment of complainants across the European Union. Much private demand for compensation stems from victims of cartel action, and rules have been crafted that allow access to the information necessary for private action while protecting leniency toward firms that cooperate with authorities to expose the cartel. The purpose of the EU initiative is only compensation for harm. Punishment and deterrence remain the province of public authority, so while private action may grow substantially, its role will continue to differ dramatically from that in the United States. Overall, despite institutional and doctrinal reform, not only does European enforcement embrace a consumer welfare standard less consistently than does the United States, but it is also more vulnerable to political influence. The ultimate decision of whether to pursue a violation is taken not just by the competition commissioner but also by the twenty-eight commissioners (one for each member state) as a group. This can inject a political element into those decisions. Nevertheless, commentators believe that the Commission is less afflicted by political influence now than it was as late as the 1990s. The increased role of economic analysis in Commission activity has undoubtedly provided constraint, as it has in the United States. Economic reasoning and economic analysis have increased rapidly in importance in the European Union, although it appears still to play a smaller role than in the United States. The ratio of economic to legal professionals in the Competition Directorate grew from one to seven in the early 1990s to one to two by 2006.110 Total economic expertise still lags significantly behind that of the US enforcement agencies. In 2005 there were only twenty PhD economists working for DG Competition,111 while the estimated total for the Antitrust Division and the Federal Trade Commission was close to 120.112

american and european perspectives on antitrust

21

The Commission appears to have moved economic reasoning to the forefront. Shortly after the new millennium, the Commission was reversed in three merger cases in which the General Court sharply criticized it for deficient economic reasoning.113 Currently, the Commission employs economic evidence (effects approach) in merger cases and cases brought under Article 101 (agreements). In its Article 102 (single-firm conduct) cases, an emphasis on effects is now being retarded by the courts, which are guided by previous case law. This appears somewhat paradoxical in light of the Court’s earlier criticism. Interests Interests may loom in competition policy both within jurisdictions and among them. The European Union presents additional complexity because most politics remains at the national state level, while in competition policy as in many other spheres the most important decisions are taken in Brussels.114 Distinguishing more general political currents from special interests cannot be done with precision, but it is clear that interests have sometimes had a marked impact on competition policy developments on both sides of the Atlantic. Despite the general insulation of the US competition policy system, it was born of populist anger. Decades later, Congress clearly tried to assist small business with little concern for efficiency as it legislated to slow the advance of the “chain store age” with the passage of the Robinson-Patman Act in 1936 during the depths of the Depression. The Miller-Tydings Act (1937)115 and McGuire Act of 1952116 were designed to allow individual states to circumvent the Dr. Miles (1911)117 prohibition on resale price maintenance, again to help small business. The economic nonviability of such prohibition at the state level confronted the political impossibility of curbing efficient distribution with national legislation; Congress banned state carve-outs in 1975,118 just in time for the “antitrust revolution.” Another major interest group, organized labor, has seldom taken much of position on antitrust,119 and, if anything, has tended to favor market power by firms as conducive of its own rent seeking. In Europe, the influence of subnational special interests only occasionally surfaces. In the struggle over the original German competition legislation, the large German firms fought against legal embodiment of ordoliberalinspired “abuse” in black-letter law, preferring administrative discretion. While these forces were largely defeated in Germany, in other states, and at the community level, another Germany-centered tension inspires political conflict today.

chapter one

22

The German Mittelstand, medium-sized manufacturing and distribution firms with great prestige as elements of German prosperity and growth, have no political counterpart in the United States, and they strongly oppose a more Chicago-oriented change in the European Union. Most of the left share these views: stronger unions are identified with national, rather than EU, and still less, US firms. There is a large literature in Europe that is almost unknown in the United States on the threat of globalization to “Rhenish capitalism,”120 and this extends to efficiency-oriented competition policy, which is seen as an element of globalization.121 A wholesale abandonment of a concern for small- and medium-sized business in a shift to an exclusive concern with consumer welfare has attracted considerable European interest and opposition. This sentiment, although perhaps diminishing, clearly retards policy convergence with the United States.122 Transnational business has generally favored consistency and convergence in competition policy. In particular, simplification and speed in merger approval has been supported. In sharp contrast, however, non-US business in particular has strongly opposed information sharing across the Atlantic that could lead to US penalties in cartel cases.123 Much of European business has also opposed the Commission’s drive to increase private litigation.124 There is also evidence of special interests when the various states confront the European Union. In the early 1990s differing interests among the member states plagued the Commission’s decision in the Aérospatiale merger decision,125 although the Commission successfully resisted pressure to allow the merger.126 At same time, the Commission’s disapproval of the merger also appears to reflect a rejection of efficiency considerations. Special interests also affect policy toward the outside world. The relatively trivial protection of exporter collusion sanctioned by the US WebbPomerene Act127 is often cited. More recent developments are also much more important. While Chicago arguments and evidence may have provided the greatest impetus for the “antitrust revolution,” many commentators also note a contemporaneous international competitive reality: the fear that stricter antitrust at home would weaken US firms in international competition at a time of great fear about “competitiveness,”128 particularly against the Japanese. This argument parallels the reluctance of much of Europe to embrace the “full competition” ordoliberalism counseled in the face of the “American Challenge” US multinational enterprises posed in the early postwar period.129 Several transatlantic conflicts have generated American suspicion that European authorities sometimes favor EU firms.130 The European Union has adamantly denied such motivation, and some direct observation suggests that overt political pressure is unlikely to have often affected EU decisions.131

american and european perspectives on antitrust

23

Economic models of trade suggest the importance of rent for national welfare: if foreigners pay more for something than it costs to make, the home country gains more than is suggested by simple comparative advantage. This truth looms very large for the United States because so many of its major exports— from entertainment to computer technology to pharmaceuticals— rest on intellectual property. Although multinational US corporate sellers greatly reduce the US Treasury’s immediate take of foreign corporate profits by the use of tax havens, most of the stockholders are nationals; this creates an immediate gain for their fellow citizens through the taxation of dividends and capital gains. If national rents from intellectual property are higher for the United States— or even if they are simply thought to be higher on both sides of the Atlantic— Europe will have a greater incentive to squeeze those profits than does the United States, even with similar concerns about retarding future innovation, because excess profits accrue mainly to foreigners. The European Union faces other parochial incentives that are more problematic in terms of its economic welfare. There is a history of favoring “national champions” in many EU states, and some firms have achieved that status at the EU level. These firms have historically been singled out for various kinds of favorable treatment on grounds that they may generate economic benefits that the firm itself cannot capture. These firms are often not global industry leaders, and their welfare may sometimes be advanced by protecting them from the firms that are. The leaders are very often US firms. Overall, the European Union appears to face greater impediments to the consistent pursuit of nonparochial competition policy than does the United States, but the magnitude of the problem is difficult to assess. The title of this book suggests its emphasis. Differences between the United States and the European Union in competition policy will dominate the following discussion because the authors believe they are typically underemphasized. This approach leads us to ignore or pay scant attention to many important competition policy issues. For example, our consideration of cartel policy relates almost exclusively to differences in procedures and penalties and not to the extraordinary growth in transatlantic (and indeed global) cooperation in this critical policy area. The last chapter provides some general discussion of both convergence and cooperation in competition policy, but the book in between concentrates on areas where differences need to be explored.

2

Welfare, Monopolization, Dominance, and Judicial Review This chapter explores the meanings that may attach to “welfare” as the primary goal of competition policy and how economic welfare and other objectives relate to two of the most important legal concepts in competition policy on either side of the Atlantic: monopolization in the United States and dominance in Europe. The chapter concludes with a discussion of the judicial review in the two systems. If “Welfare” Is the Main Goal of Competition Policy, What Does It Mean? While Europe has engaged in a continuing struggle between some recognizable concept of economic welfare and other social objectives, another tension can be seen across the Atlantic. Since the “antitrust revolution” of the 1970s, the US courts have repeatedly referred to the furtherance of efficiency as the ultimate goal of the antitrust laws.1 In the language of economists, efficiency describes the use of resources in a way that maximizes aggregate social welfare. US courts have repeatedly identified increasing efficiency with the furtherance of consumer welfare.2 The phrase consumer welfare, however, is ambiguous. If it means furthering the welfare of consumers, qua consumers, then it may sometimes produce a different evaluation from a social welfare standard in particular cases. Thus as a general standard it would pro-

chapter two

26

duce aggregate results that would be similar, but not identical, to those a social welfare standard would produce. The origin of the consumer welfare phrase is traceable to Robert Bork’s publication of Antitrust Paradox in 1978.3 In that book, Bork used consumer welfare as synonymous with efficiency and aggregate welfare by equating “consumers” with everyone, including both consumers, as ordinarily defined, and producers. Bork’s use of the phrase was cited and endorsed by Chief Justice Warren Burger in his Sonotone opinion,4 and thereafter the federal courts used the phrase to describe the ultimate purpose of the antitrust laws. This history fails to clarify the meaning of consumer welfare. In the European Union, the consumer standard completely dominates the aggregate welfare standard.5 The dispute can be demonstrated with figures 2.1 and 2.2. The demand curve shows how much successive units of output are valued by those in the community with the highest valuation. A particular price is associated with a certain level of output. In figure 2.1 the payment for amount p1, x1 is the product of those measures, while the value to all consumers together is greater because it includes the triangle above the payment rectangle: that amount is consumer surplus. One standard for antitrust simply asks whether a market change increases or decreases that consumer surplus. For example, a series of mergers could shift the market from a group of pricetaking firms selling at price p1 (equal to marginal and average cost of c1) to a monopoly selling at p2. The outcome here is very simple: the merger redistributes wealth from consumers to producers,6 but consumers lose more than producers gain. They lose p1, p2, B, D, while the sellers gain only p1, p2, B, A. The area BAD is called “deadweight loss.” It reflects the inefficiency from consumers not facing the real resource cost of an additional unit

FIGURE 2.1.

merger to monopoly power.

welfare, monopolization, dominance, and Judicial review

FIGURE 2.2.

27

efficiency and distribution trade-offs.

of production (they purchase only x2 of the good instead of x1, the amount where their marginal valuation just matches marginal cost). Another case, illustrated by figure 2.2, suggests the possibility that consumers can gain even with a merger to monopoly if the merger also decreases cost. Specifically, assume that the monopoly realizes sharply decreased cost to the level of c3. The profit maximizing price would be lowered to p3 as marginal revenue, shown by the line MR, is equated to the new cost level. Thus in this case monopoly trumps the preservation of many rivalrous firms as an avenue for the maximization of consumer surplus. This illustrates Schumpeter’s famous “gale of creative destruction”7 and may be a particularly important case in an economy marked by radically new innovations in technology. Merger outcomes could combine features of the previous two cases. What if a merger moved cost from c1 to c2 but raised price from p1 to p2? Here consumers lose p1, p2, B, D, but producers gain far more: c2, p2, B, E. Many economists and lawyers (including Williamson, Bork, Easterbrook, and Carlton)8 have argued that a merger should be allowed if it raises the total of changes in producer surplus (here profits), consumer surplus, and deadweight loss, where the third factor is always negative.9 In the present case, the former is positive and the latter two are negative, but sum is strongly positive. Law professors often argue that the test of policy ought to be whether or not industry output increases.10 In the case just sketched, output in the market of concern contracts, but it expands elsewhere; the now-redundant resources are reallocated to other uses. Many commentators (including Lande11 and Pitofsky12) assert that by allowing prices to rise, regulators abandon the fundamental objective of antitrust: consumer welfare. Most

chapter two

28

contemporary economists, however, see competition policy instrumentally, as a means to economic efficiency and growth. They seek consistency for antitrust, not with its alleged history of serving consumers but with other measures of public policy that accept some capricious income redistribution on the path of economic improvement. Total welfare adherents point out that public capital improvements and publicly funded research and development expenditures do not match payers and beneficiaries in any exact way either, but that income distribution should be the province of fiscal and not competition policy. Nevertheless, a competition policy that appears to countenance direct redistribution from buyers to sellers meets sharp political resistance almost everywhere. Exceptions seem to arise in small export-oriented economies, including Canada and New Zealand, where it is apparently assumed that maximum efficiency should be the paramount objective and that monopoly prices are largely paid by foreigners and can be moderated if conditions change. There is little doubt that the original intent of US antitrust was closer to a consumer welfare standard. Steven Salop thoroughly documents the politics surrounding the passage of both the Sherman and Clayton Acts.13 In the European Union the language of Article 101 TFEU is unambiguous about the need for practices that might otherwise be anticompetitive to pass a “fair share” of gains on to customers. Most commentators have understood this standard to apply to Article 102 as well.14 But the US legal language is far more ambiguous and even allowed Bork a kind of verbal sleight of hand whereby “consumer welfare” was defined as total welfare and declared to be the goal of US antitrust. Joseph Farrell and Michael Katz15 have recently argued that the US law would not really serve either of the two standards even if one were chosen univocally. Firm behavior is generally not controlled; both process and consequences matter. Specifically, only those actions that damage competition ever get to the stage where one of the two welfare standards comes into play. For example, monopoly is generally tolerated when legally acquired on the general rationale of promoting innovation even in specific instances where such a prospect is far-fetched. Thus the consumer welfare criterion is not served. Similarly, no serious commentator has suggested that inefficient entry be banned under a total surplus standard even though such entry may well lower total welfare. Commentators make a broader point: even if maximization of total surplus were the actual goal of antitrust, this would not necessarily inform enforcement agency actions. For example, in the case of mergers, they give a number of possible reasons to justify a narrower agency focus on consumer welfare based on the provision of a

welfare, monopolization, dominance, and Judicial review

29

counterweight to firm interests in the choice of merger plans16 and by the representation of otherwise unrepresented consumers in bargaining, lobbying, and litigation. Gary Roberts and Steven Salop17 have argued that the gap between the consumer welfare and the total welfare standards may be narrowed by a consideration of the way economies actually function. First, if taxes on profits of the merged enterprise and the rise in the value of equity holdings of consumers in that enterprise are given any weight in the analysis, a negative outcome for consumers sometimes turns around. Another important factor concerns the development of the market over time and particularly the extent to which a merger may raise prices only until the cost advantages to the merging firms have diffused throughout the industry. Roberts and Salop argue that the estimation of efficiencies and their diffusion may be no more difficult for the authorities than the estimation of concentration and barriers to entry.18 This may be true, but the additional complications do not substitute for the others. Instead, they increase the range of plausible outcomes, perhaps diminishing the likelihood of international consensus about whether a merger should be approved. The early S-C-P approach to competition policy considered not just static but also dynamic efficiency. At the time, however, growth was seen to involve mainly the adoption of new techniques science and technology afforded,19 whereas a central contemporary policy question is what industrial structure is most likely to produce new technology in the first place. Dennis Carlton has argued that competition policy authorities seldom try to forecast their results beyond two years, and this contributes to a powerful case for the use of the total welfare standard. A merger that allowed for some price increase in the short run while permitting large fixed-cost saving would be more dynamically efficient because today’s fixed costs may well be tomorrow’s variable costs. This would be recognized in a consumer welfare analysis done over a sufficient period of time, but he regards that as unlikely. A disadvantage of the use of the total welfare standard is that immediate consumer losses might be sacrificed for illusory social gains because price movements in any direction immediately following a contested market change would be allowed.20 Monopolization and Abuse of Dominance The laws of the United States and the European Union follow a similar structure. The initial provision of both laws deals with concerted activity, followed by a provision dealing with unilateral behavior. Differences in the

chapter two

30

respective approaches to unilateral behavior account for much, although not all, of the policy divergences between the two jurisdictions. Monopolization. The US law treats unilateral behavior under Section 2 of

the Sherman Act. That section prohibits monopolization and attempts to monopolize as well as combinations and conspiracies to monopolize. The initial two offenses can be the work of a single firm. “Monopolization” and “monopolize” do not equate to the possession of a monopoly despite their etymological connection. Instead, they suggest that while the possession of some level of monopoly— meaning a high market share— is a critical element in monopolizing, something more is required. That extra element is the acquisition of monopoly through anticompetitive means or the use of monopoly power to strengthen or maintain a monopoly. The first and obvious point is that “monopoly” almost never means what the Greek term implies: a single seller. Judge Learned Hand in the Alcoa case of 194521 provided the often-quoted observation that the defendant’s 90 percent market share “is enough to constitute a monopoly; it is doubtful whether sixty or sixty-four percent would be enough; and certainly thirtythree percent is not.” Subsequently, various US circuits have set the standard of monopoly between 50 and 70 percent of a market that has been somehow defined.22 The US Supreme Court has made clear that in its view firms that are too small to threaten monopolization are incapable of engaging in anticompetitive conduct, at least within the contemplation of Section 2 the Sherman Act.23 Setting a minimum market share for monopoly is perhaps the easiest challenge; more difficult is the determination of what constitutes “monopolization.” The application of the concept has varied substantially over time: in the U.S. Steel case of 1920, the Supreme Court held that “mere size” or “unexerted power” is not illegal.24 This approach was challenged by the Alcoa decision of 1945, which held that simply building capacity ahead of demand— and not any attack on rival firms— could violate the Sherman Act if a firm had sufficient market share. The tobacco oligopoly was found guilty in 1946 of violating Section 2 by acting in unison through tacit collusion, again without aggressive behavior toward rivals, but no remedy was found.25 By the DuPont case in 1956,26 the Court was considering monopolization through the lens of “monopoly power,” which it defined as “the power to control prices or exclude competition.”27 In the Grinnell case of 1966, the Court declared: The offense of monopoly under § 2 of the Sherman Act has two elements: (1) the possession of monopoly power in the relevant mar-

welfare, monopolization, dominance, and Judicial review

31

ket and (2) the willful acquisition or maintenance of that power as distinguished from growth or development as a consequence of a superior product, business acumen, or historic accident.28

But what is “monopoly power?” Abba Lerner offered an early definition that focused attention on the common characteristic of sellers in imperfect competition: some power over price,29 a situation that exists whenever a firm faces a downward-sloping demand curve for any reason. Lerner’s measure relates price solely to the elasticity of demand; the index falls as demand elasticity increases.30 Unfortunately for its policy usefulness, the Lerner index allows for any level of firm profitability. Market shares do not map in any simple way to price inelasticity, which determines the relation of price to marginal cost. And marginal cost can bear any relation to average total cost. For example, in monopolistic competition firms face downward-sloping demand and profits are still zero by definition in the long run. And complete monopolies or colluding oligopolies are not always profitable, either. They can have monopoly power but still go out of business if fixed costs are high enough. In short, Lerner’s index focuses only on allocative efficiency and not on long-run profitability— and hence market viability.31 Market viability is a central concern of industrial organization economists and lawyers. Confusingly, competition policy economists frequently use both monopoly power and market power to describe a major object of their concern, supernormal profitability.32 The commonsense meaning of monopoly power suggests something stronger than market power, and this is the usage in US antitrust case law.33 But Einer Elhauge has pointed out that while the courts may effectively use the term monopoly power to mean a “significant” or “substantial” degree of market power, market power under Section 1 of the Sherman Act is “normally defined as not just any ability to raise prices above competitive levels but an ability to raise prices ‘substantially’ over those levels.”34 Nevertheless, while such distinctions are necessarily arbitrary, they do rest on a definite dimension of harm and suggest a direction for policy priority: greatest attention to high price-cost margins in large markets. Complications abound. As a simple example, holding firm output constant, decreasing the elasticity of demand that the firm faces35 increases profits but decreases deadweight loss, creating a tension between the consumer surplus and the total surplus measure of the welfare impact of that market change. But agency statements of the goals of US competition policy today profess a dedication to the consumer surplus standard,36 and this might seem to imply an antitrust policy that would search for intervention in markets where the prospect of reducing or avoiding the summed “excess” prof-

chapter two

32

its was greatest within the limits of agency budgets.37 The pursuit of this priority would be greatly complicated by judgments about, inter alia, the extent to which the differences between prices and costs did not reflect a return on previous sunk costs and the probable future path of profits and consumer surpluses in the absence of intervention. Such concerns are vital for enlightened policy in the parts of the economy subject to rapid technological change, as explored in chapter 9. Moreover, a focus on the relation of price to cost completely ignores the special problems of resource waste that sometimes surround sheltered market positions; these include a lack of incentive or information necessary for efficient operation and expenditures solely geared to keep or attain market power. Attempted monopolization is, in effect, the penumbra of the monopolization offense. It involves anticompetitive behavior by a firm that lacks the high degree of power the courts require for the monopolization offense, but whose power is substantial and threatening to grow to the level recognized by the courts as monopoly power. Justice Oliver Wendell Holmes, in his classic description of attempted monopolization in Swift & Co. v. United States,38 identified the elements of the offense as the defendant’s intent to monopolize and the dangerous probability that the defendant would succeed. The structure of the offense draws content from those two elements. The “dangerous probability” that the defendant would gain monopoly power (in the legal sense) presupposes that the defendant already possesses some significant market power. And the two elements together presuppose a third element: a relevant market over which the defendant is seeking monopolylike power. The Ninth Circuit at one time gave an expansionary construction to the offense,39 deemphasizing the classic elements of the offense and using it against what that court perceived as anticompetitive behavior by firms lacking market power.40 This was both a misconstruction of the antitrust laws and a failure of economic analysis, since firms without power cannot restrain the market. The Supreme Court ultimately repudiated the Ninth Circuit’s approach in Spectrum Sports, Inc. v. McQuillan41 in 1993 and reimposed on plaintiffs the need to establish a relevant market and the defendant’s power in that market. Abuse of Dominance. One knowledgeable observer has attributed the failure of most foreign jurisdictions to adopt the US standard of “monopolization” to its innate ambiguity.42 He rightly notes that they have overwhelmingly opted for some version of the somewhat comparable EU standard for single firm behavior: the “abuse” of a “dominant position.” An early European Court decision attempted to define a dominant position as “ a position of economic strength enjoyed by an undertaking which enables it to prevent

welfare, monopolization, dominance, and Judicial review

33

effective competition being maintained on the relevant market by affording it the power to behave to an appreciable extent independently of its competitors, its customers and ultimately of the consumers.”43 The text continues with verbiage that is no more amenable to economic reasoning. Another passage, however, suggests that market power in the US sense may be approximated by a “dominant position”: “the fact that an undertaking is compelled by the pressure of its competitors’ price reductions to lower its own prices is in general incompatible with that independent conduct which is the hallmark of a dominant position.”44 European courts have found dominance to exist with market shares as low at 40 percent. In fact, of course, even a much higher market share might not lead to a high level of market power if that power is checked by competition, as the second quotation above implicitly suggests it is. As in the United States, it appears that those deciding on cases to pursue in Europe will use market share only as suggestive of places to look for market power, with the latter as the central concern.45 Dominance is not itself an EU offense any more than the possession of a monopoly is itself an offense under American law. Dominance is important under EU law because only dominant firms are capable of “abusing” that dominance. Abuse consists of violating any of the several clauses in Article 102 directed at particular abuses; it can also consist of violating the general injunction against (undefined) abuses. Many might question whether treating “dominant position” and the “abuse” of that position as separate ideas presents a clearer competition policy path than “monopolization,” once “monopolization” is attached to some measure of market power and its enhancement. In fact, unless “dominance” is viewed in terms of market power, which it now seems to be, it is not at all clear what it does mean. But in contemporary US practice, market power is seen as a spectrum, and, as the following chapter on merger policy will reveal, greater attention is appropriately paid to possible abuses depending on a number of attributes related to that apparent power. In the European Union, “dominance” remains more of a dividing line: it requires qualitatively different behavior from firms that fall on one side rather than the other. This would seem to demand a particularly clear definition of what a dominant firm is, and that clarity has not developed. And this problem exists independently of what the European Union regards as “abuse.” As the following chapters will reveal, despite many revisions of policy that appear to bring EU competition policy in the US direction, European practice can often be seen as favoring market participation or rivalry over consumer welfare. While the goal of the United States has become increasingly focused on efficiency construed according to consumer welfare or a

chapter two

34

total welfare standard, consumer welfare as a goal in the European Union contends with rivalry, integration across national markets, and apparently sometimes other goals as well. Both dominance and monopolization deal with firms that are large relative to their markets, so the issue of market definition becomes important. Various US circuits have set the standard of monopoly between 50 and 70 percent in contrast to shares as low as 40 percent in Europe.46 The European Union has also developed the concept of “collective dominance,” but the significance of the concept is evolving. Its past deployment has been confined mainly to situations involving some level of formal agreement among firms.47 Mergers that create a dominant position can already be blocked, but mergers that fall below that threshold might be blocked on grounds of their contribution to tacit collusion. This possibility is explicitly recognized in the EU Horizontal Merger Guidelines of 2004.48 In addition to the different market-share triggers in the two jurisdictions, the two also differ substantially on the kind of behavior that constitutes a violation of their single-firm provisions. Because the EU law governing single-firm behavior is structural, focusing on dominance, while the US law is behavioral, focusing upon monopolization, the EU law sets standards of conduct for firms whose behavior is unregulated by US antitrust law. Under US law only behavior that the courts characterize as anticompetitive or predatory can raise an inference of monopolization, whereas under EU law the kind of behavior that can be deemed abusive is wider in scope and includes refusals to deal, price discrimination, loyalty discounts, and refusals to share intellectual property. Judicial Review While “abuse of a dominant position” may be a more appealing criterion than “monopolization,” the influence of EU competition law also rests importantly on its congruence with the civil law tradition of most of the rest of the world. This implies attention to judicial review in the European Union and how it differs from US practice. The General Court and the Court of Justice are responsible for construing the Treaty and for ensuring that the Commission operates within the bounds delegated to it by the Treaty, that it employs the proper procedures, and that its decisions are supported by evidence.49 On the other hand, the Commission, as the enforcement authority, is understood to possess a margin of discretion within its proper sphere, especially with regard to complex economic assessments, that the courts must respect.50 The review standard

welfare, monopolization, dominance, and Judicial review

35

under which the courts limit their review in this way is commonly referred to as one of “manifest error.”51 In any given case, the determination of where the bounds of the Commission’s authority extend and where the court’s role begins may be difficult to sort out. This problem is not unique to the European context, however. Analogous issues arise in American administration. American observers will find that the issues involved in judicial review of decisions of the European Commission are broadly similar to the issues involved in judicial review US courts of appeal accord to decisions of administrative bodies in the United States. The European courts have not explained how the allocation of responsibilities between courts and the Commission affects judicial review in elaborate detail, as have American courts.52 In its 1998 ruling in Kali & Salz,53 the Court of Justice gave explicit recognition to the Commission’s margin of discretion described above. In that case, the ECJ, prior to invalidating certain Commission findings, took notice of the discretion the Merger Regulation conferred on the Commission: The basic provisions of the Regulation . . . confer on the Commission a certain discretion, especially with respect to assessments of an economic nature. . . . Consequently, review by the Community judicature of the exercise of that discretion, which is essential for defining the rules on concentrations, must take account of the discretionary margin implicit in the provisions of an economic nature which form part of the rules on concentrations.54

Having recognized the Commission’s discretion, however, the ECJ chose to ignore it. Rather, the ECJ ruled that the Commission was unprotected by its discretion because the Commission’s analyses of the concentration and of its market effects were flawed. The court rejected the Commission’s contention that a collectively dominant position could be inferred solely from the market shares of the parties, and it also rejected the Commission’s contention that the two largest potash distributors were structurally linked.55 The General Court followed this approach a few years later when it overturned Commission decisions in several merger cases. Thus in 2002, the General Court reversed the Commission’s rulings barring mergers in Airtours, Schneider Electric, and Tetra Laval.56 Later it reversed the Commission’s approval of a merger in Sony/BMG.57 In Airtours and Tetra Laval the Commission repeated the language of Kali & Salz referring to the Commission’s discretion in making economic assessments, while ruling against the

chapter two

36

Commission.58 In each of these cases the General Court identified so-called errors of assessment by the Commission as well as other errors involving economic issues.59 The Commission in Tetra Laval, believing the General Court had overstepped its authority by improperly intruding into the Commission’s discretion, appealed unsuccessfully to the Court of Justice. The Court of Justice, however, expressed its approval of the intense judicial review the General Court conducted. In its affirmance, the Court of Justice described the duty of the reviewing court as establishing “whether the evidence relied on is factually accurate, reliable and consistent and whether the evidence contains all the information which must be taken into account in order to assess a complex situation and whether it is capable of substantiating the conclusions drawn from it.”60 In the cases just noted both the General Court and the Court of Justice described the deference that the courts owed to the Commission as a prelude to reversing the Commission’s rulings. These decisions accordingly tell us (and are meant to tell us) that the Commission’s discretion is limited and that the courts will not accept Commission decisions that are unsupported or based upon faulty reasoning. These cases, however, do not provide criteria or standards for identifying the types of economic assessments that reviewing courts must respect. This issue came to a head in the Microsoft decision.61 This case differed from the merger decisions described above. Instead of recognizing the Commission’s margin of discretion as a prelude for reversing the Commission, the General Court’s Microsoft decision was marked by substantial deference by the Court to the Commission’s decision. Thus the Court stated: Although as a general rule the Community Courts undertake a comprehensive review of the question as to whether or not the conditions for the application of the competition rules are met, their review of complex economic appraisals made by the Commission is necessarily limited to checking whether the relevant rules on procedure and on stating reasons have been complied with, whether the facts have been accurately stated and whether there has been any manifest error of assessment or a misuse of powers.62

This is a mirror image of the language the courts employed in the merger cases referred to above, where the courts have described the Commission’s decisional discretion before invalidating its decision. Here, in Microsoft, the court described the intensive review with which the courts examine Commission decisions as a prelude to upholding the Commission decision.

welfare, monopolization, dominance, and Judicial review

37

The Court’s deferential approach toward the Commission in Microsoft appears to contrast sharply with the critical approach it took earlier toward the Commission’s merger evaluations in Airtours, Schneider Electric, Tetra Laval, and Sony/BMG. In Microsoft, it was the Commission’s “complex economic appraisals” that carried the day. Yet the Commission’s discretion over “assessments of an economic nature” was insufficient to immunize it from judicial review in the cited merger cases. One difference lies in the General Court’s familiarity and experience with the issues in the merger cases. In sharp contrast, the issues in Microsoft involved the interrelationships between competition law and intellectual property as well as the role of Schumpeterian competition. Existing law gave little guidance about these very important economic issues. Although the ECJ had struggled to reconcile competition law with intellectual property,63 no final criteria for resolving conflicts had been established. Moreover, the role of Schumpeterian competition presented a new issue on which the law was silent and over which experts disagreed. The General Court may have been unsure how to evaluate the case as a matter of law and therefore sought to escape responsibility by deferring to the Commission. The court may also have determined that the policy issues were not tethered to the Treaty in any obvious way and therefore were more political than judicial in nature. As a result, the Court may have considered it more appropriate for the Commission as the more political institution to resolve these policy issues. The European and American approaches to judicial review show close surface resemblances. Like the European courts’ powers over Treaty interpretations, American courts are the final arbiters of statutory meaning. They, like European courts, determine the limits of agency authority and ensure that the agency stays within those limits. Like European courts, they too defer to agency determinations, in some ways more than the European courts. Under the Chevron doctrine,64 US federal courts are required to defer to agency statutory interpretations where the statutory meaning is unclear or ambiguous, but the courts have the final say on whether those conditions are present. US courts also tend to defer to agency determinations on routine issues of administration and issues falling within agency expertise. They defer especially to agency evaluations at the frontiers of science where policy and prediction blend.65 The identification by the Court of Justice of a margin of discretion inhering in the Commission, especially in complex economic appraisals, and its willingness to intrude into the Commission’s exercise of that discretion when the ECJ believes that the Commission has failed to collect all of the relevant information echoes disputes about judicial review of agency deter-

chapter two

38

minations in the United States during the 1970s.66 The currently prevailing position requires judges not only to ensure that legislatively required procedures have been implemented but also to engage in a limited substantive review that is deferential toward an agency’s predictive judgments within its area of expertise,67 a position that appears generally to resemble the review described in the cited opinions by the Court of Justice. Nevertheless, the deference accorded by the General Court to the Commission’s Microsoft decision seems without parallel in the US case law on judicial review because of the magnitude of the policy issues involved. Despite transatlantic similarities in judicial review, there is a long tradition in US administrative law against deference to administrative bodies on core issues that determine the parameters of regulation. The Microsoft case, however, appears to have involved just that: the outcome subordinated intellectual property to competition law and rejected a special place for Schumpeterian competition. These are major matters of policy, and American courts would not defer to administrative decisions on such issues. Moreover, US courts possess final interpretative authority over the meaning of the Sherman Act, an authority that extends to reinterpreting its terms to keep them current with business and economic developments.68 For all of these reasons, a case such as the EU’s against Microsoft would be decided in the United States with little deference to agency determinations. Conclusion This preliminary examination of the complications of differing welfare standards, differences between monopolization and abuse of dominance, and differing review procedures underlies our exploration, in separate chapters, of mergers, exclusive dealing, price discrimination, predatory pricing, loyalty and bundled discounts, intellectual property, and dynamic competition. In these diverse areas, the rationales underlying their differing treatment under US and EU law are interconnected. For example, the rationales supporting EU treatment of loyalty discounts are related to the rationales supporting the treatment of price discrimination, of exclusive-supply contracts, and of predatory pricing, and this treatment differs sharply from the treatment of the same issues in the United States. The rest of this book explores such connections.

3

Merger Policy and Efficiencies Overview One thread can be found in almost all competitionpolicy writing that is also reflected universally in policy: a suspicion— usually abhorrence— of monopoly. Restrictions on conduct aim at the avoidance of formal and informal monopoloid behavior— and where the particular history or technical characteristics of an industry give rise to an overwhelmingly dominant firm, that firm’s conduct, including any merger activity, is almost universally given the closest possible scrutiny. One way of viewing mergers is as a kind of shortcut to market power. Although there may be persuasive efficiency arguments on the other side, both the United States and the European Union regard the creation of such power as highly suspect. In this chapter, we examine the approach of the two jurisdictions on the issue of the extent to which merger-generated efficiencies affect their antitrust evaluations. Our examination extends to horizontal, vertical, and conglomerate mergers.1 One of the most important transatlantic conflicts arose out of a nonhorizontal merger that was proposed between GE and Honeywell, in which the European Commission’s evaluation was based in part on conglomerate effects. States may be greatly affected by merger policies in other jurisdictions even if the firms involved have no production

chapter three

40

facilities in the affected state. Most major jurisdictions have responded by claiming a voice in the approval or rejection of mergers in other states. The EU’s determination to shape the terms of Boeing’s acquisition of the civilian airliner production of McDonnell Douglas in 1997 provides the most well known instance of this important source of international conflict. This case was followed in 2001 by the EU’s blockage of the merger between General Electric and Honeywell. EU policy was subsequently rethought, and despite these episodes, most mergers now seem to be evaluated similarly on both sides of the Atlantic. Many observers are optimistic that future US-EU conflicts can be minimized through increased cooperation and a growing similarity of viewpoint about which mergers should be blocked. The Principal Issues Merger policy may rest on both sociopolitical and economic grounds, and some important decisions on both sides of the Atlantic have stemmed from each.2 In addition to a possible increase in market power, mergers can be opposed as antithetical to the goals of protecting smaller firms from stronger competitors or minimizing concentrated economic power as a political threat. Explicit decision taking on these latter grounds has diminished dramatically in both Europe and America, but a residuum can sometimes be suspected even in quite recent decisions— particularly in the European Union, where the intra-EU distribution of economic activity may provide another de facto ground for approving or blocking a merger. Viewed from a perspective based solely on economic analysis— which is what competition agencies on both sides now claim as the basis for their decisions— the central issues are quite straightforward. In the Harvard paradigm in which market structure (concentration, barriers to entry, product differentiation) determines firm conduct (pricing, promotion, and product development) and hence economic performance (minimum costs, minimum price-margins, the absence of wasteful expenditure, incorporation of the best technology), horizontal mergers lead to a changed market structure in the form of a higher level of concentration of sellers.3 This may facilitate tacit collusion leading to higher prices, poorer quality, or less innovation. 4 In some circumstances, however, a merger may increase the competitive potential of the merged firm and lead to the opposite result: less tacit cooperation with other firms in the market. Those contrasting logical possibilities about a merger’s impact on interfirm cohesion were reinforced by Chicago-led attacks on the value of empirical evidence supporting the assumption of increased concentration leading to decreased competition.5 Another path can also lead from horizontal mergers to a decline in in-

merger policy and efficiencies

41

dustry performance: the phenomenon now called “unilateral effects” in the United States.6 When merging firms produce nonidentical products, their combination may create a price-increasing incentive by internalizing the impact of a change in price of one product on demand for another. In the simplest case, if product varieties A and B are close substitutes, a merger between their producers coordinates joint prices to a higher level for the same reason that a merger between previously noncolluding duopolists does.7 This result relies on no assumptions about tacitly collusive behavior between or among independent firms. Moreover, the mechanism can obtain within an industry in which the level of concentration for the broadly defined product is modest. Merger Efficiencies Our focus in this chapter is on merger-generated efficiencies, a consideration the Harvard paradigm recognized but underemphasized due to some empirical estimates suggesting that major firms in many industries had largely exhausted economics of scale and that large firms or more concentrated industries did not clearly perform better in technological progress.8 Subsequent research contested many of these findings. If mergers do lower cost or increase innovation— even without a direct impact on other firms in the industry— total welfare increases on that account. As the previous chapter explained, if merger policy rested on a goal of maximizing national income and using tax and expenditure policy to deal with distributional matters, changes in total surplus could usefully guide enforcement action. But, as detailed in what follows, this is not the case. Instead, mergers that threaten significant price increases, regardless of resource savings, are not allowed in either the United States or the European Union. The US Experience Background. The principal antitrust provision controlling merger activity is Section 7 of the Clayton Act.9 Both Sections 1 and 2 of the Sherman Act10 also govern mergers. Mergers necessarily fall under Section 1 because there is normally an underlying merger agreement between the merging firms, and a consummated merger always involves a combination between them. Under Section 1, a merger would be formally evaluated under the Rule of Reason, a standard that would yield approval either if overall concentration in the market would not be significantly affected or if, because of efficiencies resulting from the merger, the output of the merging firms would be likely to increase.

chapter three

42

Under Section 2, a merger would be condemned if it would result in a firm with an overwhelming market share, so that it would be treated, for legal purposes, as equivalent to a monopoly. It could also be condemned under Section 2’s prohibitions on attempted monopolization if the merger produced a large market share and there was evidence that the parties intended to use their power to acquire a monopoly share. Most mergers are challenged under Section 7 of the Clayton Act. Congress initially enacted Section 7 in 1914, but that provision was explicitly targeted at stock acquisitions. Despite repeated requests from the FTC, only in 1950 did Congress finally amend Section 7 to extend its reach to include asset acquisitions.11 One reason that Congress was moved to amend Section 7 at that time was the government’s loss of a Sherman Act challenge to a major acquisition by the U.S. Steel Corporation.12 Because the Clayton Act prohibits acquisitions that “may substantially lessen competition or tend to create a monopoly,” Congress believed the amendment would strengthen the government’s power to control merger activity. The first merger cases that arose under the amended Section 7 reached the Supreme Court in the 1960s.13 In these cases the Court showed a strong hostility to all forms of mergers. In its 1962 Brown Shoe decision, the Court condemned the vertical as well as horizontal aspects of a merger between two shoe companies because the merger created distributional efficiencies that, in the Court’s view, threatened less efficient small retailers.14 Thereafter, it condemned a series of horizontal mergers, largely on the ground that the market shares of the combined companies would jeopardize the continued competitive operation of the marketplace. After identifying a 30 percent postmerger market share as threatening undue concentration in its Philadelphia Bank decision of 1963,15 the Court lowered that threshold until 1966 when it condemned a merger between two grocery chains that produced a postmerger market share of a mere 7.9 percent.16 Finally, in that same year, the Court suggested in its Pabst decision that a merger could be condemned even without a showing of a relevant market.17 Starting in the mid-1960s, the court condemned a series of joint ventures, conglomerate mergers, and one vertical merger largely under a judicially created potential competition doctrine.18 The potential competition doctrine applied to cases in which the acquiring firm, although not presently competing with the firm to be acquired, was nonetheless a potential competitor.19 The Court’s hostility to mergers resulting in only modest market shares dissolved in 1974 when it began to take a more balanced stance. In that year the Court, in its General Dynamics decision,20 articulated an approach that supplies the foundation for the present treatment of mergers. There

merger policy and efficiencies

43

the Court indicated that the government fulfills its initial burden under Section 7 by providing statistics showing that the merged company will have an unduly large market share. The burden then switches to the defendants to show that the statistics are misleading, that the merged company will not in fact possess an undue amount of power to raise prices. In General Dynamics, the defendant prevailed because it was able to carry this burden.21 In two market-extension cases of that year involving banks,22 the Court insisted that it was necessary for the government to prove the existence of an economically significant relevant market,23 effectively disowning the contrary language in the earlier Pabst opinion. It also applied its potential competition doctrine with restraint. Efficiency in US Merger Policy Prior to the 1970s, efficiency was given broad recognition as a goal that should be accommodated by US antitrust law. That recognition was tempered by opposing strains of analysis that sometimes treated efficiency as an entry barrier.24 Judge Hand’s Alcoa opinion contains language that would have subordinated efficiency to other social and moral goals that, in his view, the antitrust laws were designed to promote.25 Moreover, the older view held that efficiencies are not sufficiently susceptible of proof to merit recognition in the law itself. Under that view, potential efficiencies are easy to claim but hard to establish. Efficiency claims should be weighed by enforcement agencies only as an element in exercising their discretion whether to challenge a given merger, but efficiency should have no formal place within the law. And the courts therefore should not consider potential efficiencies as relevant to assessing the lawfulness of a merger. In the 1960s, the Supreme Court suggested that no efficiency defense existed. In its Proctor & Gamble decision in 1963,26 the Supreme Court asserted that “possible economies cannot be used as a defense to illegality” in merger cases.27 The general shift in antitrust law in the 1970s in the direction of a Chicago School approach ensured that the issue of efficiencies would remain alive. Yet as late as 1978, Robert Bork argued that it was impractical for the law to inquire directly into the efficiencies particular mergers generated. Bork rather advocated general merger rules setting forth allowable market percentages that reflected the overall probable balance of efficiency and restriction of output.28 Recognition of efficiencies as a legitimate factor in merger evaluation was given increasing support in 1982 when the Justice Department issued new merger guidelines29 that provided operational approaches to market definitions and barriers to entry. The product market is

chapter three

44

defined by considering a hypothetical monopolist producing the goods or services. If that hypothetical monopolist could profitably impose a significant nontransitory price increase, perhaps 5 percent, then that is the market. If it could not, then the market is redefined to include the substitutes in product or geographic space to which consumers would have turned. The analysis is repeated until a price increase by the hypothetical monopolist would be successful and sustainable for a year. Once the market has been defined in this way, the likelihood of coordinated action is partly inferred from a Herfindahl-Hirschman Index (HHI) concentration measure. This measure adds all of the squared shares of market participants. In its original decimal form, this implies that the maximum HHI would be 1. In the raw numbers version competition policy uses, it is 100 (percent) squared or 10,000. In the 1992 Merger Guidelines, postmerger concentration of 1,000 or less generally “required no further analysis.” This same inference would apply to postmerger increase of 100 or less in markets with a postconcentration between 1,000 and 1,800 and to postmerger increases of 50 or less in postmerger markets above 1,800. Greater increases in concentration would cause concern that could still be overcome if other factors promised the maintenance of competition. The basic approach was retained by the 2010 Merger Guidelines with somewhat different categories and numbers that may appear less restrictive but were declared to more closely reflect actual agency decision making. Under guidelines in effect until 2010, concern about unilateral market effects begins with a postmerger share of 35 percent. Whatever the mechanism of price increase following a merger, it could be rendered innocuous by sufficient outside entry or product adjustment by incumbent firms. The expressed standard for entry sufficient to erase the price increase was two years or less until the 2010 guidelines, where it was left unspecified. The 1982 guidelines stated a belief that in “the overwhelming majority of cases, the guidelines will allow firms to achieve available efficiencies through merger without interference from the Department.”30 They revealed reluctance, however, to consider efficiency claims as a means for avoiding condemnation under the guidelines. Thus the Department went on to say that “except in extraordinary cases, the Department will not consider a claim of specific efficiencies as a mitigating factor for a merger that would otherwise be challenged.”31 In an accompanying footnote, the Department asserted that when it did consider efficiencies, they would have to be proven by “clear and convincing evidence.”32 The Department’s position grew less formally hostile to efficiency claims in its 1984 revision of the guidelines.33 In its introduction to the 1984 guidelines, the Department distanced itself from the just quoted language of its

merger policy and efficiencies

45

1982 version. The 1982 guidelines were described as having “a restrictive, somewhat misleading tone.”34 Moreover the earlier statement that the Department would consider efficiency claims only in “extraordinary cases” did not correspond with the Department’s declaration in the 1984 version that it “never ignores efficiency claims.”35 Consistent with its formally more accepting approach toward efficiency claims, the text of the 1984 guidelines states affirmatively that the Department will “consider” a claim that a merger was necessary to achieve “significant net efficiencies,” if the parties establish such efficiencies “by clear and convincing evidence”36 The 1992 revision of the guidelines37 incorporates the 1984 provisions almost verbatim, except that the requirement that the merger parties establish efficiencies by “clear and convincing evidence” was omitted.38 The efficiency provisions of the 1992 guidelines were rewritten in 1997.39 The tone of the revision was superficially more favorable to a showing of efficiencies. The 1997 revision still placed the burden of demonstrating efficiencies on the parties to the merger with the rationale that they are most knowledgeable on the subject. These provisions identify several types of possible efficiencies and connect those efficiencies to possible antitrust concerns. For example, they explicitly recognize that reductions in marginal cost specific to only some sellers may reduce the likelihood of coordinated interaction (that is, oligopoly pricing). The 1997 revision left a number of issues connected with efficiencies defenses unresolved. The omission of language, such as appeared in the 1992 version, favorable to fixed-cost reduction suggests that the Department had more fully embraced a pure consumer-surplus standard because only reductions in short-run marginal cost will lead to immediate price reductions and thus meet the Department’s concern that efficiencies offset the price effects of any market power the merger generated.40 A footnote in the 1997 revision does say that the Department “will consider the effects of cognizable efficiencies with no short-term, direct effect on prices in the relevant market,” but the context suggests that the efficiencies so considered must produce long-term enhancements to consumer surplus.41 The 2010 Guidelines The 2010 Merger Guidelines embody some noticeable changes from the previous versions. Some observers have made much of the fact that the 1992 guidelines explain the government’s burden as proving that a merger is “likely substantially to lessen competition,” while the 2010 version emphasizes the original Clayton Act text that employs “may,” thus seemingly showing greater fidelity to the “incipiency” standard of the Clayton Act. The

chapter three

46

language also may suggest more confidence about forecasting the course of market developments than the previous version.42 An undeniable change is the way the agency activity is explained. The 1992 guidelines state that “the Agency will first define the relevant product market with respect to each of the products of each of the merging firms.”43 The guidelines then proceed to identify market participants and their shares for the product and market concentration. The 2010 guidelines deemphasize the primacy of market definition, stressing not only that it may not be the first step in agency analysis but also that “the principles of market definition outlined below seek to make this inevitable simplification as useful and informative as is practically possible. Relevant markets need not have precise metes and bounds.”44 Consistent with the downgrading of market definition is a much stronger emphasis on unilateral effects. The original 1982 guidelines treated such effects as the result of mergers almost as an afterthought; principal attention focused on the merger’s effect on oligopolistic cohesion. Subsequent guidelines have paid ever-greater attention to unilateral effects until their 2010 discussion actually precedes the previously principal concern. Three factors seem to underlie this development. First, the original scholarship favoring the positive relationship between concentration and profitability was seriously defective, as Chicago School research demonstrated. Second, although the logic— and some evidence— of such a relation remained, careful research suggested that cohesive tendencies resulting from higher concentration could easily be swamped by myriad industry-specific characteristics. Third, better data— most notably transactions data from optical scanning— often made it possible to isolate the impact of price changes on a particular brand of a differentiated product on sales of another. There was therefore inevitably more attention drawn to “submarkets” within a broader “market.” In fact, the new guidelines may reflect a clearer explanation of agency behavior, along with some change in emphasis more than a major change in approach.45 Consider the original 1982 procedure: the market is first defined by taking the smallest number of firms and geographic areas in which price could be raised by (say) 5 percent and sustained for a year. If subsequent entry or “repositioning” by other firms was anticipated to drive prices back to their previous levels within the next twelve months, the entry was “easy”; otherwise it was not. The new guidelines do not tie “easy” entry to the time frame of twenty-four months. They say instead that “to deter the competitive effects of concern, entry must be rapid enough to make unprofitable overall the actions causing those effects.” Nevertheless, the logical path is the same. A “safe harbor” market share is no longer suggested for unilateral effects, but this may have little enforcement significance. If the merger of two

merger policy and efficiencies

47

firms removes competition between them for a certain product segment and allows for a significant and substantial nontransitory increase in price (SSNIP) for a year, then such a “submarket” has in fact experienced merger to monopoly by traditional guidelines logic. But one might surmise that exactly because these are typically varieties of what most would regard as a single product, such pricing opportunism would lead to reactions by other incumbent firms either before a year had passed or shortly thereafter, implying a situation of “easy entry” and the absence of an antitrust problem. Nevertheless, the mechanism is entirely industry specific and is largely unconnected with market shares of more broadly defined products. Moreover, none of this can ignore that enforcement resources are scarce and should aim at maximum effect, which is presumably an increase in aggregate consumer surplus. Minor increases in varietal monopoly power in otherwise innocuous or procompetitive mergers must necessarily be ignored. And nothing in the new guidelines suggests otherwise. The impact of own-price and cross-price elasticity on the effect of mergers has generated an enormous literature, each strand of which grows from alternative approaches to defining a market or forecasting the impact of a merger in situations of market power prior to the merger. “Critical loss” analysis,46 mentioned in the 2010 guidelines for first time, attempts to answer the question: what is the percentage unit sales decrease that would make a certain percentage price increase unprofitable? This question can be used for market definition by extending it over a larger group of firms until the expected unit loss moderates to profitability. But the approach is more than a trivial restatement of the original guidelines market logic because it focuses on the key element driving profits down: the price-cost margins of the firms involved. Any loss of units from a price increase is greater as the margin from which the price rise begins is greater. On the other hand, a high margin suggests price inelasticity and hence relatively slight response in unit terms for a still higher price. The importance of firms’ original price-cost margins to their postmerger behavior lies at the heart of an approach called Upward Pricing Pressure (UPP), which attempts to combine the increase-braking impact of existing margins with a consideration of a merger’s impact on marginal cost. The UPP approach consolidates these concerns into a single number that can be calculated using only price-cost margins and estimates of marginal cost shifts.47 (The term appears in the 2010 Merger Guidelines with no technical discussion.) Market definition is not an element in the exercise. Critics have noted that the technique only identifies a tendency for prices to rise and without additional information cannot provide estimates of how much prices will rise. Other proposed approaches to forecasting the price effects

chapter three

48

of mergers either posit one or more demand function shapes (to check for consistency)48 or involve more elaborate simulation.49 Efficiency Standards Broadly speaking, the three iterations of the Merger Guidelines that were issued in Republican administrations appear somewhat more open to influence by a total surplus standard than was the penultimate revision of the efficiency provision that was issued in the Clinton administration.50 The same tension seems to have extended into the Bush and Obama administrations. Some Bush appointees saw the standard implied by the efficiency section of the 1997 US guidelines as more open to a consideration of total welfare because savings are recognized that may not be immediately passed along to the consumer.51 In contrast, efficiencies that are not passed on do not figure into the 2010 guidelines, and the overall tone of that document appears to some observers clearly more “litigation friendly” than the ones they replaced, perhaps signaling a kind of “counterrevolution” to the Chicago School.52 The Bush administration’s assistant attorney general made no such suggestion in his critique of the guidelines, however. He did not favor all of the changes and objected especially to the incorporation of “upward pricing pressure” as premature. Nevertheless, he saw merit in much of the emendation. His overall assessment was that “the proposed Guidelines will remedy some, though not all, of the limitations of the previous Guidelines. However, unless the many caveats to some of the techniques discussed in the proposed Guidelines are clearly understood, the proposed Guidelines run the risk of giving courts and foreign antitrust agencies a false impression of the reliability of some modes of merger analysis.”53 For the purposes of this book, another characteristic of the revised document stands out. As will be seen, the de-emphasis on market definition, greater emphasis on the sufficiency of entry to render previous price increases unprofitable, and the absence of numerical safe harbors for unilateral effects all bring the new guidelines into closer accord with their European counterpart.54 Markets, Mergers, and Efficiency in the US Courts Most US courts assert that efficiency concerns underlie the antitrust laws and give verbal recognition to an efficiencies defense in merger cases.55 These courts probably are equating “efficiencies” with increasing consumer surplus.56 No published court of appeals decision, however, has yet upheld a merger solely on the ground that it furthered efficiency.57 This dearth of

merger policy and efficiencies

49

cases may reflect judicial caution about acceptable evidence to establish an efficiencies defense. A number of courts have recognized the substantial cost savings that a merger would generate and yet have failed to accord controlling weight to that evidence because they believed those savings could be generated without the merger.58 Efficiencies defenses have been successful in a few lower court decisions, although in these cases the mergers might have been upheld on the basis of other defenses in the absence of those efficiencies.59 In the widely publicized Staples case involving the merger of two officesupply superstore chains, the district court tentatively evaluated a proffered efficiencies defense, but rejected it on the ground that the efficiencies were insufficient to avoid a price increase projected by the government’s expert.60 The Eighth Circuit, in its Tenet Health Care decision,61 perhaps came the closest of any court of appeals to using an efficiencies rationale in support of a merger. In that case the court vacated the lower court’s injunction against the merger, ruling that the FTC had not presented adequate evidence of a relevant geographic market and that the lower court failed to consider evidence of efficiency. It is unclear whether the court would have accepted evidence of cost savings as sufficient to overcome the inference of anticompetitive effects the FTC drew from increased concentration. A significant number of judicial decisions that discuss the efficiencies defense have indicated that a pass-on of a portion of merger-generated efficiencies to consumers is a required part of that defense.62 These decisions thus provide support for a merger-evaluation standard keyed to consumer surplus. But the real criterion could be stated more clearly. If there is little or no risk of an increase in market power from a merger in an imperfectly competitive market, part of any marginal cost saving will necessarily be passed on. On the other hand, if market power is increased but is just compensated for by a decline in marginal cost, consumers are no worse off and the merger should still be allowed. Certain estimates are very rare, of course. Instead, the courts seem generally to make probabilistic calculations using a kind of intuitive “sliding scale”: if there is a prospect of market power increase then there must be a likelihood of a compensating decline in marginal cost so that price does not rise. Such a “sliding scale” approach can be seen in the DC Circuit Court’s analysis of efficiency in the Heinz case.63 Contrary to the other merger decisions referred to above, the Heinz court ignored the possibility that the merged firm might pass on some of the merger-generated cost savings to consumers. The case concerned a proposed merger between Heinz, the secondranking producer of baby food in US sales volume, with Beech-Nut, the third-ranking seller. Heinz supplied 17.4 percent of the US market and BeechNut supplied 15.4 percent. The dominant seller was Gerber, with 65 percent

chapter three

50

of US sales.64 The merger would have substantially increased concentration, from three producers to two, and would have increased the HerfindahlHirschman Index from a premerger level of 4,775 by 510 points.65 This substantial potential increase in the HHI index exerted a profound effect upon the court’s decision. While the concentration-producing aspects of the merger were clear enough, it also appeared that the merger would result in significant efficiencies. The Beech-Nut plant was old and inefficient, while the Heinz plant was new, highly efficient, and underutilized. Had the merger taken place, the production that was currently performed at the Beech-Nut plant would have been diverted to the more-efficient Heinz plant. According to the lower court, that consolidation of production in the Heinz plant would have reduced the production cost of the Beech-Nut share of the production of the two companies by 43 percent.66 The court of appeals cut this figure to 22.3 percent, however, on the basis that the 43 percent figure referred to “variable conversion cost” and that “variable conversion cost” was only part of Beech-Nut’s overall variable manufacturing cost. The court then stated that even the 22.3 percent cost reduction was not an accurate statement of the mergergenerated cost savings because the cost savings resulting from the merger should be assessed against the entire production of the merged entity.67 Although the court ended its discussion here, it appears that Beech-Nut’s actual production would be approximately 47 percent of the total production of a combined Beech-Nut/Heinz merged entity.68 Therefore, the savings would be at least 22.3 percent of 47 percent or a cost savings of 10.48 percent of the production costs, measured against the output of the combined entity. Economic analysis suggested that somewhere between 50 and 100 percent of the cost saving was likely to be passed on to consumers even in the more highly concentrated postmerger market.69 The DC Circuit justified its decision by endorsing a high standard of proof for the efficiencies defense in cases like Heinz where concentration levels were high. Indeed, in such cases the court required proof “of extraordinary efficiencies.”70 This imposition of a high standard of proof carries dire implications for other cases like Heinz that involve both high concentration and substantial efficiencies. The Heinz ruling casts doubt upon the general availability of an efficiencies defense for a merger under the consumer surplus standard. Nevertheless, the Heinz court left some room for future development. In demanding proof of “extraordinary efficiencies” from mergers in highly concentrated markets, the court was employing a “sliding scale” approach to the proof of efficiencies by demanding a higher degree of proof of efficiencies in merger cases as the degree of market concentration increases.

merger policy and efficiencies

51

A careful reading of the US Guidelines may provide significant guidance. The 2010 guidelines contain the following language that is unrevised from the previous version: The Agency will not challenge a merger if cognizable efficiencies are of a character and magnitude such that the merger is not likely to be anticompetitive in any relevant market. To make the requisite determination, the Agency considers whether cognizable efficiencies likely would be sufficient to reverse the merger’s potential harm to consumers in the relevant market, e.g., by preventing price increases in that market.71

This is a clear statement of a consumer welfare standard. The 2010 guidelines go on reiterate the previous guidelines’ rejection of the total surplus alternative with unrevised language: “in conducting this analysis, the Agency will not simply compare the magnitude of the cognizable efficiencies with the magnitude of the likely harm to competition absent the efficiencies.”72 The ensuing explanation implies that the agencies will consider not just their best single estimates of increased market power and diminished costs but also the array of possible outcomes in both spheres and the probabilities of those outcomes. Merger Policy in the European Union Prior to the adoption of the Merger Regulation in 1989, Articles 101 and 102 were the only legal provisions governing competition law. Article 102 prohibits the “abuse” of dominant position, but holding a dominant position is not illegal. Article 101, however, would apply to invalidate any merger agreement that would prevent, restrict, or distort competition within the internal European market. Under Article 101(3), however, the European Commission would have the authority to exempt any agreement that: contributes to improving the production or distribution of goods or to promoting technical or economic progress, while allowing consumers a fair share of the resulting benefit, and which does not: (a) impose on the undertakings concerned restrictions which are not indispensable to the attainment of these objectives; (b) afford such undertakings the possibility of eliminating competition in respect of a substantial part of the products in question.

chapter three

52

These same considerations were later incorporated into the Merger Regulations of 1989 and 2004.73 The insistence that agreements under Article 101(3) furthering “production or distribution of goods or to promoting technical or economic progress,” must also provide consumers with a “fair share” of the resulting benefit implies a consumer welfare standard. Nevertheless, just how the modifying clause constrains the Commission from granting an exemption is unclear. When the agreement in question merely results in a cost reduction for the participating firms and no market power, their marginal cost curves would move downward. If such firms account for a negligible proportion of industry output, the resulting cost reduction would be reflected only in an increase in output by the cooperating firms and above-normal profits. There would be no price reduction74 and no perceptible gain for consumers— although they would suffer no loss. If the method of cooperation were also available to others, however, then one would expect the new technique to be widely replicated with a resulting cost reduction throughout the industry. In such a case, price would fall with attendant consumer benefits. Of course, a merger such as the one described might not need the Article 101(3) exemption because it would not prevent, restrict, or distort competition. But other mergers that took place in imperfectly competitive markets could generate downward shifts in marginal cost as well as increased market power. In those cases, the issue would be which effect predominated. Article 101(3)’s clause (a) contains a further restriction, analogous to the judicially constructed “least restrictive means” condition that has been found in US antitrust law.75 In the United States, the least restrictive means condition developed in application of the Rule of Reason. A restraint was unreasonable if there were less restrictive means available to accomplish the same objective. The condition raises a problem, however, because one can almost always find some less restrictive means to accomplish an objective with sufficient study and the wisdom of hindsight. The US courts have become increasingly sensitive to the unreasonableness of treating business executives as if they were omniscient at the time they acted. In 1989 the Council adopted a Merger Regulation76 that expanded the coverage of its competition law over mergers. The Commission must be notified of intended mergers with a “Community dimension,” which turns on the volume of sales in Europe and worldwide.77 This left mergers between smaller firms to be handled by the national laws of the member states. DG Comp employs methods to define relevant markets in a roughly similar way to those used in the United States.78 Cultural differences across Europe, however, sometimes lead to less substitution in response to relative price

merger policy and efficiencies

53

changes and hence to more narrowly defined markets than is true in the United States.79 Until the early 1990s the European concern with “dominance” completely eclipsed attention to either oligopolistic interdependence or unilateral effects.80 If a handful of firms of roughly equal size exhausted a market, a merger between any two of them would not be blocked unless it created a “dominant position.” If, for example, dominance were considered to be a 25 percent share, then in theory an industry of ten equal-size firms could apparently have engaged in five mergers, each to a 20 percent share without violating the law. Such consolidation would always have attracted agency attention in the United States.81 Since the early 1990s, the Commission, after having persuaded the General Court to accept the concept of “collective dominance” under Article 102,82 incorporated that concept into enforcement of the Merger Regulation (in the Nestle-Perrier case83). The concept is referred to in the EU’s 2004 Horizontal Merger Guidelines,84 and those guidelines themselves devote twentyone paragraphs to the analysis of coordinated effects.85 The Commission has thus formally moved closer to the approach of the US merger guidelines that have always recognized the tendency of a merger to further oligopolistic coordination.86 As in the United States, however, sufficiently easy entry could obviate concerns about the size of merging firms relative to the total market.87 Analogous to the exemption provision in Article 101 of the Treaty, Article 2(1) of the 1989 Merger Regulation presents as one of the considerations for evaluating a merger “the development of technical and economic progress provided that it is to consumers’ advantage and does not form an obstacle to competition.”88 Thus while efficiencies are recognized, the Merger Regulation appears unambiguous about the requirement for an expected fall in price.89 Moreover, the second part of the phrase could be seen as an expression of concern about a possible threat of the merged entity to its rivals based on its new efficiency. In fact, even when efficiencies have been estimated to be sufficiently large to induce a seller to lower its price, the Commission has not always looked favorably upon them. In the de Havilland merger case of 1991,90 the Commission not only rejected an efficiencies defense but also counted the lower costs and resulting lower prices that would have been available to the merged firm as a reason for disapproving a merger.91 The Commission has often made decisions about mergers and also enforced them in a complex political environment. In particular, although the Commission has sought to bind the European Union ever more tightly together, the national governments that sometimes pressure it have frequently

chapter three

54

had views that differ from the Commission— and from one another. On some occasions, the positions of national governments on issues of competition policy have emphasized efficiency over rivalry more than has the Commission.92 In general, however, national governments are likely to be primarily concerned with the probable impacts of mergers on the location of activity, and particularly immediate gains and losses of jobs— concerns that weigh the interests of workers and suppliers ahead of consumers and lower prices. Within the Commission itself, DG Enterprise and Industry, which must deal with policy issues surrounding industrial policy, has exerted pressure to further its aims. DG Enterprise and Industry, France, and Italy all lobbied to influence the Commission’s evaluation of the de Havilland merger.93 After the 1991 de Havilland decision, the Commission— which had resisted pressure from DG Enterprise and Industry in its ruling— bowed to DG Enterprise and Industry’s concerns by amending its rules to obligate the competition commissioner henceforth to inform other commissioners after he decides to open a second-stage investigation.94 This increased the opportunity for formal consideration of factors beyond competition in the Commission’s final disposition of the case. Immediate political considerations may have played a role in the several instances in which European merger enforcement has treated efficiencies as negative factors because of their potential for damage to competitors. This was the case in de Havilland,95 where the Commission evaluated an acquisition by an Italian-French aerospace venture of a Canadian subsidiary of Boeing. The Commission believed that the acquisition would have given the merged company an advantage vis-à-vis its European rivals in managing currency risk. But because the Commission believed this cost advantage would be employed in undercutting less efficient rivals who would ultimately be driven from the market, it disapproved the merger. Again in the Boeing/ McDonnell-Douglas merger of 1997, the Commission approved the merger only on the condition that Boeing surrender exclusive-supply contracts that it had with three large US air carriers (American, Continental, and Delta).96 Although the foreclosure effect of these contracts was only an estimated 11 percent of the market,97 the Commission thought those effects would be exacerbated when other airlines demanded similar contracts because of the cost savings they generated to the aircraft purchasers (as well as to Boeing itself). As a result of Boeing’s surrender of those supply contracts, both its own costs and the costs of its customers apparently increased. The Commission’s imposition of this cost-raising condition cannot easily be reconciled with a pursuit of consumer surplus maximization.

merger policy and efficiencies

55

In the Commission’s 2001 GE/Honeywell decision,98 a major ground for the Commission’s attack on the merger was the likelihood that the merged company would be able to bundle products and offer prices lower than rivals could offer.99 This is an example of what in the European Union is known as “portfolio effects.” The Commission treated supposed merger-generated efficiencies, in this case the avoidance of double marginalization of complements, as negative factors in its evaluation.100 The presence of extensive direct supplier-purchaser bargaining in the aircraft industries made such a motivation or result unlikely, however. The Commission alternatively relied on GE’s financial strength to cross-subsidize the packages.101 Because the Commission’s own scenario involves immediately lower purchaser prices and assumptions of foreclosure without any plausible predation story, the case appears as an extreme example of rivalry as a preferred competition policy goal.102 Although the General Court affirmed the Commission’s rejection of the merger, it rejected the Commission’s bundling analysis on the ground that it lacked support in the record.103 The GE/ Honeywell decision is an important symbol of US-EU differences in competition policy and will be discussed further. The Commission’s decision in the Metso/Svedala merger of 2001104 also conforms to this pattern. In that case, the merging firms were engaged, inter alia, in the production of rock-crushing equipment. Because the Commission determined that as a result of the merger the merged entity would acquire increased economies of scale,105 it approved the merger only when the parties agreed to significant divestment.106 The Commission’s stated objection to the merger as originally planned focused on the merged company’s alleged ability “to engage in targeted competitive actions” against its smaller competitors.107 In Danish Crown/ VestjyskeSlagterier of 1999,108 the Commission provided a somewhat different rationale for rejecting a merger that would have apparently benefited consumers. In that case, involving the proposed merger of two meat processors, the Commission stated: “The creation of a dominant position in the relevant markets . . . means that the efficiencies argument put forward by the parties cannot be taken into account in the assessment of the present merger.”109 This case suggests that under the then governing Merger Regulation of 1989, the Commission saw no positive role for efficiencies in merger evaluation. If the merger in Danish Crown/ VestjyskeSlagterier had not created a dominant position, the merger would have been compatible with the common market and would have been approved. It is precisely in the case where a dominant position would be created and the merger would normally be disapproved that recognition of efficiencies as a positive factor is needed. If, in this case, merger-generated

chapter three

56

cost savings would have benefited consumers through lower prices, then the merger should have been approved. In the five merger cases discussed above (de Havilland, Boeing/ McDonnell-Douglas, GE/Honeywell, Metso/Svedala, Danish Crown) the Commission appeared to treat merger efficiencies as problematic. These cases were all decided under the 1989 Merger Regulation that alludes to efficiency only in the cited passage that approves of “the development of technical and economic progress provided that it is to consumers’ advantage and does not form an obstacle to competition.”110 In each of these cases, the Commission disapproved a merger or conditioned its approval because of the presence of efficiencies that would have enabled the merged company to undercut its rivals. Although the current (2004) Merger Regulation contains precisely the same language to recognize “progress” (Article 2(1)(b)), Recital 29 and the new Regulation’s Guidelines both emphasize a merger’s possible contribution to efficiency. The Merger Regulation, Its Revision, and the EU Guidelines. Under the Merger

Regulation, in either its original 1989 or its current revised 2004 form, the Commission determines whether mergers and acquisitions (which it defines as “concentrations”) are “compatible” or “incompatible” “with the common market.”111 Under the 1989 Merger Regulation, a concentration is “incompatible with the common market” if it “creates or strengthens a dominant position as a result of which effective competition would be significantly impeded in the common market or in a substantial part of it.”112 Under the 2004 Merger Regulation, a concentration is incompatible with “the common market” if it would significantly impede effective competition in the common market or in a substantial part of it, in particular as the result of the creation or strengthening of a dominant position. The Commission’s 2004 guidelines on horizontal mergers indicate that a concentration with a market share of 25 percent or less should be presumed to be compatible with the common market.113 The creation or strengthening of a dominant position for purposes of the Merger Regulation, therefore, involves a postmerger firm with more than a 25 percent market share.114 Although the guidelines state that several mergers resulting in market shares between 40 and 50 percent have led to the creation or strengthening of a dominant position (and that in some cases even mergers involving shares below 40 percent have produced that result), the Commission seems to be saying that a 25 percent share is a safe haven and is not suggesting that mergers involving market shares above 25 percent but below 40 percent are problematic.

merger policy and efficiencies

57

In late 2001, a green paper dealing with reform of the EU Merger Regulation raised the possibility of incorporating into the new regulation the substantial-lessening-of-competition language of the Clayton Act.115 In the end, however, this language was not adopted, and much of the language of the new Merger Regulation is virtually identical to the older regulation.116 Contemporary observers saw two major reasons why the new regulation was not changed more substantially: it was not seen as necessary for a change in policy, and retaining much of the basic language contributed to the continuity of EU competition law.117 While the language of the revised Merger Regulation closely resembles the language of the original Merger Regulation, the new language is intended to produce a substantial shift in policy. The revised regulation promulgates a substantive test that aims to determine whether a merger would “significantly impede effective competition in the common market or in a substantial part of it, in particular as the result of the creation or strengthening of a dominant position.”118 This language reverses the position of the two clauses, raising the “significantly impede” clause (the SIEC test) to a level of independent significance. As previously noted, the EU Horizontal Merger Guidelines endorse the concept of collective dominance119 and discuss the problems of coordinated effects120 and noncoordinated effects (unilateral effects).121 The discussion of factors contributing to their success strongly parallels the US guidelines’ discussion of unilateral effects and tacit collusion. Moreover, like the US Merger Guidelines, the EU Horizontal Guidelines measure concentration using HHI thresholds. These thresholds are only slightly more lenient than those in the 1992 and 1997 US guidelines— which were understood to be lower than those actually used for enforcement. The possible efficiencies of a merger are emphasized. Recital 29 of the 2004 regulation states: “It is possible that the efficiencies brought about by the concentration counteract the effects on competition, and in particular the particular harm to consumers, that it might otherwise have.”122 The regulation’s accompanying EU guidelines’ recognition of efficiencies closely resembles the treatment across the Atlantic. They must benefit consumers, be merger specific, and be verifiable.123 Moreover, the Commission has invoked efficiency as a supporting reason in a number of cases for ruling that a merger did not pose a significant impediment to effective competition and thus met the standards for approval under the Merger Regulation. The EU Non-Horizontal Merger Guidelines and the GE/Honeywell Case. It will

be recalled that the Commission in its GE/ Honeywell decision was concerned that the merger would produce a combined company that would

chapter three

58

market complements (jet engines and avionics), would have an economic incentive to offer these products in a bundle, and that rivals would be unable to meet the bundle prices. This is one reason why the Commission disapproved the merger. This rationale produced a profound shock in American antitrust circles. That decision, involving a widely watched merger between major international companies, rendered the conflict between European and American antitrust policy highly visible. In response, the Organization for Economic Cooperation and Development (OECD) organized a policy roundtable in 2001 on the antitrust evaluation of portfolio effects.124 The US Department of Justice (DOJ) submitted to the roundtable a twenty-five-page memorandum on its analysis of “range effects,”125 the term it believed best described the Commission’s approach in GE/ Honeywell and other conglomerate merger cases. The DOJ found three theories of range effects in the European Commission cases: (1) the merger would create economies of scale and scope that rivals could not match, (2) the merged firm would gain a decisive advantage over its rivals by virtue of its greater size and financial resources, and (3) the merger would facilitate tying or bundling of complementary products. The DOJ rejected the first rationale as inconsistent with the objectives of antitrust law. It rejected the second rationale on the ground that there was no empirical support that size alone confers any significant competitive advantage. It then addressed the third rationale (about bundling complements) in the greater part of its submission. The DOJ argued that the cases where bundling would foreclose rivals and produce negative welfare effects were uncertain and required that at least seven conditions be met, making it virtually impossible to predict a negative outcome with confidence. The DOJ went on to say that proof that all of those conditions have been met “requires making guesses about the future conduct of the merged firm, its customers and its rivals that are beyond the capability of even the most prescient competition authority.”126 Therefore, prohibiting a merger on the ground that it would enable anticompetitive bundling would eliminate the efficiency improvement of avoiding double marginalization and would otherwise potentially deter a large class of efficient mergers on the basis of highly speculative and unprovable theories of competitive harm. As noted above, the General Court’s affirmance of the Commission’s GE/ Honeywell decision rejected the Commission’s portfolio count as unproved. Three months after its GE/Honeywell decision, the Commission, in Tetra Laval/ Sidel,127 disapproved a merger between two companies involved in different branches of liquid-food packaging, again employing a conglomerate (portfolio effects) analysis. In the Commission’s view the dominance

merger policy and efficiencies

59

of Tetra in cartons and the technological skill of Sidel in machinery for plastic containers would facilitate the merged company offering bundles to customers who were gradually expanding their use of plastic containers. This bundling (of cartons and plastic container machinery) would assist the merged company, which already had dominance in cartons, to acquire dominance in the plastic-container machinery industry as well. Although the Commission conceded that the two product lines involved were not complements in the economic sense “but technical substitutes,” it nonetheless contended that the two product lines provided a stronger incentive for leveraging its dominance in cartons into machinery for plastic containers.128 The Commission’s Tetra Laval/Sidel decision was reversed by the General Court in 2002,129 ruling that the Commission had not proved its case. On appeal by the Commission, the Court of Justice affirmed the General Court,130 observing that the effects of a conglomerate merger are generally considered to be “neutral, or even beneficial, for competition on the markets concerned.”131 As a result, the ECJ demanded convincing evidence of the circumstances producing the alleged anticompetitive effects.132 The rejection by the General Court of the Commission’s conglomerate allegations against both the GE/ Honeywell and Tetra Laval/Sidel mergers on failure-of-proof grounds raises the question of whether the Commission’s portfolio effects positions had been put to rest. Subsequent to its rulings in GE/Honeywell and Tetra Laval/ Sidel, the Commission, in 2004, issued its revised Merger Regulation that explicitly recognizes efficiencies as positive factors in merger evaluation. The EU Horizontal Guidelines, issued the same year, confirm the importance of efficiencies in its thirteen paragraphs devoted to the analysis of efficiencies. The new Merger Regulation and the EU Horizontal Guidelines suggested to many that the policy conflict revealed in the GE/Honeywell case was ending. In 2008 the Commission issued guidelines on nonhorizontal mergers.133 These guidelines deal with both vertical and conglomerate mergers. Part IV of these guidelines deals with vertical mergers. Here the Commission recognizes the procompetitive potential of vertical mergers. Moreover, this Part explicitly recognizes the avoidance of double marginalization as a potential efficiency of vertical mergers.134 And the Commission has pointed to this avoidance as a positive factor in its assessment of vertical mergers.135 Because this issue (double marginalization in supplier-purchaser transactions) is closely related to the contentious issue in GE/ Honeywell (double marginalization in sales of complementary products), a reader of (only) Part IV might think that the Non-Horizontal Guidelines confirm the end of the USEU policy conflict. Part V, however, shows that the policy conflict continues.

chapter three

60

The Non-Horizontal Guidelines reveal that the Commission has continued to view mergers involving producers of complementary producers with suspicion. Thus in paragraph 101 of these guidelines, the Commission explains: Where a supplier of complementary goods has market power in one of the products (product A), the decision to bundle or tie may result in reduced sales by the non-integrated suppliers of the complementary good (product B). If further there are network externalities at play, this will significantly reduce these rivals’ scope for expanding sales of product B in the future. Alternatively, where entry into the market for the complementary product is contemplated by potential entrants, the decision to bundle by the merged entity may have the effect of deterring such entry. The limited availability of complementary products with which to combine may, in turn, discourage potential entrants to enter market A.136

The economic analysis contained in paragraph 101 is not incorrect, but it focuses on what the DOJ concluded were the less likely anticompetitive effects rather than the more likely procompetitive effects the bundling of complements generated. The facts described are reminiscent of GE/ Honeywell. In the circumstances described nonintegrated competitors in product B would suffer reduced sales if they failed to reduce their prices or if they failed to team up with a producer of product A to offer a competitive bundle. But if they failed to do either, then the US position is that the seller offering the bundle deserves to replace them because its bundle increases consumer (and total) welfare. Other paragraphs of Part V also recognize the effect on complementary product prices when a single seller of both products internalizes the effect and reduces the price for a bundle.137 A reader of the NonHorizontal Guidelines may wonder whether paragraph 101 was included for purposes of analytical thoroughness or as a reassertion of the Commission’s position in GE/Honeywell. This difference in approach continues. In a 2009 symposium, the economist James Langenfeld criticized the Justice Department for its failure to revise its merger guidelines on nonhorizontal mergers since their issuance in 1984.138 Langenfeld complained that, as a result, current economic theoretical insights were being neglected. But in sharp contrast, Gregory Werden, a DOJ economist, countered that the EU (and Australian) Non-Horizontal Guidelines: outline the possible anticompetitive effects of such mergers as identified by the economic literature, but they say little about the

merger policy and efficiencies

61

conditions under which action will be taken against proposed nonhorizontal mergers. The reason is that the economic literature does not lend itself to practical application. None of the multiple theories of anticompetitive effects has general applicability under readily identifiable conditions.139

Werden’s comments echo the DOJ’s position in its 2001 submission to the OECD policy roundtable. That position rejects the inclusion in antitrust guidelines of economic theories that do not fit the facts of common business events. The DOJ, as the 2001 submission makes clear, is concerned that the inclusion of purely theoretical insights into antitrust guidelines carries the potential for deterring socially beneficial business behavior and for enabling officials to challenge such behavior. This difference in perspective may explain the fact that the US guidelines are less than 3,200 words, while the EU guidelines are more than four times as long. The US guidelines treat potential competition and vertical mergers but do not claim to be exhaustive, stating merely that “this section describes the principal theories under which the Department is likely to challenge non-horizontal mergers.”140 Conclusion Overall, the treatment of unilateral and collusive merger problems as well as efficiencies now appears quite similar in the United States and the European Union as matters of declared policy. Unfortunately, not enough pivotal cases have been decided since 2004 to establish the extent of EU policy change.141 Some of the most notable, Inco-Falconbridge,142 and Ryanair-Aer Lingus,143 confronted the Commission with claims of efficiency that were rejected, but the levels of concentration involved were very high, and it is unlikely that US authorities would have handled them very differently.144 As has been true in the United States, the standard for demonstrating efficiencies and their favorable impact on prices appears to become increasingly stringent as one or both previous shares are very high.145 In the area of vertical and conglomerate mergers and on mergers involving potential competition, the extent of agreement is unclear. The United States remains highly skeptical that such mergers typically harm welfare and has retained its 1984 guidelines despite the admitted possibility that anticompetitive outcomes can sometimes arise. Its position has consistently been that those conditions are too idiosyncratic to be captured in useable guidelines. The European Union has followed another path. The 2008 guidelines for nonhorizontal mergers trace a number of possible lines of harm, but the document also stresses that such mergers are typically benign. Only

chapter three

62

time will reveal whether this apparent transatlantic difference between the stated enforcement postures— rather close convergence on horizontal mergers and much wider differences on vertical and conglomerate mergers146— will actually lead to important differences in enforcement. The EU guidelines’ greater expressed concern for anticompetitive outcomes could either serve as a means of reconciling a more efficiency-oriented future agenda with a pattern of enforcement that prevailed until recently or as an indicator of de facto attention to the welfare of rivals. Merger concerns loomed large in the US-EU competition policy coordination agreement of 1991.147 Following the Boeing McDonnell-Douglas and GE Honeywell disputes, DOJ, FTC, and DG Comp developed a short document called “Best Practices on Cooperation in Merger Investigations” in 2002; a somewhat revised version appeared in 2011.148 The document calls for maximum feasible cooperation at every stage of a merger investigation, noting that waivers of confidentiality by the putative merging parties can greatly facilitate such cooperation. No substantial merger disputes have arisen since such cooperation was so specified.149 In addition, while studies based on stock market movements confirmed that the widespread view that EU merger regulation discriminated against foreigners through the 1990s, the data reveal no such discrimination in the period since the early years of the new century.150

4

Price Discrimination An Introduction to the Following Five Chapters In the next several chapters, we examine the antitrust law of the United States and the European Union on price discrimination, predatory pricing, exclusive-supply contracts, loyalty rebates, and bundled discounts. We will find that the concepts employed in the two jurisdictions are often similar, even though the two jurisdictions often reach different results. We will point out that the antitrust analyses of these several subjects is sometimes overlapping, that the analysis of predatory pricing in both jurisdictions is used as a basis for evaluating loyalty and bundled discounts and may enter into their evaluation of other restraints, that the different approaches of the two jurisdictions to exclusive-supply contracts also carry over to their approaches to loyalty rebates and bundled discounts, and that a jurisdiction’s approach to exclusive-supply contracts may be affected by its approach to price discrimination and to predatory pricing. Initially we wish to draw attention to several overlapping analytical points that will be developed in the following chapters; these involve predatory pricing, price discrimination, and exclusive-supply contracts. Predatory pricing occurs when a firm sells below cost in order to drive its rivals out of the market or to discipline them for deviating from a pattern of supracompetitive prices.

chapter four

64

Robert Bork understood predatory pricing to be a rare event, and the US Supreme Court has repeatedly cited Bork on its rarity. More recent theoretical work and empirical evidence has challenged this assumption. However rare such pricing may or may not be, predatory pricing analysis underlies much antitrust law. Although American and European predatory pricing analyses differ significantly, there are also major similarities. In both the United States and the European Union, pricing below cost is part of predatory pricing analysis. Both jurisdictions exhibit some uncertainly about which alternative measures of cost are appropriate for these purposes. A major part of predatory pricing analysis involves the “as efficient competitor” standard. A firm that sells below cost interferes with the competitive process because selling below cost can drive out equally or more efficient rival sellers. A seller cannot drive an equally or more efficient firm from the market, however, by prices that equal or exceed its own costs. The “as efficient competitor” test also forms part of the analysis that antitrust authorities in both the United States and the European Union employ in different but related areas of antitrust concern, such as those involving loyalty rebates, bundled discounts, and exclusive dealing. Antitrust scholars such as Richard Posner have suggested the use of the “as efficient competitor” test as a guide to evaluating behavior under Section 2 of the Sherman Act. Should predatory pricing analysis or its embodiment in the “as efficient competitor” test be used as the ultimate criterion of lawfulness throughout a wide range of behavior beyond predatory pricing? The Third Circuit, in LePage’s Inc. v. 3M Corp., thought not. That court determined that behavior— bundled discounts in the case before it— could be ruled unlawful, even though the defendant’s behavior did not involve predatory pricing, and it was not established that it failed the “as efficient competitor” test. By contrast, the Antitrust Modernization Commission in proposing antitrust standards for evaluating bundled discounts drew on predatory pricing analysis and the “as efficient competitor” test. The Sixth Circuit has employed the “as efficient competitor” test to assess the lawfulness of exclusive-supply contracts. To what extent are these US institutions implicitly saying that the “as efficient competitor” test is, or should be, the exclusive determinant of lawfulness? The European Commission has also incorporated the “as efficient competitor” test in its guidance1 for abusive behavior, but it appears to treat that test as a nonexclusive guide to legality. The scope of the “as efficient competitor” test, that is, the degree to which it is accepted as an exclusive determinant of lawfulness, is one of the most pressing issues in antitrust. Because of its importance and because of divisions of opinion on the role of the test, it would not be surprising if the United States and the European Union arrived at different positions. Be-

price discrimination

65

cause that test plays a part in the evaluations of a number of subject-matter areas, different views in the United States and the European Union about the exclusiveness of that test can help to explain divergence between the jurisdictions in those areas. Price discrimination is another subject-matter area whose analytical approaches carry over to other areas. Economists have tended to view price discrimination that involves selling at different marginal prices to different buyers through the lens of static analysis, where it is seen to generate resource misallocations and attendant deadweight losses. A dynamic analysis changes these results. The freer a firm is to set price, the more it is able to match or undercut prices of rivals. This freedom to price is likely to foster the greatest degree of price competition possible, driving down prices and expanding welfare. American and European antitrust currently take different approaches to price discrimination. In the United States, price discrimination is primarily governed by Section 2 of the Clayton Act as amended by the RobinsonPatman Act. At mid-twentieth century, this legislation was construed to erect widespread barriers to price discrimination, both on the primary line where the discriminator’s rivals feel the effects and on the secondary and tertiary lines where the discriminator’s customers or their customers feel the effects. At this time, the American law reflected traditional static economic analysis that saw price discrimination as welfare reducing. In the last half of the twentieth century, however, American law shifted. Under the Supreme Court’s 1993 Brooke Group decision, no pricing, discriminatory or otherwise, can be challenged for its primary-line effects unless it is predatory. Lower courts dealing with effects on the secondary and tertiary lines have begun to reinterpret the language of the Robinson-Patman Act to increase the importance of market-wide effects and to decrease the importance of effects on particular dealers. In Europe, antitrust has taken a different direction. Article 102(c) contains a provision that is designed, like the US Robinson-Patman Act, to protect a supplier’s customers from being subjected to competition from rivals who have received more favorable supply prices. Thus during the 1960s, both jurisdictions followed similar approaches to price discrimination generating effects on the secondary line. But while the United States drastically reduced the Robinson-Patman Act’s impact on protecting direct competitors in recent decades, the European approach to primary-line price discrimination has remained significantly more protective of rivals than the US (Brooke Group) approach. As will be shown in the following chapters, the different approaches to price discrimination may also help to explain differences in other antitrust areas.

chapter four

66

Finally, an economically rational seller, that is, a seller seeking to maximize its profits, possesses an incentive to lower price whenever the profit from its increased sales exceeds any loss of accumulated margin on its existing sales base. This is particularly relevant to evaluating exclusive-supply contracts, but it also applies to price discrimination and discounting. A wide variety of behaviors are explainable by sellers’ incentives to reduce price in order to increase the quantities sold. Some such behaviors have troubled antitrust enforcement authorities in the past and some still do. First, to the extent that a seller can identify different segments of its demand and can market its product separately to those several segments (without raising the danger of arbitrage), it will do so. It will charge the profitmaximizing price to each segment, thereby engaging in price discrimination. Second, it will often offer goods at lower prices to those who will buy larger quantities. Sometimes larger volume reduces cost, as large-volume sales generally avoid some selling expenses or shipping costs, and because large volumes facilitate planning, they may reduce manufacturing costs. But the incentive to reduce prices on transactions involving large volumes is not solely related to those reduced costs. Rather, it is a function of bargaining between the seller and large customers. The greater volume a buyer is willing to take, the greater is the seller’s incentive to lower the price in order to make the sale. As the volume increases, the additional sales compensate the seller for the loss of margin on its preexisting sales base: the greater the sales to a particular buyer, the lower the price the seller is willing to take to clinch the sale. These considerations are relevant to exclusive-supply agreements. If a buyer that has previously purchased 40 percent of its needs for a product from a particular seller is willing to increase its purchase volume, it may be able to secure a lower price by doing so, if the larger volume more than compensates the seller for its loss of margin on the preexisting sales volume. Such considerations, among many others, may contribute to exclusive contracts, and the lower buyer prices in a competitive reselling environment will lower final purchaser prices. All price-cutting to resellers does not result from cost saving, marketwide price discrimination, or usual bilateral bargaining. Benjamin Klein and Kevin Murphy 2 have explained how exclusive-supply or increased-supply arrangements with dealers generate incentives on suppliers to reduce prices that otherwise would be lacking. The distributor may function essentially as an agent for its customers as a group by overriding some preference for variety to make larger volume offers to suppliers in response to lower prices from them: the demand the suppliers face becomes more elastic and the wholesale price drops. Klein and Murphy further explain how buyers may

price discrimination

67

be able to play off suppliers against one another to the benefit of the final purchaser. They concede, however, that this mechanism does not obviate usual concerns about the foreclosure of distribution to efficient rivals. The law, both in the United States and in the European Union, has developed standards for evaluating the lawfulness of exclusive-supply contracts. These laws aim at situations in which a supplier is using exclusive-supply contracts to foreclose or impede rival suppliers from competing, so that the supplier can raise price and restrict output in the manner of a monopolist. We discuss some theories about how suppliers can use exclusive-supply contracts in this way. But all these theories are based on the assumption that the number of buyers (dealers or final users) is limited, and that significant entry barriers prevent new buyers from entering the market. Price Discrimination: An Economic and Legal Background The practice of selling the same good at different prices— generally referred to as price discrimination— has not fared well in the legal systems of modern economies, including the United States and the European Union. In the United States, the original Clayton Act 3 and its later amendments attacked price discrimination.4 During the era of transportation regulation, various laws and regulations that governed railroad, motor vehicle, and air transport rates targeted price discrimination.5 In Europe, Article 102(c) of the TFEU assails price discrimination.6 Those jurisdictions and scores of others also target price discrimination in international trade through antidumping and other trade laws.7 Is price discrimination the social evil these laws appear to assume? This is not a new issue; lawyers and economists have been concerned with it for decades.8 Lawyers— like the public at large— traditionally have tended to see price discrimination as “unfair” and therefore objectionable. In sharp contrast, economists, following Arthur Pigou, have tended to view price discrimination most often through the lens of static monopoly analysis, where price discrimination is seen as prone to generating resource misallocations, with attendant deadweight losses.9 Under Pigou’s third-degree price discrimination, where buyers face differing but fixed price, the only difference with linear demand curves is a diversion of sales from purchasers who value the product more to purchasers who value the product less— unless buyers are brought into the market by discrimination who are completely shut out absent discrimination. Therefore, price discrimination may well reduce welfare. Overall, much of the highly developed economic literature demonstrates the potential of price discrimination to transform consumer surplus into seller profit— transformations that (under static analysis) re-

chapter four

68

duce consumer welfare and often reduce aggregate welfare. Thus under a static monopoly perspective, even current US antitrust law, which is now widely understood to be designed to further either aggregate welfare or consumer welfare, should forbid or discourage price discrimination. Thus from a static monopoly perspective, both lawyers and economists can view price discrimination as antisocial: it is unfair, and it often reduces welfare. Results change dramatically in a more realistic dynamic analysis. Prohibitions on price discrimination introduce rigidity to pricing decisions. The freer a firm is to set price, the more it is able to match or undercut prices of rivals. This freedom to price as each firm thinks best serves its own interest and is likely to foster the greatest degree of price competition possible, thus expanding welfare. These considerations have generated reevaluations of price discrimination among economists. Since the antitrust revolution of the 1970s, lawyers, too, have increasingly focused on the potential of price discrimination to intensify price competition and to foster consumer and aggregate welfare. Thus a dynamic analysis can change the perspectives of both lawyers and economists and unite them in an appreciation of the potential of price discrimination to foster competitive behavior. George Stigler provided a prime source for economists’ attention to the potential for price discrimination to stimulate enhanced competition is his 1964 article on oligopoly.10 Stigler presented a theory tying the vulnerability of oligopolistic pricing to the market-share dispersion of its members because small sellers can cheat with less chance of detection. As Stigler argued, oligopolistic pricing breaks down when members depart from the target pricing levels that its members have tacitly agreed upon. While it is in the collective interest of its members to maintain the oligopolistic price, each firm has an incentive to increase its profits by increasing its own sales, if it can do so without immediately undermining the oligopoly. In any particular case, therefore, the question is whether an oligopolist can reduce its price in order to make an attractive sale while keeping information about its pricecutting from the other sellers. If the firm shades prices to some buyers and not to others, this is price discrimination as price competition. In real situations this is likely to increase welfare under a consumer surplus and under a total surplus standard.11 Stigler also points out that if the probabilities of detection for any one undercutting sale are approximately the same, those sellers who concentrate their sales efforts on larger buyers maximize the ratio of their sales increase to the risk of detection.12 In Stigler’s words: It follows that oligopolistic collusion will often be effective against small buyers even when it is ineffective against large buyers. When

price discrimination

69

the oligopolists sell to numerous small retailers, for example, they will adhere to the agreed-upon price, even though they are cutting prices to larger chain stores and industrial buyers.13

Stigler presents a case in which practices that disadvantage smaller resellers serve the general welfare. Such selective price-cutting is a mechanism through which the behavior of sellers, seeking to expand their profits, is likely to lead to a general reduction in the price level. Competition policy should therefore encourage that practice. More recently, William Baumol and Daniel Swanson have argued that not only is price discrimination compatible with, and often favorable to, competition, but also that competition may compel its use for firm survival. They note that in industries characterized by high fixed costs, especially industries characterized by repeated high-fixed-cost investments, competition may compel producers to engage in differential pricing across customers.14 When arbitrage is impossible, third-degree price discrimination (selling at different prices to different buyers unrelated to the number of units purchased) is a way of increasing revenue relative to cost— that is, of increasing profit. Baumol and Swanson point out that when such industries are characterized by intense competition, each firm is likely to find its price structure under strong downward pressure, resulting in each firm engaging in price discrimination across “distinct and non-trade compartments” but earning zero profits in the long run.15 Legal Treatment of Price Discrimination in the United States The Original Clayton Act. In 1914 Congress passed the Clayton Act, which addressed price discrimination generally in Section 2.16 Responding to complaints about the behavior of the Standard Oil Company and the American Tobacco Company,17 Congress prohibited discrimination in prices where the effect was likely to lessen competition or tend toward monopoly. Committee reports from both the House and Senate describe these companies as having been engaging in behavior now called predatory pricing.18 These reports also asserted that these companies were supporting their predatorily low prices in some markets from monopoly revenues that they were earning in other markets.19 Scholars have since pointed out that this congressional understanding was flawed and that the companies probably were not in fact acting predatorily.20 Congress addressed discrimination again in the Antidumping Act of 1916.21 In that legislation, and in subsequent antidumping legislation beginning with the Antidumping Act of 1921, Congress sought to prohibit foreign

chapter four

70

producers from selling their goods in the United States at lower prices than those sellers were charging in their home markets.22 In the 1916 act, the prohibition took effect only upon proof that the seller possessed a predatory intent.23 Later legislation eliminated that intent requirement.24 The legislative history of the early twentieth-century antidumping acts shows that Congress viewed the practice of foreign firms selling in the United States at prices below their home-market prices as unfair.25 In the two decades following the end of World War I, Congress revisited domestic and international price discrimination. Congress’s failure to include a predatory intent element in the 1921 Antidumping Act resulted in a law whose sole object was to protect domestic producers from international competition.26 Under third-degree price discrimination, aggrieved final purchasers are presumably those facing higher prices than those offered elsewhere. Several antidumping acts, however, offered Americans the opposite: “fairness” to domestic producers required that consumers always paid top dollar. The Robinson-Patman Act: Secondary-Line Effects. The emergence of chain stores, principally in the grocery and drug-store sectors, and the onset of the Great Depression propelled Congress once against to side with producers against consumers when it enacted the Robinson-Patman Amendments to the Clayton Act in 1936.27 The proponents of the legislation were members of the United States Wholesale Grocers Association who were losing business to the new chains; they were joined by small retail grocery enterprises, druggists, and food brokers who found themselves in competition with the chains.28 The chains were efficiently run,29 and because they were able to purchase in bulk, they often were able to obtain their inventories at lower cost than were their more traditional competitors.30 Nevertheless, the FTC, after studying the operation of the chains, had issued a critical “Chain Store Report.”31 The “Report,” combined with the lobbying efforts of wholesalers, small retailers, and their trade associations, persuaded Congress to respond. Congress explicitly sought to eliminate a competitive advantage that it believed the large chain stores unfairly held over the traditional smaller retailers.32 Although the term unfair does not appear in the Robinson-Patman Act, it was intended to create a level playing field between the chains and their smaller competitors. As several courts have stressed, “it is fairness, as Congress perceives it, that Robinson Patman is all about.”33 During the incubation period of the Robinson-Patman Act, large-scale retailers such as The Great Atlantic & Pacific Tea Company (A&P) clearly benefited from both greater efficiency and superior bargaining power.34 Over the entire US economy, innovative resource savings in distribution during

price discrimination

71

this period rivaled other economic advances in quantitative importance.35 Moreover, to the extent that the low prices large-scale retailers obtained reflected cost differences, there was no price discrimination in an economic sense.36 Nevertheless, many saw superior bargaining power as an illegitimate advantage. John Kenneth Galbraith outlined the opposing position. He cited the buyer bargaining to lower purchase price as an example of “countervailing power” between large buyers and sellers.37 Strong and increasing retail competition passed most of the savings through to the final purchaser. The FTC aggressively enforced the Robinson-Patman Act until the 1970s. The Supreme Court’s 1948 decision in FTC v. Morton Salt Co. assisted both the FTC’s enforcement efforts and plaintiffs in private lawsuits by erecting a presumption of illegality whenever proven that a defendant supplier sold goods at different prices to competing merchants.38 Because both the bargaining power of chains purchasing in bulk and distribution cost savings drew concessions from suppliers,39 the act protected small retailers at the expense of consumers. Primary-Line Effects. The original version of Section 2 of the Clayton Act

was directed against predatory pricing, but the addition of language in the Robinson-Patman Act expanded Section 2’s focus to include a concern with protecting the reselling customers of the discriminating seller and their own customer resellers. The Robinson-Patman Act also bolstered the Clayton Act’s original concern with protecting the rivals of the discriminating seller from the impact of its low prices. Indeed, in the three decades after the enactment of the Robinson-Patman Act, the FTC and the courts began to direct Section 2 against price discrimination that primarily affected the seller’s rivals, even when the seller was not acting predatorily as Congress envisioned in 1914.40 As amended, Section 2 prohibited primary-line effects: nonpredatory price discrimination that adversely affected the rivals of the discriminating seller. This expansion in scope was related to the confusing language that Congress employed in the Robinson-Patman Act, which made price discrimination unlawful where the discrimination might “injure, destroy, or prevent competition with any person who either grants or knowingly receives the benefit of such discrimination, or with customers of either of them.”41 This language was intended to prohibit discrimination that imposed a competitive disadvantage upon the customers of the discriminating seller.42 The choice of words was particularly unfortunate, however, because in referring to injuring “competition” with the customer, the provision appears to equate harm to the customer with harm to competition. In so do-

chapter four

72

ing, the wording assigns a meaning to “competition” that is significantly at variance with common understanding. By an analogous construction of the statutory language, the reference to injuring, destroying, or preventing competition with “any person who grants . . . such discrimination” would equate harm to rivals of the discriminating seller with harm to competition, thereby making such discrimination unlawful. For a time extending into the late 1960s, this construction appears to have influenced the way the courts approached claims of primary-line injury.43 As antitrust observers have long remarked, competition— as business persons, economists, and the public at large generally understand it— involves business firms attempting to take sales away from their competitors by undercutting them or surpassing them in the quality or attractiveness of their products.44 Whenever a firm succeeds in taking business away from one of its rivals, it has “harmed” or “injured” that rival. Injuring rivals by diverting business away from them is competitive activity par excellence. Yet the FTC and the courts soon found exactly that kind of activity unlawful under the Robinson-Patman Act. Within the first decade after the Robinson-Patman Act was enacted, the FTC ruled that a company had unlawfully discriminated because its lower prices “tended to divert trade to the respondent from its competitors.”45 On review of the FTC’s order, the US Court of Appeals for the Second Circuit upheld the FTC, asserting “that these findings supported the [FTC’s] order is too obvious to admit of discussion.”46 In the course of its opinion, the court said that the statutory language “no doubt mean[s] that the lower price must prevent, or tend to prevent, competitors from taking business away from the merchant which they might have got, had the merchant not lowered his price below what he was charging elsewhere.”47 Under the Second Circuit’s approach, proof that a seller offered two prices was sufficient to raise a presumption of unlawfulness.48 The defendant seller could overcome that presumption by proving its low prices did not in fact divert sales away from its competitors, or otherwise bringing itself within one of the act’s defenses.49 Although other circuits did not always deem any diversion of sales to be unlawful, along with the FTC, they tended to find a seller’s discriminatorily low prices increasingly problematic, as those prices deeply undercut the seller’s rivals and significantly altered market shares.50 Thus the FTC’s 1957 ruling against Anheuser-Busch’s localized price reduction in Saint Louis took this approach.51 In that case, the FTC equated a substantial diversion of business with competitive harm: No other circumstance [than the discriminatorily low price] will account for the fact that, while respondent more than tripled its sales,

price discrimination

73

most of its competition suffered such serious declines. This almost speaks for itself. Respondent’s gains could only have been made at the expense of competition since the total sales in the St. Louis market did not increase by any such substantial amount as the sales of respondent and the small combined increase in sales by all of the other competitors could not begin to account for the losses experienced by Falstaff, G. B. and G. W. Respondent’s price discriminations manifestly resulted in a substantial diversion of sales from competitors to itself.52

This use of the act to protect competitors became especially perverse when the FTC and the courts condemned local price reductions that reflected scale economies. In several cases, a business firm that reduced local prices in order to increase its sales from a plant with significant scale economies was condemned under the Robinson-Patman Act.53 In these cases, courts equated the changes in the local market share that resulted from the seller’s high-volume, low-price sales with injury to the seller’s rivals and thence with harm to competition. In 1967 the use of the Robinson-Patman Act to protect rivals of a discriminating seller reached its apogee in the now infamous case of Utah Pie Co. v. Continental Baking Co.54 There, the Supreme Court construed the act to protect a local supplier against a national rival’s attempt to enlarge its local sales by geographically limited price reductions. The Court referred to “radical price cuts” and “drastically declining price structure” as indicative of competitive harm.55 In reality, however, the localized price cuts gave rise to a period of intense price competition that eroded the market share of the locally dominant seller (who nonetheless continued to operate profitably throughout the period in question) and substantially expanded the total volume of product sold by all companies.56 The Retrenchment of Robinson- Patman to Secondary-Line Effects. The Robinson-Patman Act came into disfavor with antitrust revolution of the 1970s, which brought the courts to focus the antitrust laws on efficiency, and thus the generation of income and wealth, rather than on producerprotecting fairness or even rivalry.57 In its Brooke-Group Ltd. v. Brown & Williamson Tobacco Corp. decision of 1993, the Supreme Court reconsidered the structure of Section 2 of the Clayton Act.58 The Court concluded that under the Robinson-Patman Act, cases concerned with so-called primaryline harm must meet the same standards for proof of predatory pricing as do cases under the Sherman Act.59 Although the Court did not parse Section 2’s language, it effectively held that it contains two sets of provisions.

chapter four

74

The first set dates from 1914 and is directed only against predatory pricing. This interpretation is likely based upon the apparent intent of Congress at the time of enactment.60 The second set of provisions is composed of the language directed against discrimination affecting resellers contained in the Robinson-Patman Act and concerns only secondary-line effects. This language is confined to that of Robinson-Patman Act’s main objective: deterring discrimination that disadvantages some business firms purchasing from a discriminating seller vis-à-vis their rivals. Even under this new approach to the interpretation of the RobinsonPatman Act, its anticompetitive potential remains substantial. Primary-line harm is no longer a matter of concern unless discriminatorily low prices are below the measures of cost the courts employ to identify predatory pricing.61 Yet the act continues to make discrimination that disadvantages business customers vis-à-vis their rivals unlawful. Strict enforcement of the act would likely impose rigidity upon pricing that would discourage price competition and foster oligopolistic pricing behavior, effects that run counter to the procompetitive policies of the other antitrust laws. Indeed, the Supreme Court has always recognized the possibility of conflict between the RobinsonPatman Act and the Sherman Act; this underlies its indication that in cases of conflict, the procompetitive policies of the Sherman Act should prevail.62 The FTC no longer sees enforcement of the Robinson-Patman Act as a priority.63 Recently, the Antitrust Modernization Commission has called for its repeal because it hinders competitive behavior. 64 Although courts continue to apply the act, they construe it narrowly. Indeed, a number of lower courts have taken new interpretative approaches that have breathed elements of flexibility into the act.65 As a result, few plaintiffs successfully recover under the Robinson-Patman Act.66 Nevertheless, the act continues to burden the handful of unlucky firms targeted as defendants with substantial litigation costs.67 Price Discrimination and the Law in the European Union Price Discrimination and Article 102(c) of the TFEU. Articles 101 and 102

both contain provisions that target price discrimination. Article 101(1)(d) specifically bars agreements that “apply dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage.”68 Article 102 uses virtually identical language to prohibit dominant firms from “applying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage.”69 On their face, these provisions appear to prohibit price discrimination by a supplier between two competing purchasers because a

price discrimination

75

higher price to one dealer might well be deemed to competitively disadvantage it vis-à-vis the other. The focus of these provisions is thus upon protecting purchasers from what in the United States is termed secondary-line harm under the Robinson-Patman Act. It is not at all surprising that the TFEU would incorporate policies similar to those of the Robinson-Patman Act because the drafters of the antitrust provisions of the Treaty drew freely from the models provided by American antitrust law, despite much content that was uniquely European. Moreover, as we noted in chapter 1, until the US antitrust revolution, the decisions of the European Commission and of the Court of Justice embodied policies that resembled those underlying the US decisions of that period.70 During drafting of the EC Treaty of Rome in the late 1950s, the RobinsonPatman Act was widely seen as a major component of US antitrust law, and enforcement action by the FTC was large and growing. The FTC and courts repeatedly construed the act to prohibit price discrimination that would confer “competitive advantages” on favored buyers.71 The widespread public acceptance in Europe of the policy goals of the Robinson-Patman Act may have been recognized by the drafters of the competition law provisions of the TFEU. In Europe, however, the provisions of Article 101(1)(d) have a more limited application than the analogous provisions of the Robinson-Patman Act. While the US law extends to all sales, the coverage of Article 101 is limited to agreements or other concerted action. Given that European antitrust authorities consider ordinary sales transactions unilateral actions, they do not fall within the scope of Article 101.72 European authorities concentrate their concern about price discrimination largely on sales by dominant firms under Article 102. While this emphasis appears to be an advance over the American approach, which has taken no account of the size or prominence of the discriminating seller, the modest share threshold for dominance still means that price discrimination among competing customers by myriad large and successful firms is potentially vulnerable to attack under Article 102. Europe offers a rich case law on price discrimination as one form of abuse of a dominant position. In United Brands Co. v. Commission of the European Communities of 1978,73 a foundational Article 102 case, the Court of Justice ruled that United Brands contravened Article 102(c) by selling bananas to several national distributors at different prices. Because those distributors each resold their banana inventories in different local markets,74 they were in fact not in competition with one another, and thus none of them could be disadvantaged in competition with any of the others. It is unclear whether the ECJ failed to understand the competitive relationships among the distributors or whether the ruling was intended to show that

chapter four

76

competitive disadvantage was not required under Article 102(c) despite its language. The case is complicated by the ECJ’s mistaken interpretation of United Brand’s pricing practices as an intentional division of the banana market along national lines, a practice that, in the ECJ’s view, strikes at the heart of the single-market objective of the Treaty.75 In another foundational case, Hoffmann-La Roche & Co v. Commission of the European Communities (Hoffmann-La Roche) of 1979,76 the ECJ condemned so-called fidelity rebates as violations of Article 102(c).77 These consisted of rebates conditioned upon a purchaser’s buying all or a large percentage of its requirements from the seller. In that case, the ECJ was primarily concerned with the effects of fidelity rebates impeding the seller’s rivals from selling to its customers. The ECJ thus appeared to use Article 102(c) as a means for protecting the seller’s rivals, even though the Article is explicitly concerned with protecting the seller’s customers. Viewed from the perspective of the US Robinson-Patman Act, the ECJ focused on primary-line effects— effects on the rivals of the discriminating seller. The ECJ nonetheless employed a legal provision directed at secondary-line effects— effects on the seller’s customers— to support its condemnation of Hoffmann-La Roche’s discrimination.78 A significant part of EU price discrimination case law is concerned with protecting competitors of a dominant firm. The European courts are especially concerned with a dominant firm’s discount practices that impede rivals from selling to the dominant firm’s customers. These practices are considered exclusionary. They include “fidelity” or “loyalty” rebate systems, such as those involved in Hoffmann-La Roche (in which rebates are keyed to a customer purchasing a specified percentage of its requirements from a seller), as well as “target” or “objective” rebate systems in which rebates are keyed to the customer’s satisfaction of sales objectives set in absolute amounts. By contrast, so called quantity discounts— in which the price to all buyers is reduced on a uniform schedule as the quantity purchased increases— have generally been upheld as lawful, even when granted by a dominant firm. In its later decision in British Airways plc. v. Commission of the European Communities (British Airways), however, the General Court observed that the implicit justification for quantity discounts is that they reflect the lower unit costs often incurred by sellers in large-volume transactions.79 Accordingly, the General Court hinted that when a dominant firm’s criteria for granting such a rebate reveal that it is not cost related, the rebate may be viewed more like a fidelity or target rebate impeding rivals from selling to the dominant firm’s customers.80 Moreover, discount systems in which the discounts are computed on sales over a long reference period have been deemed to have similar exclusionary effects because the value of the discount

price discrimination

77

significantly increases over the length of the period and thus exerts growing pressure on the buyer to remain with its current supplier.81 These and related issues are considered again in chapters 6– 8. Although the Court of Justice has condemned fidelity and target rebates under Article 102(c), it has also condemned fidelity and target rebate systems as abuses under Article 102’s general language, without invoking clause (c).82 Indeed, in British Airways, the General Court asserted that a dominant supplier’s fidelity rebate system requiring customers to obtain their supplies exclusively or almost exclusively from that supplier is abusive and therefore in violation of the basic prohibition of Article 102.83 Additionally, that General Court also ruled that the rebate system was in fact discriminatory and imposed competitive disadvantages upon customers within the meaning of Article 102(c).84 In its 2007 judgment affirming this decision, the Court of Justice discussed and upheld both rulings.85 Commentators have criticized the European authorities for misusing Article 102(c) as a means for protecting the rivals of dominant firms.86 They contend that Article 102(c) should not apply absent of a showing that buyers have been placed at a competitive disadvantage. Instead, in many cases the court appears to be concerned with primary-line effects, patently not a matter dealt with by Article 102(c).87 Comparing Article 102(c) with the Robinson-Patman Act. Congress adopted the Robinson-Patman Act in response to the complaints of small business firms of unfair exposure to the competition of large chains stores able to obtain their supplies at lower prices than were their smaller rivals. Congress wanted to neutralize the bargaining power of the chains vis-à-vis their suppliers, who were frequently small- or medium-size companies.88 Although the Robinson-Patman Act directs most of its provisions against the discriminating sellers, its premise is that buying power is misused at the purchaser level. By contrast, Article 102(c) applies only to large sellers that can meet the criteria for “dominance” as used in the TFEU. Thus although both Article 102(c) and the Robinson-Patman Act were ostensibly designed to prevent buyers from being competitively disadvantaged, the two provisions actually direct their focus in opposite directions. Article 102(c) focuses on the pricing behavior of powerful sellers while the core concern of the RobinsonPatman Act is upon the purchasing power of powerful buyers. Although in earlier years the Robinson-Patman Act was also employed to protect the rivals of a discriminating seller, just as Article 102(c) is used today, current interpretations of Robinson-Patman protects a seller’s rivals only against predatory pricing.89 Overall, the dominant thrusts of US and European legal concerns about

chapter four

78

price discrimination aim in different directions. US law represents a historic desire to protect small retail merchants from the competition of powerful buyers. European law, on the other hand, appears focused on protecting initial sellers from the competition of their powerful, and inevitably often nonnational, rivals. EU Price Discrimination beyond the Context of Article 102(c). Article 102(c) is not the only provision in Article 102 that European authorities have used to target price discrimination. It is but one of four clauses that describe particular types of behavior that fall within that Article’s general prohibition against abuses of dominant position. And the structure of Article 102 makes clear that abuse can take forms other than those referred to in the four clauses. As indicated above, the European Commission and the European courts have targeted price discrimination as an abuse under Article 102’s general clause. In doing so, these authorities have frequently described this abuse as involving “selective” price cuts, a phrase that is literally coextensive with all price discrimination.90 European authorities have seen selective price-cutting as subject to the prohibition against “abuses” of a dominant position because they have viewed it as a tool for deterring entry by rivals or for forcing them to exit the market, and thus as a device for obtaining or preserving a dominant position.91 As discussed below, European authorities appear to have understood at least some forms of selective price-cutting as posing a threat to a competitive market structure analogous to the European approach to predatory pricing involving prices above average variable cost and below average total cost. In the EU case law, both predatory pricing and selective price-cutting have been seen as fostering dominance and thus as a threat to competitive market conditions. In Akzo, a leading case from 1991, the abuse consisted of both predatory pricing and selective price cuts.92 Because the selective price cuts were at below-cost levels, the independent significance of selective price-cutting was unclear. In the Hilti case later the same year,93 however, the General Court clarified that issue. In that case, the Commission had earlier characterized selective price-cutting as abusive, without hinging that characterization on whether the prices were below cost.94 Thus when the General Court affirmed the Commission’s decision it appeared to uphold the Commission’s position that above-cost selective price-cutting was abusive. In Compagnie Maritime Belge of 2000 the ECJ reinforced this understanding when it held that a shipping conference (with 90 percent of the relevant market) abused its dominant position when it employed “fighting

price discrimination

79

ships” to offer carriage at reduced rates in competition with rivals, although the reduced rates were not shown to have been below cost.95 More recently, the Court of Justice has reviewed an allegation of abuse involving selective price-cutting in Post Danmark A/S v. Konkurrencerådet,96 a 2012 decision of the Court of Justice. Post Danmark involved a reference to the ECJ by the Danish Højesteret. The case involved the Danish post office, which was competing with a rival (Forbruger-Kontakt) in the distribution of unaddressed mail. Both Post Danmark and Forbruger-Kontakt performed this task by contracting with distributors. Post Danmark offered prices to former customers of Forbruger-Kontakt that were lower than the prices offered to its own preexisting customers, behavior similar to that in Akzo and Hilti. Moreover, because the prices offered were lower than Post Danmark’s average total costs, the case also raised predatory pricing issues. This, of course, was an issue discussed by the Court of Justice in Akzo. The ECJ ruled in favor of Post Danmark. It acknowledged that documents before the ECJ showed that the complaint “was based on the suggestion that Post Danmark, by a policy of low prices directed at certain of its competitor’s major customers, might drive that competitor from the market in question.” But it also recognized that in the proceedings before the Danish tribunal, it “could not be established that Post Danmark had deliberately sought to drive out that competitor.”97 In the circumstances, the ECJ stated that price discrimination “cannot of itself suggest that there exists an exclusionary abuse.”98 On the predatory pricing issue, the ECJ observed that the Danish tribunal used an incremental-cost framework, different from the Akzo (average-variable cost) framework, to carry out a price-cost comparison. The ECJ nonetheless accepted the incremental-cost framework of the Danish tribunal, using incremental cost as a substitute for average variable cost in the ECJ’s Akzo line of decisions. The ECJ then ruled that pricing behavior “cannot be considered to amount to an exclusionary abuse simply because the price charged to a single customer by a dominant undertaking is lower than the average total costs attributed to the activity concerned, but higher than the average incremental costs pertaining to the latter.”99 The ECJ’s rulings, on the selective-pricing issue and on the predatory pricing issue, appear to be based on intent, a matter more fully discussed below. The European Commission, which instituted the cases against Akzo, Hilti, Irish Sugar, and Compagnie Maritime Belge, all of which involved selective price cuts, appears to have remained concerned with price discrimination (including selective price cuts) at least as late as 2005, when its discussion paper on Article 82 [102] was issued. That paper contains several references to price discrimination and selective price-cutting,100 but the

chapter four

80

Commission’s guidance (which was based on the discussion paper), does not mention either price discrimination or selective price-cutting explicitly. The guidance does, however, refer to the “selectivity” of the conduct in question in its discussion of standards relating to all (price and nonprice) conduct. In this passage, the guidance cites Irish Sugar as a reference for selectivity.101 So there is ground for believing that the commission remains hostile to at least some “selective” price cuts. All of the cases just discussed reflect the way the European Commission and the European courts analyze the dominant firm’s intention. Intent plays a role in the predatory pricing cases, and these cases are instructive about how the EU authorities approach selective price-cutting. The Court of Justice has identified two possible types of predatory pricing. The first type occurs when a dominant seller offers its goods at prices below average variable cost.102 The Court of Justice views such behavior as unambiguously predatory and hence abusive. The second type of possibly predatory pricing occurs when a dominant seller offers its goods at prices that are below average total cost but above average variable cost. Courts do not treat this behavior as presumptively predatory because there are legitimate economic rationales for such behavior. Such sales, for example, can minimize losses in a situation of falling demand. In the case of above-average-variable-cost pricing, there must be proof that the sales were part of a plan to eliminate a competitor before they will be deemed an abuse.103 Thus according to the Court of Justice, the second category of pricing is ambiguous and there must be evidence of the seller’s intent before the pricing can be condemned as abusive. The ECJ’s insistence upon evidence of intent as a means of resolving the ambiguous nature of the firm’s behavior in the predatory context has been repeated in the context of discriminatory pricing challenged under the general clause of Article 102. In those cases in which the ECJ has condemned selective price-cutting, the firm in question has directed its price-cutting toward customers of one or more rivals.104 In the view of the Court of Justice and other EU authorities, this targeting has revealed the dominant seller’s intention to injure the particular rivals that are threatening its dominance. With the actor’s intent seemingly clarified, its actions have been treated as abuses, forbidden by Article 102. Thus in Akzo, the Court of Justice construed the company’s selectively low prices to the customers of its rival Engineering and Chemical Supplies (Epson and Gloucester) Ltd. (ECS) as evidence of its intention “to adopt a strategy that could damage ECS” and thus constitute abuse.105 European competition law has treated selective price-cutting and predatory pricing similarly because both practices have been seen to threaten the

price discrimination

81

maintenance of a competitive market structure. Yet the danger in prohibiting various forms of low pricing is that unless the prohibitions narrowly embrace only unambiguously anticompetitive behavior, the prohibitions themselves can create price umbrellas under which inefficient sellers receive protection from legitimate price competition. This result conflicts with the core purpose of competition law. The EU prohibition against selling at below-average-variable-cost prices (albeit somewhat different from the full US predatory pricing approach) is finely tuned to target anticompetitive behavior. The Court of Justice correctly states that such pricing has no legitimate economic rationale.106 When the EU authorities target selective price-cutting, however, they cannot claim to be limiting their sanctions to behavior that is unambiguously anticompetitive. Indeed, when a firm lowers its price to respond to a rival’s incursions on its market, that behavior constitutes the very price-competitive behavior that competition laws are designed to foster. The EU authorities thus have been operating with a false dichotomy. In the arena of market competition, it is impossible to draw a distinction between intent to take sales away from a rival and intent to injure the rival by doing so. Evidence that a dominant firm intended to injure a rival by diverting sales away from it, therefore, is nothing more than evidence of intent to compete. The underlying flaw in the traditional EU analysis lies in the premise that price competition is legitimate when market-wide and uniform, but that price reductions targeted to the areas of intense rivalry are suspect. Post Danmark suggests a possible reformulation of that premise in which extrinsic evidence to eliminate a competitor, not the mere targeting of a rival’s customers, is requisite to a finding of abuse. Even with the broadest interpretation of Post Danmark, however, the EU approach to price discrimination remains significantly different from the US approach in its apparent failure to appreciate price discrimination’s procompetitive potential. Many American antitrust observers would view the attempts by EU authorities to distinguish between competition on the basis of market-wide pricing and competition employing selective price reductions as an unfortunate repetition of the US experience under the Robinson-Patman Act during the middle of the twentieth century before the Supreme Court reinterpreted the that act as no longer protecting firms from the price competition of rivals. When courts construe Article 102 of the Treaty to prohibit selective price-cutting— just as when courts across the Atlantic were construing the Robinson-Patman Act to target nonpredatory primary-line injury— they do so in pursuit of a policy of “fairness” to rivals of the favored seller. That “fairness,” however, comes at the expense of consumer or total welfare and therefore arguably at the expense of society. Insofar as selective price-cutting

chapter four

82

operates as a mechanism for breaking down supracompetitive pricing, the social cost of “fairness” is a reinforcement of anticompetitive pricing and a reduction of social welfare. As in many other realms, the political feasibility of cleaving ever closer to a welfare standard in judging price discrimination is easier in the United States than in Europe. In Europe, competition-law authorities often appear more concerned with the welfare of incumbent economic actors than with the competitive process itself. Price Discrimination: An Overall Assessment This chapter argues that price discrimination is an important element in competition and does not deserve to be viewed with suspicion. Firms in perfect competition cannot engage in price discrimination, but they cannot sell differentiated products, either. When firms do sell differentiated products, they face downward-sloping demand curves, which imply some discretion over price. That discretion may be used differentially across units sold, and across purchasers, yielding price discrimination. Yet profits may be only normal or even negative. This situation approximates the current predicament of the airline industry. More typically firms in industries that might otherwise have quasi-collusive excess profits based on entry barriers and recognized mutual dependence can be destabilized by the ability of participants to nibble at one another’s markets through selective price competition rather than charging prices such that no purchasers receive any better deal than any others. Finally, in some cases, firms may well use targeted discrimination to hinder the competitive progress of rivals who would benefit if they could not be singled out for special attack. This behavior can include certain kinds of discounts and rebates. They deserve special scrutiny from competition authorities, and they will be considered in chapters 7 and 8. Rules about price discrimination should be evaluated in relation to their likelihood to improve welfare by either the consumer surplus or total surplus standard; only the former is now feasible in Europe and probably in the United States as well. The United States should repeal the RobinsonPatman Act, and the European Union should abandon its comprehensive prejudice against price discrimination. Both jurisdictions should critically explore claims that a seller’s discriminatory prices, which may adversely affect particular rivals of that seller, are damaging or are threatening competition in the market.

5

Predatory Pricing The paradigm case of predatory pricing involves a large seller pricing its goods below cost in order to drive its rivals from the market and to acquire a monopoly for itself. Variations include pricing below cost in order to discipline rivals who have undercut supracompetitive oligopoly prices, where below-cost pricing is intended to coerce the nonconforming firms back into oligopoly compliance. Predatory pricing is unlawful in both the United States and the European Union, although the two jurisdictions employ somewhat different approaches to that issue. Predatory Pricing in the United States Section 2 of the Sherman Act treats predatory pricing as monopolization when practiced by a monopolist and as attempted monopolization when practiced by a near monopolist. It can also constitute a violation of Section 2 of the Clayton Act when it is accompanied by price discrimination. Indeed, the first official recognition of the practice appears in the legislative history of the Clayton Act in 1914. Both the Senate and House reports on the original section of that act describe the practices of Standard Oil and American Tobacco as “lowering the price of their commodities, oftentimes below the cost of production in certain communities or sec-

chapter five

84

tions where they had competition, with the intent to destroy and make unprofitable the business of their competitors, and with the ultimate purpose in view thereby of acquiring a monopoly in the particular locality or section in which the discriminating price is made. Every concern that engages in this pernicious practice must necessarily recoup its losses in the particular communities or sections where their commodities are sold below cost or without a fair profit by raising the price of this same class of commodities above their fair market value in other sections of communities.”1 It was not until the late twentieth century that the case law developed standards for distinguishing unlawful predatory pricing from lawful aggressive competitive pricing. Indeed, as late as 1967, the Supreme Court’s decision in Utah Pie Co. v. Continental Baking Co.2 revealed with embarrassing clarity that the Court was unable to distinguish aggressive pricing from predatory pricing. In 1974 Harvard professors Donald Turner and Phillip Areeda published “Predatory Pricing and Related Practices under Section 2 of the Sherman Act,”3 an article that was designed to help courts to make this distinction both accurately and at the pretrial stage of litigation. In their article, the authors stated that although they believed predatory pricing was a rare occurrence,4 they wished to provide tools for the courts to dismiss unmeritorious predatory pricing claims quickly. They reasoned that any price equal to, or above, short-run marginal cost should be treated as nonpredatory by the courts because marginal-cost pricing characterizes competitivemarket equilibrium and because no firm can drive more efficient rivals out of the market by pricing its own goods at (or above) its own marginal cost. Because it is notoriously difficult to determine marginal cost, however, Areeda and Turner suggested that courts use average variable cost as a surrogate for marginal cost. They also argued that predatory pricing requires the predator firm to recoup its losses from monopoly profits it earns from successful predation.5 Accordingly, they argued that any plaintiff asserting a predatory pricing claim should establish a likelihood of recoupment. The Areeda and Turner article stimulated proposals from other academics for alternative ways of identifying predatory pricing.6 Because of its simplicity, apparent ease of use, and the reputation of its authors, however, their test quickly gained the largest following. Shortly after the appearance of that article, several circuit courts embraced it, although with their own modifications. Thus the Ninth Circuit, contrary to the recommendations of Areeda and Turner, made the determination of which costs were fixed and which were variable a jury question. The Ninth and other circuits also treated the proposal’s use of average variable cost as the dividing line between predatory and nonpredatory pricing as presumptions, but as rebut-

predatory pricing

85

table presumptions. In Transamerica Computer Co. v. IBM Corp., the Ninth Circuit ruled that in some circumstances (such as limit pricing or entry deterrence) even prices above average total cost could be deemed predatory.7 That position was rejected quickly by the Sixth Circuit, which otherwise followed the Ninth Circuit’s version of the Areeda-Turner standard.8 The Eleventh Circuit, differing with the Ninth Circuit’s position in Transamerica Computer, took the position that prices above average total cost are per se lawful, but otherwise followed the Ninth and Sixth Circuits in ruling that prices between average total cost and average variable cost would be treated as presumptively lawful, requiring a plaintiff to produce evidence to rebut the presumption of lawfulness.9 This additional evidence could be either subjective or objective, and the evidence must be stronger the closer the defendant’s prices are to its average total costs.10 The Eleventh Circuit’s standard, although cast in different words, is essentially the standard the Ninth Circuit adopted for prices below average total cost. These Circuit Court approaches treating prices at levels between average variable cost and average total cost as only presumptively lawful but as vulnerable to a predation claim if the plaintiff musters sufficient additional evidence bear a resemblance to current predation rules in the European Union. In general those rules will treat such pricing as predatory if they are shown to be part of a plan to eliminate a competitor. Three years after the Areeda-Turner proposal, Robert Bork argued in his influential Antitrust Paradox that predatory pricing was so rare that it was unwise for the courts to develop rules for dealing with it in the manner Areeda and Turner proposed. Although the Supreme Court has effectively rejected Bork’s suggestion of not developing rules governing the subject, Bork’s skepticism over the existence of predatory pricing has influenced the Court. Indeed, in its 1986 Matsushita decision, its first predatory pricing decision after the publication of the Antitrust Paradox, the Court cited Bork on just this point.11 In its other decisions involving predatory pricing claims, the Court has also described predatory pricing as rare and highly risky for the predator.12 Indeed, in Matsushita, the Court, clearly influenced by Bork’s skepticism, raised the standard of proof applicable to such claims, making the economic plausibility of the claim a critical element and allowing the courts to evaluate that economic plausibility.13 Although many economists have come to the view that predatory pricing is much more common than Bork postulated, the Court has incorporated Bork’s position into its framework for dealing with predatory pricing. In its Brooke Group decision of 1993, the Court accepted the parties’ designation of the appropriate cost measure as average variable cost but

chapter five

86

also referred to the appropriate cost measure as “incremental” cost. Most antitrust observers have understood the Court’s reference to incremental cost as meaning marginal cost or to a closely connected short-run cost such as average variable cost or Baumol’s average avoidable cost. As we point out below, however, the Court’s reference is ambiguous and is open to an alternative interpretation. In Brooke Group, the Court indicated that prices at levels equal to or exceeding incremental cost are per se lawful. Under the common interpretation equating incremental cost with marginal cost or a close relative, Brooke Group modifies the predatory standards of the Ninth and Eleventh Circuits. Under the standards articulated in Brooke Group, the plaintiff must establish that the alleged predator has priced its goods below incremental cost and that there is a likelihood that the defendant will recoup its losses in the postpredation period. Predatory pricing cases had been in decline since the publication of the Areeda-Turner article in 1974. Brooke Group reaffirmed the recoupment requirement articulated in prior cases, and success in predatory pricing cases declined further.14 Profitable predatory pricing apparently depended on very special circumstances. Following the exclusion of one or more firms, the market structure must be favorable to a price increase: high concentration and high entry (or reentry) barriers and sufficient capacity by the predator to meet demand at the elevated prices. Many simple models focused on a single market make recoupment appear highly unlikely and therefore cast doubt on predation as a motive for the behavior observed. But when firms operate in multiple markets contexts, the calculations may look very different; aggressive behavior in one market may provide an effective warning across many others and make recoupment very likely.15 More recently, the Court has expanded the application of the Brooke Group standards to related issues, such as allegedly predatory bidding. In Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co.,16 the Court ruled that predatory bidding claims must be evaluated by the same standards as predatory pricing claims. In that case, the plaintiff sawmill operator alleged that the defendant rival sawmill operator had driven up the price of red alder logs by its excessive purchases or by paying higher prices than necessary. The Supreme Court analyzed the issue as predatory buying, which it saw as the mirror image of predatory pricing. So long as the defendant sells its output at prices that equal or exceed its costs, it is acting lawfully. The Court’s analysis was based on its assumption that the inputs the alleged predator purchased (logs) ultimately would be disposed of in the output market. On this assumption, the Brooke Group standard (requiring sales below incremental costs) can be applied. So long as the seller is earning rev-

predatory pricing

87

enues in the output market that covers its costs (including its allegedly high costs in the input market), there is no legally cognizable predation. Refining Predatory Pricing Standards. When the Supreme Court in Brooke

Group described incremental cost as an appropriate measure of cost for purposes of predatory pricing analysis, it implicitly recognized that the relevant cost should be computed for the period of the alleged predation. This was one of the flaws in the original formulation of the Areeda-Turner proposal that used average variable cost as the effective cost standard, since average variable cost is normally computed for a firm’s entire output. Moreover, the incremental cost for the predation period would not distinguish between fixed and variable cost as Areeda and Turner proposed but rather would include all of the costs that the alleged predator incurred to produce the predatory increment. In a highly influential article, William Baumol17 grappled with these and other issues and made several proposals for refining the Areeda-Turner standards. Areeda and Turner proposed a marginal cost standard as a theoretical ideal for separating predatory from nonpredatory pricing, and because of the impracticality of determining marginal cost they proposed using average variable cost as a “surrogate” for marginal cost. Baumol pointed out that Areeda and Turner were struggling to find ways of determining whether an alleged predator was or was not covering its out-of-pocket costs on the products it was selling at allegedly predatory prices. He suggested that “average avoidable cost” (AAC) on the allegedly predatory increment is superior to the average variable cost standard then used by most federal courts to assess predation. AAC reduces or eliminates problems surrounding which costs are variable and which are fixed because AAC includes all costs (whether fixed or variable) that could be avoided by not producing the allegedly predatory increment.18 Baumol also noted that AAC could be used to resolve the problem common costs presented, that is, costs common to the product at issue and other products. Because there is no nonarbitrary way to apportion common costs among products, the average variable cost of a particular product cannot reflect common costs without arbitrarily apportioning some portion of them to the product in issue. Here Baumol suggested identifying avoidable costs of the entire package of products to which the common costs related. Whether the avoidable cost of the package exceeded the revenue the package generated would help to determine whether the defendant was pricing predatorily. Under Baumol’s proposed approach, a plaintiff would first identify the product or service that he believes is being priced predatorily. The plaintiff

chapter five

88

would also identify any package of products or services (to which the common costs related) that he believes is being priced predatorily. The defendant then could defend on the ground that the revenues from both the product and the package exceeded their respective avoidable costs.19 The Department of Justice embraced Baumol’s suggestion. Although the DOJ case against American Airlines in 2003 was unsuccessful, the Tenth Circuit accepted the propriety of using an average avoidable cost standard.20 The DOJ embraced that standard in its 2008 report on single firm conduct,21 where it seemed to favor its use as an exonerating standard for predation. The Obama administration dissociated itself from that report in May of 2009.22 The EU commission has also adopted AAC in lieu of average variable cost although only as a marker for clear indication of predation. In other words, AAC has replaced average variable cost (or short-run marginal cost) as a signal for likely violation,23 but much other disagreement remains about the appropriate predatory pricing standard. Broadening the Scope for Predatory Pricing. The Supreme Court has, in sev-

eral opinions, cited Bork for the rarity of predatory pricing. In recent years, however, economists and legal practitioners have become increasingly aware that predatory pricing need not be limited to situations in which a firm forces the exit of its rivals by incurring large losses in a price war (the situation Bork envisioned). Patrick Bolton, Joseph Brodley, and Michael Riordan have influentially emphasized that besides that expensive form of predation, research by many scholars has identified other, less expensive forms of predation, such as financial-market predation, reputation predation, testmarket predation, and cost-signaling predation.24 Financial-market predation occurs when a predator exploits the reliance of its rival on financial institutions, forcing prices down and reducing the revenues of the rival, undermining the expectations of the financial institutions supplying its ongoing capital requirements.25 Reputation predation occurs when the predator acts as a classic predator in one market in order to establish a reputation as a predatory or aggressive competitor in other markets. In this way, the predator incurs some losses in the market of immediate concern but profits from deterring competition in the other markets, which in the aggregate may be many times larger than those in the market where the predatory pricing occurred. Test-market predation occurs when a predator forces prices down to unprofitable levels when a potential entrant is testing demand in the predator’s market. Successful test-market predation would either mislead the potential entrant into a false belief about demand conditions or would render its test marketing unusable. Cost-signaling predation occurs when the predator reduces price in order to mislead a rival or potential entrant

predatory pricing

89

into believing that it has attained costs that a rival could not meet, leading to the rival’s exit or lack of entry. Bolton and his colleagues argued that although the case law was capable of incorporating these new insights into predation, judges were failing to act upon them because they were still attached to the belief that predatory pricing was rare, a belief leading them to skepticism toward the new insights. This may be too harsh a judgment, as financial predation appears to have been recognized as early as 1984 in the Laker litigation.26 In any event, the Bolton, Brodley, and Riordan article has been widely read and discussed, bringing developments in predatory pricing theory to the attention of a wide audience. It is probably accurate to conclude that US courts now (more than a decade later) will recognize the newer dynamic forms of predation that article identified when cases involving them are brought. Bolton and his colleagues argue that pricing above long-run average incremental cost (LAIC, a calculable version of total cost) should insulate a firm from attack for predatory pricing, but that pricing below that level but above AAC ought to be treated as predatory in some cases. LAIC. In their advocacy of a broadened scope for predation analysis, Bolton, Brodley, and Riordan also urged a dramatic modification of the price-cost standard. As noted, the Supreme Court in Brooke Group had referred to incremental cost as the “appropriate measure of cost” for predation analysis. Bolton and his coauthors seized on the ambiguity of “incremental cost” to argue against an exclusively short-run interpretation of that term. In their view, long-run incremental cost would be a proper measure for industries like software, where most of the costs are incurred up front (in the research and development stage) and where short-run costs are negligible. If the Supreme Court were willing to treat its references to incremental cost as open to a long-run construction, then differences between the United States and the European Union over predatory pricing analysis would be significantly narrowed. Bolton, Brodley, and Riordan’s suggestion is (following Baumol) to treat pricing below AAC as presumptively predatory, to treat pricing between AAC and LAIC as presumptively lawful, and to treat prices above LAIC as conclusively lawful. If we treat LAIC and AAC, respectively, as equivalent to average total cost and average variable cost, we would have a formulation that would meet the position of the Court of Justice. The European Commission has already made this interpretive step, essentially accepting the Bolton, Brodley, and Riordan proposal.27 Several points should be made about the LAIC concept. First, use of the term incremental to refer to an average concept invites confusion. Economists typically use the word incremental to mean marginal, and average

chapter five

90

marginal is confusing. Nevertheless, Baumol used “average incremental cost” (AIC) to refer to the cost of any product-specific outlay that is caused by a specific output increment, including sunk outlays.28 Bolton, Brodley, and Riordan took the concept, acknowledged its use by Baumol, and called it “long-run average incremental cost,” or “LAIC,” on the ground that this designation specifically identifies the cost as long run.29 This approach makes an important contribution to useable economic theory by dealing with “nonmarginal” increments (or decrements), that is, substantial changes in output that correspond to particular situations and cost configurations.30 Second, there is an underemphasized distinction between recurrent and nonrecurrent costs. Much of the front-end expenditure on the development of a new product is nonrecurrent. Once these costs have been incurred, their strict duplication by other firms for exactly that product represents social waste. LAIC includes these nonrecurrent costs. This may create a tension between interfirm competition and social welfare; short-run marginal-cost pricing would immediately further social welfare. Bolton, Brodley, and Riordan deal with this issue by allowing for a defense of prices below LAIC on efficiency grounds. Third, much of the sunk costs of product development are not recorded as capital expenditures, and so LAIC can only be approximated. Further, even if product development were booked as investment, its appropriate amortization would remain problematic because the firm’s pricing strategy pattern depends on the behavior of other firms. The pace at which a firm should (be obliged to) recover such costs is therefore innately problematic. The public monitoring of such cost recovery is unpromising. But posing the problem in this form is misleading in dynamic industries, anyway, because a failure to continuously incur sunk development costs assures oblivion. Opportunity Costs. Among the most troublesome issues in the US approach

to predatory pricing is the matter of opportunity costs. US law is torn between an approach in which rules are based on modern economic analysis and a concern to make the rules as simple as possible, so that business firms can observe them and generalist judges can apply them. One area where this double goal of economics-based rules and simplicity of application generates tension is opportunity costs. Economists generally understand that a profit-maximizing firm takes opportunity costs into account when making a decision. The firm, in other words, is not merely deciding whether a particular course of action is profitable; it is deciding whether that course is more profitable than other mutually exclusive options. As applied to predatory pricing analysis, opportunity costs reflect the “sacrifice” of foregone revenue. So viewed, they help verify that the alleged

predatory pricing

91

predator is investing in the exit of the “prey” from the market, and these costs must be recoupable in order for the predatory scenario to be rational. Yet if opportunity costs are to be incorporated in predatory pricing analysis, then a determination of predation may require the fact finder to assess the likely profits available to the alleged predator from the use of resources elsewhere. Instead of a simple determination of whether the firm is selling below its (accounting) costs, the fact finder will have to assess the profitability of alternative behaviors and treat the sacrifice of those alternative profits as opportunity costs. Taking this perspective, the determination of whether the firm has sold below cost becomes a problem of managerial decision making. This approach turns the simplicity of the Areeda-Turner formulation into a complex problem that seems to challenge the decisional capabilities of juries and judges. This dilemma has generated different results from the courts. Most courts have rejected opportunity costs as an element in the determination of cost for predatory pricing analysis. The Ninth Circuit and some district courts31 have ruled that opportunity costs are not to be considered in assessing predation, while the Sixth Circuit has accepted evidence of opportunity costs as part of the determination of predation.32 Baumol has again provided a needed analytic refinement.33 If the focus were placed not on the alleged predator’s sacrifice but on the costs an equally efficient rival would incur, then the predator’s loss of revenues from alternative lines of activity become irrelevant. The relevant costs are those that a rival must match. In the American Airlines case,34 the government charged American Airlines with predatory pricing on certain routes out of Dallas/ Fort Worth when American diverted aircraft to those routes. This greatly expanded American’s capacity on those routes, severely disadvantaging rivals. But the revenue foregone by American as a result of diverting aircraft to the Dallas/ Fort Worth airport was irrelevant to the ability of a rival to compete with American from that airport. The relevant benchmark was the cost of aircraft leases. The American Airlines case also illustrates the difficulties generalist judges face trying to apply superficially simple predatory pricing rules that become complex in their application. The court indicated that it was persuaded by the above-cited Bolton, Brodley, and Riordan article that predatory pricing can no longer be considered a rare event, and it accepted Baumol’s suggestion that average avoidable cost was an acceptable proxy for marginal cost. Yet the court appears to have reached an incorrect result because it could not sort out an overlap of average variable and average avoidable cost in the government’s evidence about the cost of American’s capacity addition.35 In a number of recent decisions, the Court has indicated its awareness

chapter five

92

that its predatory pricing rules leave room for strategic behavior aimed at imposing unfavorable costs on rivals, and that sophisticated economic analysis might be able to identify that potential. But it has concluded that unless the defendant is selling output below cost and is likely to recoup its losses, the courts are unable to deal with it, without generating unacceptable risks of type one errors and thus chilling procompetitive behavior. Predatory Pricing in the European Union Akzo. The European Community first dealt with predatory pricing in 1985 in the Commission’s Akzo case.36 Akzo was a large chemical company that produced organic peroxides for the polymer industry as well as benzoyl peroxide compounds for use in the commercial baking of bread and other flour or milling additives. On a complaint by a rival firm, Engineering and Chemical Supplies (Epson and Gloucester) Ltd. (ECS), the European Commission instituted a proceeding against Akzo to determine whether Akzo had engaged in predatory pricing. The behavior in question involved Akzo’s substantial underpricing of ECS to dissuade the latter from entering the general European market for organic peroxides for the plastics industry. Before the Commission, Akzo defended its actions on a theory seemingly based on Areeda and Turner’s proposed predatory pricing analysis.37 Akzo contended that so long as its prices exceeded its average variable costs, that pricing should be treated as lawful. The Commission characterized Akzo’s position as depending “entirely on the mechanical application of a per se test based on marginal or variable cost”38 and listed three factors in addition to short-term efficiency that it said should be brought to bear in assessing Akzo’s pricing: First, the broad objectives of EEC competition rules set forth in Article 3(f) and thus the maintenance of “an effective structure of competition”; second, longer-term strategic considerations; and, third, the role of price discrimination in subsidizing a dominant producer’s low prices offered to a rival’s customers. Although the Commission determined that Akzo’s definition of variable costs was flawed and that Akzo in fact failed the very average-variable-cost test that it was promoting,39 the Commission based its decision on Akzo’s intent to coerce ECS from entering the plastics market. The Commission inferred Akzo’s anticompetitive intent from documentary evidence (including internal communications), from selective price cuts to ECS’s customers, from Akzo’s abandonment of its prior practice of fully recovering its costs in selling flour additives, and from its enabling of price cuts in flour additives with below-cost transfer prices from its plastics and elastomers division.40 In language that is remarkably similar to that used by the US Supreme

predatory pricing

93

Court’s description of monopolizing behavior in Brown Shoe, the Commission stated that “a small competitor” must “be protected against behavior by dominant undertakings designed to exclude it from the market not by virtue of greater efficiency or superior performance but by an abuse of market power.”41 The Commission concluded that even if short-term efficiency were the goal of the Treaty’s competition provisions, a dominant firm that prices above average variable cost but below average total cost will eventually eliminate “smaller but possibly more efficient firms.”42 On review, the Court of Justice affirmed the Commission.43 Like the Commission, the ECJ ruled that “costs” are not “the decisive criterion for determining whether price reductions by a dominant undertaking are abusive.” The ECJ supported this conclusion in the same way the Commission did: It referred to Article 3(f) and the need to prevent the impairment of an effective structure of competition; it referred to the strategic aspect of pricecutting; and it adopted the Commission’s conclusion that there “can be an anti-competitive object in price-cutting whether or not the aggressor sets its prices above or below its own costs, whatever the manner in which those costs are understood.”44 As did the Commission, the ECJ elaborated on the significance of the dominant firm’s price-cost relations: Prices below average variable costs . . . by means of which a dominant undertaking seeks to eliminate a competitor must be regarded as abusive. . . .45 Moreover, prices below average total costs, that is to say, fixed costs plus variable costs, but above average variable costs, must be regarded as abusive if they are determined as part of a plan for eliminating a competitor. Such price can drive from the market undertakings which are perhaps as efficient as the dominant undertaking but which, because of their smaller financial resources, are incapable of withstanding the competition waged against them.46

Tetra Pak. The Court of Justice later revisited predatory pricing in Tetra Pak,47 a case involving the production of milk cartons. In that case, the General Court had determined that Tetra Pak had sold nonaseptic cartons in Italy at prices that were below its average variable costs from 1976 to 1981. In 1982 Tetra Pak sold those cartons for prices above average variable cost but less than average total cost.48 In this case, the Court reaffirmed its earlier ruling that sales at prices below average variable cost are always deemed abusive and that sales at “prices below average total costs but above average variable costs are only to be considered abusive if an intention to eliminate [a competitor] can be shown.”49 In the proceeding of the General Court, it

chapter five

94

was determined that Tetra Pak intended to eliminate a competitor.50 In these circumstances, the Court of Justice ruled that it was unnecessary to establish that Tetra Pak had “a realistic chance” of recouping its losses.51 In denying the need to show a likelihood of recoupment, the Court of Justice followed the ruling in the General Court.52 This determination that a showing of a likelihood of recoupment is unnecessary to the establishment of a predatory pricing case, of course, differs from the approach taken in the United States, where the US Supreme Court has made that condition essential. The European Commission, the Discussion Paper, and the Guidance. The Eu-

ropean Commission issued a 2009 guidance dealing with an assortment of abuses of dominant position, including predatory pricing.53 Prior to issuing that guidance, however, it released a discussion paper in December 2005 on the application of Article 82 of the Treaty to exclusionary abuses.54 The discussion paper provides a more extensive treatment of the material that was later incorporated into its guidance. At the same time, the Office of the Chief Economist issued another discussion paper on predation.55 The discussion papers reveal the thinking of Commission staff and illuminate some of the positions taken in the guidance. The predation discussion paper is based heavily on American academic authorities. In particular the authors follow Bolton and colleagues in arguing for a view of predation that embraces not only the expensive classic predation in which the predator drives out rivals by selling below cost for a sustained period, but also describes often less expensive strategic forms of predation such as financial predation, reputation predation, price-signaling predation, and test-market predation.56 They follow Baumol’s recommendation for substituting average avoidable cost for average variable cost.57 They also propose that proof of likely recoupment should be part of the showing that enforcement authorities make in a predatory pricing case,58 thus aligning themselves with US decisions and rejecting the approach taken by the Court of Justice. Other positions taken by the predation discussion paper appear to be incompatible with current US law. The paper favors treating the “sacrifice” of profits made by a predator in changing the market structure in a less competitive direction as an investment whenever a course of action is chosen that is less profitable than other options open to it even if the enforcement authorities are unable to establish that the firm was selling below any measure of cost.59 Despite their embrace of a sacrifice test and their willingness to apply a profit-maximizing standard to determine sacrifice, the predation discus-

predatory pricing

95

sion paper also asserts that the most direct way to test for profit sacrifice is by comparing the incremental costs and revenues of an alleged predatory practice.60 Finally, the authors reject the “as efficient competitor” standard61 that is a fundamental component of the prevalent US analytical approach to predatory pricing as well as to a range of other antitrust issues. They see merit to protecting less efficient competitors on the ground that, even though less efficient, those firms may be exerting competitive constraints on the predator.62 The Article 82 (102) discussion paper adopts many of the positions of the predation discussion paper, including broadening the understanding of the scope of predatory pricing,63 the importance of recoupment,64 and the substitution of average avoidable cost for average variable cost.65 The Article 82 (102) discussion paper, however, adopts an “as efficient competitor” standard.66 In its 2009 guidance on enforcement priorities, the Commission embraces the “equally efficient” competitor 67 standard of the Article 82 (102) discussion paper and US antitrust law. From the “equally efficient competitor” test the Commission has derived tests of predation that turn on the relation of an alleged predator’s prices to its costs. As in Baumol, AAC is defined to include fixed costs incurred during the period under examination, and long-run average incremental cost (LRAIC) includes any product specific fixed costs ever incurred. We have previously commented on this concept whose abbreviation we shorten to LAIC.68 The guidance states that failure to cover AAC “indicates that the dominant undertaking is sacrificing profits in the short term and that an equally efficient competitor cannot serve the targeted customers without incurring a loss.”69 By contrast, “failure to cover LRAIC indicates that the dominant undertaking is not recovering all the (attributable) fixed costs of producing the good or service in question and that an equally efficient competitor could be foreclosed from the market.”70 Thus failure to cover AAC excludes equally efficient competitors, whereas failure to cover LAIC could exclude but does not necessarily do so. The guidance adopts a profit “sacrifice” approach to predatory pricing.71 Besides adopting a standard keyed to actual losses (such as selling below AAC), the sacrifice approach extends further. Thus the Commission says that it will investigate whether the alleged predatory conduct led in the short term to net revenues lower than could have been expected from reasonable alternative conduct.72 The guidance treads carefully on the issue of recoupment. It will be recalled that the likelihood of recoupment is an element of a predatory pricing

chapter five

96

case in the United States, but that the Court of Justice has taken the opposite position, ruling that a likelihood of recoupment need not be shown to establish a predatory pricing case. The predation discussion paper viewed a likelihood of recoupment as an element in an enforcement case, while the Article 82 discussion paper favors presuming the likelihood of recoupment from dominance. In the guidelines, the Commission comes close to adopting a recoupment requirement but avoids saying so directly. It says that it will presume consumer harm (and thus merit the Commission’s enforcement action73) if the dominant undertaking can reasonably expect its market power to increase after the predatory period.74 That, of course, would be a circumstance that would be essential to recoupment.75 So the guidance, contrary to the position of the Court of Justice, effectively requires a showing of a likelihood of recoupment in a predatory pricing case. The Lawfulness of Price “Squeezing” in the Two Jurisdictions Both jurisdictions have recently dealt with a price-squeeze theory of predatory pricing. Under this theory, an integrated firm that sells inputs to rivals with whom it competes in the output market can subject those rivals to a price-cost squeeze by narrowing the spread between the input prices that it charges its rivals and its own output prices at which the rivals must sell in order to be competitive with the integrated firm. One of the US government’s complaints against Alcoa in that well-known post– World War II monopolization case was that Alcoa sold aluminum ingot at high prices to its rivals in the aluminum fabrication market while it was selling fabricated aluminum products at low prices.76 Today a common complaint against local incumbent telephone companies is similar. The incumbent telephone company owns the telephone lines through which broadband services are delivered to subscribers. Rivals in the retail distribution of broadband service, therefore, must purchase access to those lines from the incumbent, yet the rivals must meet the retail broadband price of the incumbent. The rivals therefore are vulnerable to a price squeeze: if the price of access in the wholesale market is too high in relation to the retail price of service, the rivals’ profits will be reduced or eliminated. The US Supreme Court faced this issue in Pacific Bell Telephone Co. v. Linkline Communications, Inc. in 2008.77 In that case, the FCC had decided that because of competition in the broadband market from cable systems and satellite providers, it was no longer necessary to require incumbent telephone companies to provide access to their telephone lines to rivals in the broadband market.78 Because the incumbent company was under no duty to

predatory pricing

97

deal, the Court ruled, it had no duty to provide access at any particular price level. Moreover, there was no claim that the incumbent was selling broadband service below its costs. So the issue an allegation of a price squeeze raises appears to be governed by the general standard governing predatory prices: regardless of the level of its input prices, a supplier’s prices will be treated as potentially predatory only if its output prices fall below its incremental costs. The EU Court of Justice also has faced the price-squeeze issue in the distribution of broadband cable services. In Deutsche Telekom AG v. European Commission,79 the Court of Justice ruled in 2010 that price squeezing is an abuse of dominant position, a ruling that was the opposite of that of the US Supreme Court’s ruling in Linkline. Employing an analysis closely related to predatory pricing, the Commission had ruled that the incumbent telephone company’s pricing was abusive if the difference between the prices the incumbent charged customers in the retail market for broadband and the prices it charged competitors buying in the wholesale market was negative or insufficient to cover the product-specific costs to the dominant operator of providing its own retail services on the downstream market. This approach was upheld by the General Court 80 and approved by the Court of Justice. Although the Commission’s test related to predatory pricing standards, it differed significantly from them. The Commission was focused upon whether an equally efficient competitor would be able to compete with the integrated telephone company. This depended upon the difference between the upstream prices that the unintegrated rival paid for the input and the integrated company’s prices in the retail market that the rival must meet. If the integrated company’s supply prices are sufficiently high, the rival will be unable to compete. But high supply prices are not predatorily low retail prices. In a case where the supply prices are sufficiently high, the unintegrated rival will be unable to compete, but not because the integrated company is selling services below cost. The difference between the EU and the US approaches appears to be twofold. First, the US position assumes that there is adequate competition in final product market to render the participation of more purchaser-resellers of landline broadband services irrelevant. This may or may not be an accurate reading of the competitive situation in the United States. Moreover, the level of intramodal competition for broadband service in Europe may differ from the US situation. Second, regardless of the level of competition as the final European buyer experiences it, this appears to have been irrelevant to the decision of the EU Commission and courts. What mattered was apparently “fairness” to market participants other than the final buyer.

chapter five

98

Comparing the US and EU Law on Predatory Pricing Both US and EU institutions recognize that competition on price is the core of competitive behavior. EU institutions, however, assume that pricing behavior can generate anticompetitive results more often than US institutions do. Under US law, a case of predatory pricing requires, first, sales below an appropriate measure of incremental cost. The courts are gradually elaborating this requirement. The original Areeda-Turner proposal of using average variable cost as a marginal-cost surrogate is probably morphing into an average avoidable cost standard that will generally be higher than average variable cost. Second, a predatory pricing case requires a likelihood of recoupment. To show a likelihood of recoupment, it is generally necessary for the plaintiff to show a third factor: that entry barriers protect the affected market; otherwise, new entry would compete away the supracompetitive returns on which recoupment is based. EU law differs on each of these requirements. Under EU law the pricecost relation is important but apparently not necessarily decisive. Under the EU case law, recoupment is not critical, at least in some circumstances. Accordingly, the entry barrier issue— which plays a supporting role to recoupment— is necessarily less critical. The Commission seems to be moving in the direction of requiring a showing of the likelihood of recoupment in a predatory pricing case but is constrained by existing case law. US law is especially concerned to avoid chilling vigorous price competition by casting legal doubt on price undercutting. The need to show likelihood of recoupment in order to establish a predatory pricing case performs the role of raising the proof requirements to ensure that only genuine cases of predatory pricing reach the trial stage. Conversely, a sustained period of low prices benefits consumers, who will not be subjected to a subsequent period of supracompetitive prices if recoupment is unlikely. Thus pricing below cost without recouping the losses is deemed a positive benefit to consumers. Several opinions of the US Supreme Court have expressed a concern to avoid false positives. Accordingly, US law does not recognize a “sacrifice” theory of predatory pricing in which the alleged predator incurs an opportunity cost by using its assets in a market where its returns are lower than they would be in an alternative use. This “sacrifice” theory, in the apparent view of the Court, is too prone to error in application.81 The approach of US antitrust law to predatory pricing is based on several premises. First, it has been assumed that is rare, but this view may be changing. The second premise embraces aggressive price competition as a normative goal. The third premise is that it is easy for courts to mistake ag-

predatory pricing

99

gressive price competition for predatory pricing unless the court has clear lines drawn around predation. The fourth premise is that the rules should be drawn to sharply limit false positives. The fifth premise is that there are limits on the judicial abilities to identify predatory pricing without generating an undue number of false positives. And the sixth premise (common to the entire corpus of US antitrust law, with the exception of the RobinsonPatman Act) is that the law should aim at protecting competition and not competitors. Under the case law of the European Union, predatory pricing can be established by a showing that the alleged predator priced below average variable cost.82 Alternatively, predatory pricing can be established by showing that the predator priced above average variable cost but less than average total cost if the prices in question are part of a plan for eliminating a competitor. The Court of Justice has dispensed with a need for showing a likelihood of recoupment in at least some circumstances. Both the Court of Justice and the European Commission have indicated a willingness to recognize predatory pricing in circumstances when the predator is not selling below its costs, but these circumstances are unclear. The Commission has stated that it accepts a profit “sacrifice” theory under which a firm might be vulnerable to a charge of predatory pricing as a result of deploying its assets in ways that failed to maximize its immediate profits. The EU approach to predatory pricing appears to be focused on selling below some measure of cost or even at selling at prices that appear to deviate from immediate profit maximization without special concern for the market effects of that behavior. At least in circumstances when an alleged predator is selling below its average variable cost, the Court of Justice feels it is unnecessary to establish a likelihood of recoupment. That was the circumstance in Brooke Group where the US Supreme Court ruled in the opposite way. The Court of Justice also is willing to find predatory pricing when the alleged predator is selling above its average variable cost, so long as it is shown as a part of a plan to eliminate a competitor. This also conflicts with the most usual interpretation of US law under Brooke Group in which pricing above “incremental cost” is lawful and the appropriate measure of “incremental cost” is assumed to be average variable cost. Moreover, the elimination of a competitor would not, in itself, demonstrate a likelihood of recoupment. US courts would demand more impact be shown on the market, that is, the elimination of a sufficient number of competitors to significantly raise the likelihood of recoupment. Finally, neither the EU courts nor the Commission has shown a concern with false positives. To American observers, the EU rules on predatory pricing appear to ignore some of the basic concerns US courts articulate. By dispensing with a

chapter five

100

showing of a likelihood of recoupment, they are not using that issue to raise proof burdens and thus help to ensure that only legitimate cases of predatory pricing proceed to litigation— a concern that the role of private litigation always pushes to the fore in the United States. More significantly, indeterminate predatory pricing standards may chill aggressive but competitive price-cutting. The EU rules, however, become quite understandable if a major intended function is to protect rivals from subjection to damaging price competition. In that case, there is no need for a recoupment requirement; the elimination of a single competitor would suffice and there would be no need to demonstrate any further market effect. This standard— protecting rivals from damaging price competition— is basically a standard of “fair” competition. It is a standard that focuses on the welfare of producers and from which consumers may benefit derivatively. Conclusion There is clearly room for some reconciliation between the US and EU positions. Brooke Group appears to block any finding of predation at prices above average total cost, and it requires that a likelihood of recoupment be shown. The need for recoupment clearly enjoys some support among Commission economists. The Commission has employed AAC as a substitute for average variable cost, and the withdrawal of a DOJ document that seems to favor a safe harbor for prices above AAC could presage a more careful consideration of the possibility of predatory prices between AAC and ATC (LAIC) if market conditions were favorable to successful recoupment through the exclusion or disciplining of rivals. In both jurisdictions, “profit sacrifice” could be treated as one relevant consideration in the determination of a dominant firm’s possibly predatory strategy.

6

Exclusive-Supply Contracts Introduction The United States and the European Union currently take quite different antitrust approaches to the evaluation of a series of vertical practices: exclusive-supply contracts, single-product rebates, and bundled rebates. This chapter and the two following will explore these practices in turn. All three practices share the common feature that something constrains the immediate access of all sellers to all buyers at the next level of commerce.1 In the European Union little distinction is made between exclusive-supply and tailored discounts. Exclusive-supply contracts and rebating schemes can involve a single product or a group of products. Exclusivesupply arrangements involve a buyer’s contractual obligation to purchase the product or products in question exclusively from a single supplier. The practice under which distributors accept a contractual obligation to handle their supplier’s full line of products becomes a form of exclusive dealing when the distributors are obliged to handle only their supplier’s products. Despite the connections among the products, full line forcing is generally evaluated in the US under a version of the Rule of Reason rather than the more stringent standards applicable to tying arrangements. When a supplier distributes a line of products through exclusive-supply contracts, its

chapter six

102

bundle may consist of goods with independent demands (farm tractors and lawn tractors), they may be complements (tractors and plows), or they may be substitutes (farm tractors of different sizes). The possible threat to competition from certain bundling pricing practices discussed in chapter 8 turns partly on the cross elasticity of demand among goods within the bundle as well as the degree of monopoly power for one or more of the included goods. Time is an essential dimension in exclusive dealing because the remaining length of the contract meters the penalty for its breach and measures the opportunities (which come when the contract is up for renewal) to switch suppliers without penalty. In a somewhat parallel fashion, when a buyer faces multiple competing sellers of essentially the same product or products, the period over which a volume rebate scheme is calculated (for a buyer of given size) determines both the penalty-free opportunity to change suppliers and the temporal pattern of changing disadvantage that competing sellers face.2 A final analytic point lies in distinguishing between direct foreclosure of rivals from the ultimate market and foreclosure of their access to a market impeding their abilities to distribute their goods. If behavior forecloses certain intermediate distribution channels or increases the cost of employing them, a full competitive analysis may require that attention be directed to whether the cost and effectiveness of other channels are adequate to maintain competitive constraints on the principal players in the market in question.3 This issue has come up in the US cases and is discussed below. Many writers divide possible harm to competition from vertical practices into two major categories: predation and exclusion. These terms are used variously in the literature; predation is usually linked to practices aimed at specific rivals, while exclusion is often used to refer to action denying all rivals access to a part of the distribution system. Both practices can involve assorted behaviors, and we think the most useful focus lies in the impact of such practices on either the price or costs of the affected rivals. We identify anticompetitive practices that lower the price the rival must meet to enter or remain in a market with predation and those that raise rivals’ costs as exclusion. Some practices may do both. Indeed, an antitrust issue of growing importance is the extent to which practices deemed to constitute exclusion under traditional approaches (such as the “substantial share” test applicable to exclusive-supply contracts discussed in this chapter) can be converted to issues of prices and related costs and then evaluated under the decisional criteria applicable to predation. Thus, for example, in the Nicsand case discussed below the initial Sixth Circuit panel used traditional exclusion standards, but the en banc court that analyzed the case under predation standards reversed that decision. In the en banc court’s view, the

exclusive-supply contracts

103

defendant 3M Corp. attracted all the major buyers because it offered them better prices. Nicsand, the excluded supplier, could have won the day by offering more attractive prices than 3M did. Nevertheless, many practices that appear as predation or exclusion are in fact largely or entirely something else: they are efficiency enhancing or they are a means of price discrimination or both. Efficiency may be enhanced inter alia by lowering selling expenses for the seller and search costs for the dealer, providing incentives for suppliers to train dealers who will devote special attention to the seller’s product line, and encouraging more advertising by sellers who need not fear that dealers will dilute their efforts by pushing other brands when customers arrive.4 Sorting out the effects of vertical practices is one of the major challenges in contemporary competition policy, and a large part of current transatlantic differences stem from the wide gap in thinking about price discrimination discussed in chapter 4. Exclusive-Supply Arrangements: The United States The anticompetitive potential of exclusive-supply contracts a supplier imposes lies in its ability, in some circumstances, to impede rivals’ access to distribution channels as effective as those of an incumbent firm. This could be the case where all channels, or channels of sufficient quality, are limited and not easily expandable. Exclusive-supply contracts are governed by Section 1 of the Sherman Act and Section 3 of the Clayton Act.5 Where their foreclosure potential creates a “dangerous probability” that the supplier will acquire a monopoly, they can violate the attempt clause of Section 2 of the Sherman Act. Finally, when they enable a firm to acquire or maintain a monopoly, they can violate the monopolization clause of Section 2 of the Sherman Act as well. Exclusive-supply contracts also carry the potential for generating a variety of efficiencies.6 Generally, these contracts are lawful, unless they foreclose a “substantial share” of the market to competition by rivals. The US case law governing exclusive contracts uses an approach initially employed in a 1949 decision of the Supreme Court, Standard Oil Co. v. United States (Standard Stations).7 Two years earlier, in International Salt Co. v. United States,8 the Court had condemned a tying arrangement because the volume of business affected was not “insignificant or insubstantial,”9 as measured by dollar volume.10 In Standard Stations, the question before the Court was whether to employ the same test for exclusive dealing. The argument for doing so was powerful because the identical statutory language of Section 3 of the Clayton Act governed both tying and exclusivesupply arrangements.

chapter six

104

The Court first reviewed the beneficial effects of these contracts (risk reduction and the consequent facilitation of long-term planning).11 It then chose not to evaluate them under the test that it had employed in International Salt to evaluate tying contracts. Instead, the Court ruled that exclusivesupply contracts were governed by a “substantial share” test. Requirements contracts would be upheld unless they foreclosed a substantial share of the market to rivals.12 Thus for exclusive-supply contracts, the Court chose to look to a proportionate impact rather than, as in the case of tying contracts, ruling them illegal whenever they involved a substantial dollar amount of commerce.13 The Court later reaffirmed the application of the “substantial share” test to requirements contracts in Tampa Elec. Co. v. Nashville Coal Co.14 In Tampa of 1961, the Court impliedly recognized that the length of an exclusive contract was relevant in assessing its exclusionary potential but nonetheless approved a twenty-year contract where it involved only a small share of the market. Tampa can be read as the beginning of a judicial focus upon the market position of the supplier. It distinguished earlier rulings on the ground that the instant seller lacked market power and lacked “the myriad outlets with substantial sales volume, coupled with an industry-wide practice of relying upon exclusive contracts” as was the case in Standard Stations.15 In Standard Stations and Tampa, the arrangements were challenged under Section 3 of the Clayton Act. An exclusive-supply contract whose foreclosure effects do not run afoul of Section 3 will not violate Section 1 of the Sherman Act.16 In Standard Stations the Court ruled that the defendant’s distribution contracts failed the substantial-share standard even though only 6.7 percent of the relevant product (gasoline) was governed by those distribution contracts. The facts that the defendant’s rivals also distributed their gasoline through exclusive-supply contracts and that 55 percent of the total gasoline sold in the relevant market was distributed through exclusive contracts appear to have influenced the Court’s decision. In a footnote, the Court adverted to the limited number of strategically located outlets available to rival suppliers, thus emphasizing the foreclosure aspects of these contracts.17 In Tampa, the defendant’s contract as the exclusive supplier of coal to a Florida utility involved less than 1 percent of the coal in the relevant market, a share the Court ruled to be insubstantial.18 Jefferson Parish Hosp. Dist. No. 2 v. Hyde,19 a 1984 case dealing with tying arrangements, added an important refinement to these earlier decisions. In this case, the Court made clear that unless a defendant possessed a market share of over 30 percent, tying arrangements would be evaluated under the Rule of Reason.20 Historically, the courts have treated tying arrangements as more suspect than exclusive-supply contracts because the

exclusive-supply contracts

105

courts recognized the potential efficiencies in requirements contracts at least from the Standard Stations decision of 1949. But for decades thereafter, they viewed tying arrangements as serving “hardly any purpose beyond the suppression of competition.”21 The Court’s decisions in International Salt and Standard Stations illustrate its different approaches to tying and exclusive-supply arrangements. Today the Court’s strong hostility to tying arrangements has diminished,22 and they are now only a per se violation when the seller possesses power in the market for the tying product, a substantial amount of sales in the tied product market is affected, and the seller is unable to establish the efficiencies of the arrangement. Therefore, when the Court insisted on Rule- ofReason evaluation of tying arrangements in Jefferson Parish, its decision implied that exclusive-supply contracts involving less than 30 percent of the market would also be evaluated by the Rule of Reason where the combination of the insignificant market power and the efficiencies generated by those contracts would generally ensure their lawfulness. Thus there is an implicit requirement that the supplying firm possess some degree of market power in order to be capable of violating either Section 3 of the Clayton Act or Section 1 of the Sherman Act. Thus in her concurring opinion in that case, Justice Sandra Day O’Connor addressed exclusive dealing directly, concluding that “Exclusive dealing is an unreasonable restraint on trade only when a significant fraction of buyers or sellers are frozen out of the market by the exclusive deal.”23 The recent exclusive-supply cases strengthen this conclusion. In Stop & Shop Supermarket Co. v. Blue Cross & Blue Shield of Rhode Island,24 a 2004 decision by the First Circuit, that court indicated that “for exclusive dealing, foreclosure levels are unlikely to be of concern where they are less than 30 or 40 percent.”25 In the Microsoft antitrust case, the district court (which ruled for the government on almost all counts in the complaint) declined to find unlawful Microsoft’s exclusive agreements with Compaq, AOL, Internet content providers, internet software vendors, and Apple because the foreclosure confronting Netscape, Microsoft’s rival, was not shown to have amounted to 40 percent of the browser market.26 Even though Microsoft was able to control the most efficient distribution channels, Netscape was nonetheless able to reach customers through other, albeit less efficient, distribution channels (such as downloading from the internet and through retail stores).27 On appeal, the DC Circuit referred in dicta to “the roughly 40% or 50% share usually required to establish a § 1 violation.”28 A number of cases have further elaborated the substantial share analysis by examining the distribution alternatives available to competing suppliers to reach the ultimate consumer market. As noted above, this issue

chapter six

106

was touched on in Standard Stations itself, where the Court adverted to the limited number of strategically placed outlets available to Standard’s rivals. That issue was the focus of the Ninth Circuit’s attention in Omega Environmental, Inc. v. Gilbarco, Inc.29 of 1997. There the court identified the foreclosed market (in petroleum dispensing equipment) as the percentage of Gilbarco’s market share sold through its authorized distributors with whom it had exclusive-supply contracts. Since 70 percent of Gilbarco’s sales were through distributors and it captured 55 percent of the total market in a particular year, the court concluded that its exclusive distribution contracts foreclosed about 38 percent of the total market. Further, the court ruled that because the foreclosure was at the distributor level, a higher standard of proof of foreclosure is required because there were alternative ways of reaching the final consumers, such as direct sales or sales through servicing contractors. And the alternatives need not be as efficient as the defendant’s set of distributors. In this case, the court concluded that the alternatives available to Gilbarco’s rivals effectively eliminated any foreclosure effect that its exclusive distributorship contracts might otherwise have had. This conclusion was reinforced by the short duration and easy terminability of the contracts. Other cases have reached similar conclusions, although the DC Circuit’s handling of Microsoft’s exclusive arrangements, discussed below, is in some tension with these rulings. The extent of possible foreclosure diminishes over the life of an exclusivesupply contract. A buyer who breaks an exclusive-supply contract is liable for the seller’s lost profits. Thus a breach at the beginning of the contract will make the buyer liable for all of the profits the seller could have earned throughout the remainder of the contract. But the closer the breach is to the end of the period covered by the contract the less is the buyer’s potential liability. A rival seller must not only offer the buyer a better price but also a price that additionally compensates the buyer for the damages the buyer will be assessed. Thus once the buyer has entered into a requirements contract and during most of that contract’s term, a rival seller must offer the buyer a price below the principal supplier’s cost, since the damages for breach are measured by the supplier’s profit (which is the difference between the contract price and the supplier’s cost), and the price must be sufficiently low to compensate the buyer for these damages. Yet as the end of the contract period approaches, this condition weakens, since the rival may prevail if it offers a new requirements contract at a price that is not necessarily below the principal supplier’s cost but that is below the principal supplier’s price. At the end of the contract period, the rival merely has to undercut the price of the incumbent, if the product is homogeneous. A corollary to

exclusive-supply contracts

107

this analysis is that exclusive-supply agreements are less suspect as their length diminishes. Special Rules for Monopolization or Attempted Monopolization under Section 2 There are special rules for evaluating exclusive-supply contracts under Section 2 of the Sherman Act. Most antitrust issues involving exclusive-supply contracts will involve Section 3 of the Clayton Act or Section 1 of the Sherman Act, where the market power of the supplier will be in issue as well as whether the exclusive contracts constitute a “substantial share” of the relevant market. When these contracts are challenged under Section 2, additional issues become relevant: whether the contracts create a “dangerous probability” that the supplier will achieve a monopoly (for the attempt clause), and whether the contracts enable the supplier to acquire or maintain a monopoly (for the monopolization clause). Section 2 has enabled the government to prevail where it has been unable to prove the existence of agreements between the supplier and its distributors (as in Dentsply, discussed below) and when the government failed to convince the trial court that the supplier’s contracts constituted a substantial share of distribution (as in Microsoft, also discussed below). The Third Circuit has been a major locus of new antitrust developments on exclusive arrangements under Section 2. In Dentsply of 2005,30 the Department of Justice challenged the lawfulness of the distribution arrangements between Dentsply, a fabricator of artificial teeth, and its dealers. Dentsply possessed a market share of 75 to 80 percent (measured on a revenue basis) in the artificial teeth market but only a 67 percent share on a unit basis. These shares were sufficient for the court to infer that Dentsply held an effective monopoly in this market.31 The conduct at issue was Dentsply’s refusal to sell to dealers that added lines of artificial teeth from rival manufacturers. Because distribution through dealers was more efficient than direct sales to dental laboratories, and because the lower court had rejected Dentsply’s justification as pretextual, the court held that Dentsply’s exclusion of rival suppliers from its own dealer network violated Section 2 as monopoly maintenance. In the court’s view Dentsply’s exclusive dealer network was a means through which Dentsply was unlawfully maintaining its effective monopoly. The Third Circuit’s Dentsply decision bears a striking resemblance to the DC Circuit’s 2001 ruling in Microsoft on monopoly maintenance. In Microsoft that company prevailed on a Section 1 challenge to its exclusive arrangements with internet service providers, just as Dentsply prevailed in

chapter six

108

the government’s Section 1 attack on its distribution arrangements. In both cases, the government chose not to appeal the Section 1 ruling and prevailed on a Section 2 assault against the exclusive arrangements. Indeed, the Third Circuit justified its decision by citing and explaining the Microsoft precedent.32 The Third Circuit also relied on its then recent monopolization decision in LePage’s as authority for its Dentsply ruling.33 Dentsply as well as the LePage’s decision on which it partially relies, however, are in some tension with the Sixth Circuit’s en banc ruling in Nicsand, Inc. v. 3M Co., to be discussed later in this chapter, where 3M’s exclusive arrangements with retailer customers, coinciding with its acquisition of a monopoly in the automotive do-it-yourself sandpaper market, were treated as the result of aggressive competition and the excluded seller’s foreclosure as due to its failure to offer sufficiently attractive prices to those customers. In the DC Circuit’s Microsoft decision, that court ruled that Microsoft’s exclusive-supply arrangements with Compaq, AOL, Internet content providers, internet software vendors, and Apple violated Section 2, even though, as previously observed, the district court in that case had found those contracts lawful under Section 1 and the government had not appealed that ruling. The court of appeals, however, as part of a monopolization analysis, focused on their impact in denying Netscape, Microsoft’s browser rival, a sufficient installed base to develop into a platform in its own right, and thus eroding Windows platform monopoly. Under the court’s approach, Microsoft bore the burden to prove that these exclusive-supply arrangements were justified by their efficiencies, a burden that Microsoft failed to carry. The court took the view, contrary to the district court, that because these distribution channels were the most cost-efficient ones (apart from installation by computer manufacturers), Microsoft acted unlawfully in denying these channels to Netscape.34 Although the court recognized the 40 to 50 percent share usually governing the application of Section 1, it thought that in some circumstances a monopolist could violate Section 2 with a smaller share. Because the issue involved Microsoft’s action designed to impede Netscape developing into an alternative to Microsoft’s platform monopoly, this was such a circumstance. In 2012 the Third Circuit revisited exclusive-supply in Meritor, LLC v. Eaton Corp.,35 where it reaffirmed and elaborated its earlier rulings. This case involved heavy-duty truck transmissions produced by Eaton Corp., for many years the only North American producer of these transmissions. In 1989 Meritor entered the market and began offering manual transmissions, and in 1999 Meritor formed a joint venture with ZF Friedrichshafen, a leading supplier of heavy-duty transmissions in Europe, to compete with Eaton. Eaton’s response was to enter into long-term agreements with the four di-

exclusive-supply contracts

109

rect purchasers, domestic truck manufacturers. Although truck buyers still could select Meritor transmissions, Meritor’s market share fell to 8 percent, a share too low for the success of the joint venture, which was dissolved in 2003. Meritor’s share subsequently fell further, and it left the market in 2006. The court rejected Eaton’s defense that the exclusive agreements were governed by predatory pricing standards. The mere fact that Eaton was selling its transmissions at above-cost levels did not immunize it to an antitrust evaluation under the “substantial share” analysis traditionally applied to exclusive-supply contracts. According to the court, the exclusive contracts required purchasers to take between 80 and 97.5 percent of their requirements from Eaton. This was sufficient to reduce Meritor’s share to 4 percent from 8– 14 percent a few years earlier. The result was foreclosure that exceeded the standards applicable to all of the applicable antitrust provisions: Section 3 of the Clayton Act and Sections 1 and 2 of the Sherman Act. Indeed, when Meritor left the market, the foreclosure was complete. Both Dentsply and Microsoft took the position that a monopolist might be bound by stricter standards than would apply under Section 1, although that language in Dentsply was probably dicta because Dentsply’s share substantially exceeded the 40 to 50 percent share used in evaluating exclusivesupply agreements under Section 1. The Meritor decision follows a fortiori from Dentsply and Microsoft because the contracts in Meritor were with the final purchasers thus leaving no room for a contention (as in Gilbarco) that there were alternative ways for rivals to reach the downstream market. And the share foreclosed in Meritor exceeded even the very large share involved in Dentsply. Exclusive-Supply Arrangements: The European Union EU competition law in Article 101(1) broadly prohibits agreements and concerted practices that adversely affect competition, but Article 101(3) provides grounds for exempting agreements and practices from the prohibition. Article 101(3) essentially converts Article 101 into a Rule-of-Reason provision by legitimating agreements that enhance efficiency and allow consumers a “fair share” of the resulting benefit and do not eliminate competition in a substantial part of the product in question. Under the 1999 block exemption regulation, a firm would be allowed exclusive dealing if its share of the total market were 30 percent or less and there were no other questionable elements to the agreement.36 Article 102 forbids “dominant” firms from abusing their dominant position. Besides its prohibition of abuse in general, Article 102 contains four specific clauses directed at particular forms of abuse. Clause (c) forbids “ap-

chapter six

110

plying dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage.” The most obvious application of this language would prohibit price discrimination by dominant firms that would impose higher prices on some customers that are competing for resale with rivals who are charged lower prices. This clause necessarily has application to exclusive-supply contracts when the supplier offers lower prices to buyers who enter into exclusive arrangements. In most of the European cases that have condemned exclusive-supply contracts, the seller has offered lower prices to customers that have accepted exclusivity, so clause (c) has been applicable. But while the focus of clause (c) is on protecting the seller’s customers from discriminatory supply prices, the general clause of Article 102 has been construed to treat exclusive-supply contracts as abuses of dominant position because they foreclose rivals from access to the seller’s customers. In Hoffmann-La Roche, perhaps the leading case on exclusive-supply arrangements, the Court of Justice stated that “obligations . . . to obtain supplies exclusively from a particular undertaking  . . . are incompatible with the objective of undistorted competition within the Common Market, because . . . they are not based on an economic transaction which justifies this burden or benefit but are designed to deprive the purchaser of or restrict his possible choices of supply and to deny other producers access to the market.” In the language of EU competition law, exclusive-supply arrangements are referred to as “single branding.” As the above quote from Hoffmann-La Roche indicates, single branding by a dominant firm is presumed abusive because of its foreclosure effects. Many of the single-branding or exclusivesupply cases that have come before the courts have involved a rebate system the court has condemned because of its capacity to induce customers to remain loyal to their original supplier, denying rivals access to their patronage. In Hoffmann, the ECJ observed that “exceptional circumstances” might make an exclusive-supply arrangement permissible. Such exceptional circumstances, the ECJ suggested, could exist when an agreement met the conditions set forth in Article 101, and in particular paragraph (3) dealing with the generation of efficiencies.37 In later cases the courts have indicated that discounts for quantity keyed exclusively to the volume of purchases would be deemed presumptively lawful and not to produce a foreclosure effect. Thus price reductions available to all buyers pursuant to a uniform schedulereducing price as the quantity purchased increases are treated as prima facie lawful. This treatment is based on presumed scale economies. If increasing the quantity supplied results in lower costs for the supplier, the General Court said in Michelin II, “the latter is entitled to pass on that reduction to the customer in the form of a more favourable tariff.”38

exclusive-supply contracts

111

Hoffmann connected the efficiency issue under Article 102 to the provisions of Article 101(3). The Commission’s discussion paper follows this approach to Article 102 by limiting efficiency justifications to those that meet the conditions of Article 101(3). Adopting what is ostensibly a consumerwelfare criterion, the paper takes the position that no claimed efficiency justification is acceptable for behavior that makes consumers worse off. The Guidance on Article 82 and Effects-Based Analyses The Commission’s guidance on Article 82 is an example of the Commission’s use of an “effects” approach as opposed to a “form” analysis that is still being followed by the courts. The difference, as one commentator has explained, is that under a form-based approach a practice’s anticompetitive effects could be “inferred from its intrinsic features,” but an effects-based approach requires the “verification of competitive harm, through a detailed assessment.”39 This can be restated in terms more familiar to American readers: a form-based approach is a per se approach under which certain categories of behavior are deemed unlawful. An effects-based approach is analogous to the use of the Rule of Reason: the lawfulness of the behavior under review depends upon an assessment of its economic effects. Under a form-based approach, the use of exclusive-supply contracts by a dominant firm is considered inherently anticompetitive. The Commission’s guidance, however, broadens the focus to include an assessment of market conditions. Thus in a passage reminiscent of the “substantial share” analysis US courts use, the Commission’s guidance identifies the extent of the allegedly abusive conduct as important in the determination of the foreclosure effect: In general, the higher the percentage of total sales in the relevant market affected by the conduct, the longer its duration, and the more regularly it has been applied, the greater is the likely foreclosure effect.40

As observed, these factors were used in the US Supreme Court’s Tampa decision to assess the competitive consequences of exclusive- supply contracts and have since been regularly employed by the US lower courts for this purpose. An approach to competition law based on effects would move EU law closer to the Rule-of-Reason evaluation US courts and enforcement authorities follow. Commentators have given the Commission high marks for introducing new flexibility into EU competition law by focusing on the actual effects of

chapter six

112

behavior. Yet the move raises a number of questions. How will the pursuit of an effects-based approach by the Commission affect the relations between the Commission and the courts, which still pursue a form-based approach? If the courts continue to apply a form-based approach, then will judicial review of the Commission’s decisions ignore the Commission’s effects analyses? That is, will the courts say that the Commission’s effects analyses were unnecessary because the behavior in question falls into a category the courts treat as inherently anticompetitive? Beyond these questions, however, is the extent to which the Commission’s guidance is in fact freed from a form-based approach. Even an approach that focuses on economic effects will generally employ presumptions to allocate burdens of proof. That is the role “substantial share” in US antitrust law plays. It allocates to the defendant the burden of justifying its exclusive-supply contracts on efficiency grounds. Thus it would not be surprising if the guidance, although predominately an effects-based document, incorporated form-based analyses at the margins. The commentators have begun the task of searching the guidance for indications of where the Commission frees itself from showing effects and retains the right to rely on a truncated or form-based analysis.41 Some Important Theoretical Issues Connected with Exclusive Supply The original Chicago position on exclusive-supply contracts held that they must be efficiency increasing or else the parties would not have entered into them. A monopolist or near monopolist supplier could not employ exclusive-supply contracts to deter the entry of a potential rival (by denying the potential rival a market for its goods) because its customers would not willingly discourage the entry of a competing supplier whose presence would erode the power of their current supplier to impose supracompetitive prices on them.42 Moreover, the monopolist supplier could not pay his customers (say, in reduced supply prices) to enter into exclusive-supply contracts either when their foreseeable effects would be to deter an otherwise likely entry. In the circumstances supposed, the customers would demand compensation for their loss of consumer surplus, an amount that would exceed the monopolist’s profits. Since under this analysis, exclusive-supply contracts cannot be misused to strengthen a monopoly; their only use lies in the generation of efficiencies. A number of “post-Chicago” theorists have identified flaws in the “Chicago” approach. One possibility is a collective-action problem that may impede customers from exercising their aggregate bargaining power in negotiations with a monopoly supplier who plays purchasers off against each

exclusive-supply contracts

113

other.43 In this scenario, the incumbent seeks to maintain exclusive-supply relations with enough distributors to block an entrant from reaching minimum efficient scale. Although entry would be in the purchasers’ interest, the incumbent prevails because the distributors are unable to coordinate their response. The purchasers may be either dealers or final purchasers. Second, the Chicago analysis does not consider the possibility that agents outside of the exclusive dealing contract may have rent exacted from them.44 Thus, for example, the existing incumbent supplier and its distributors could, in theory, write contracts that allow a more efficient firm to enter and to provide a cheaper source of supply for the distributors at cost. For example, the distribution contracts could permit the distributors to purchase from an alternative source if they (or their alternative source) paid an entrance fee to the incumbent. So long as the incumbent can capture the benefit of the entrant’s efficiencies, both the incumbent and the distributors can be better off. This model fits the situation in which a dominant supplier enforces its right to damages under exclusive-supply contracts in the amount of its lost profits against dealers that breach those contracts in order to buy their supplies from a more efficient alternative supplier. But this strategy can also produce inefficiency when such contracts do in fact deter the potential entrant and the higher cost source of supply persists. Many of these post-Chicago models stress the extent to which exclusive distribution arrangements can exclude entrants where there are economies of scale. The exclusion works because the exclusivity is granted selectively and creates a coalition between the incumbent supplier and favored distributors against excluded distributors.45 The incumbent and the favored distributors then share the arrangement’s monopoly profits. The potential entrant is excluded because the excluded distributors do not purchase in the aggregate sufficient product for the entrant to reach minimum efficient scale. Many models fail to distinguish between demand for resale and final demand. This simplification may work where exclusivity applies to geographically isolated markets, but this is not appropriate where formal or de facto exclusivity is practiced in sales to buyers who compete with one another. Some models therefore stress the role of downstream competition, especially in homogeneous products. In such markets, even a single uncommitted distributor carries the potential for shortcutting the incumbent’s attempt to deny distribution outlets to a potential entrant. If the potential demand in the downstream market is sufficiently great, the purchases (for resale) of the uncommitted distributor can provide the entrant with the sales it needs to attain scale economies. In this scenario, the Chicago analysis accurately describes the predicament of the incumbent supplier.46 Conversely, as product differentiation rises (that is, cross-elasticity of demand among competing

chapter six

114

distributors falls), that analysis becomes less applicable. The latter situation may prevail in many circumstances where the good in question is incorporated into another differentiated product. The various economic models produce varying results depending, inter alia, on the competitive means and opportunities of downstream firms, the sequence of actor moves, the exact terms of the exclusive contract, and the size of entrant’s fixed costs relative to the size of the entire market.47 Many models that consider markets in which exclusive arrangements account for a substantial share of purchases share common features. First, it is typically assumed that the number of buyers is limited. As pointed out in the preceding paragraph, this makes sense for final demand, but it may often be an unwarranted assumption for resellers. Second, there is typically one incumbent firm and thus no competition within the provider market. Finally, the barrier to entry rests solely on economies of scale, which is a convenient but unrealistic analytical assumption. In fact, nearly all real entry takes place in markets with substantial established buyer preferences that only some combination of time, expenditure, and attractive offers can overcome. Uncertainly thus permeates entry conditions and must therefore play a large role in each actor’s strategy. Despite these limitations, however, these models demonstrate how exclusive contracts can convert potential efficiency gains into rents absorbed by parties other than final purchasers. In common antitrust parlance, contractual arrangements can raise rivals costs.48 Raising-rivals’-costs models invariably involve the use of monopoly power to diminish competition. Although exclusive-supply contracts and their close cousins, fidelity- and bundled-rebate arrangements, are vertical restraints, their welfare significance lies in their impact on the horizontal relationship between a dominant supplier employing these devices and its rivals. Properly employed, the raising-rivals’-costs line of analysis can help to identify the market in which the restraint is being imposed.49 These models thus alert the analyst to possibilities, the plausibility of which depends on an evaluation of just how much competitive restriction is actually in play. Litigation in the Sixth Circuit in 2007 involved an antitrust claim involving several of the factors identified in these models.50 Nicsand, a supplier of do-it-yourself automotive sandpaper, originally held a 67 percent market share until it began losing share to 3M, the only other supplier of such sandpaper. Both Nicsand and 3M sold to a highly concentrated retailer market, where six large retailers made 80 percent of the purchases. Most of these retailer customers wanted to handle only one brand of the product. 3M began eroding Nicsand’s share when it offered large up-front payments to customers for switching to 3M under multiyear exclusive-supply arrange-

exclusive-supply contracts

115

ments. As Nicsand lost market share to 3M, its costs rose and eventually it withdrew from the market and filed for bankruptcy. In an antitrust suit challenging the lawfulness of 3M’s contracts under Section 2 of the Sherman Act, the initial Sixth Circuit panel discussed some of the theoretical points identified above. The panel thought a collectiveaction problem among the distributors enabled 3M to sign up most of the distributors to exclusive-supply arrangements and that 3M reinforced the distributors’ disarray with large up-front payments, raising the costs of its rival.51 On these possibilities, the court ordered the case to proceed to trial. But when the en banc court reconsidered the matter, it reached the opposite result. That court ruled that 3M’s payments to the distributors should be seen as price reductions, and that because 3M’s prices (adjusted to include the up-front payments) exceeded its costs, they were not predatory and therefore lawful.52 On the exclusive agreements themselves, the en banc court took the view that Nicsand could have secured some or all of those contracts for itself, if it had underbid 3M.53 In the en banc court’s view, the events underlying the litigation constituted vigorous price competition. Vertical arrangements generally benefit the cooperating parties, so the issue is whether any of these vertical arrangements impose illegitimate costs on nonparties, such as the supplier’s rivals, and, if so, what is the impact on welfare? More precisely, is the supplier using exclusive-supply contracts or related vertical arrangements to monopolize a market to which the supplier’s rivals need access in order to compete? Thus, as Timothy Brennan suggests, the focus should be whether a complement market to the supplier’s product— such as the market in distribution facilities— has been monopolized.54 This viewpoint makes it easier to direct analysis to the critical restraint and to examine its operation. Thus the problem exclusive-supply contracts raise is whether rival suppliers (or new entrants) have access to dealer and/or distributor services necessary to compete, or whether the current supplier has been able to deny such access. This parallels the traditional approach, exemplified by Standard Stations, which made the lawfulness of exclusive-supply contracts turn on whether they covered a “substantial share” of the relevant market. Brennan takes his analysis somewhat further. On the premise that exclusive-supply contracts between the supplier and the distributors adhering to them are significant because of their horizontal effects, Brennan would treat the set of a supplier’s several exclusive-supply contracts as analogous to a horizontal merger among distributors. In this vein, he would use the merger guidelines to assess the lawfulness of these contracts. Thus under his approach the merger guidelines effectively determine what is a “substantial share” for purposes of foreclosure analysis. The guidelines also add

chapter six

116

additional focus to the question of whether or not the complement market (distribution) is expandable and thus not foreclosable. Using an approach modeled on the merger guidelines to evaluate exclusive-supply contracts, as suggested by Brennan, further suggests that whenever concentration is sufficiently high in the complement market to raise a prima facie bar under the guidelines, additional exclusive-supply contracts would be permitted only if the efficiencies those contracts generated were large enough to offset the negative effects of that increased concentration. Those negative effects would be higher distribution costs incurred by actual and potential rivals from the supplier’s exclusivity practices. The Brennan perspective is useful, but it can also confuse. The relation between concentration and possibly anticompetitive outcomes in contemporary merger analysis rests on two kinds of coordinated action: either increased recognition of mutual dependence among horizontal competitors or the unilateral effects of bringing substitutes under common control. Brennan does not develop analogous considerations in his complement monopolization argument. Instead, he stresses that the use of a merger-like focus on concentration could provide markers for possible exclusion. But exclusion appears analogous to barriers to entry, a critical factor in merger analysis and distinct from concentration levels, so the problem of evaluating the difficulty faced by a disadvantaged upstream seller seems repackaged more than resolved. Conclusion Overall, the antitrust assessment of exclusive-supply contracts in Europe is now similar to that of the United States. As the prior discussion explains, the lawfulness of exclusive-supply contracts turns on whether more than a substantial share of the market has been foreclosed in both the United States and the European Union. This result is reached directly in the United States and indirectly in the European Union. In the European Union this result is reached through an initial determination of whether the supplier holds a dominant position in the market, whereas in the United States concern focuses on the extent of the total market involved in exclusive contracts. By contrast, and as more fully developed in the following two chapters, both jurisdictions apply a different analysis to single-product and bundled discounts, an analysis relating price to cost and derived from their approaches to predatory pricing. In both situations, the antitrust concerns are the same: whether a supplier’s rivals are being denied access to critical distribution facilities or input purchasers.

7

Single-Product Loyalty Rebates: Is a Large Gap Narrowing? Discounting from list price is a familiar and widely used technique to increase sales. One of the most familiar is the quantity-discount schedule under which a seller attaches successively lower prices to its goods as the quantity purchased increases.1 Quantity discounts can also be tailored to the circumstances of individual buyers. These entitle a buyer to a discount on all of its purchases over a certain period, so long as the buyer’s purchases meet a preset target. These individually tailored discounts are the subject of this chapter. Because the buyer is not entitled to the discount until its purchases reach the target, the discount necessarily comes in the form of a rebate. Where the target is near the buyer’s total requirements of the product, as is often the case, the effects of this discount resemble those of an exclusive-supply contract. While many models of single-product rebates involve sales either to atomistic final buyers or intermediaries without market power, the rebates examined here (because they are tailored to each customer) are generally made to buyers large enough to make such individual tailoring practicable, and they are frequently made in sales to firms with substantial buying power. Hence they typically represent a combination of third-degree price discrimination by the seller and (in the case of large buyers) bilateral bargaining between the seller and buyer. These “power buyers”2 add a major ele-

chapter seven

118

ment of analytic complexity to realistic models of both single-product and bundled rebates.3 Discounts tailored to the circumstances of particular buyers can be called “target rebates” (because the rebates are often earned after the buyer’s accumulated purchases reach a specified target amount), or “fidelity” or “loyalty” rebates (because they have the effect of maintaining the buyer’s loyalty to the seller as a source of supply). The term loyalty rebates will be used here. Although loyalty rebates have been the subject of antitrust concern in Europe for some time, only a few US cases have considered their lawfulness, and they have generally been upheld. The US Experience While the European authorities generally view loyalty rebates granted by dominant firms as unlawful, US courts have been reluctant to condemn them. European and US courts as well as their respective antitrust enforcement authorities tend to focus on different aspects of these rebates. European authorities usually consider exclusive-supply contracts— “single branding” in EU rhetoric— by dominant firms abusive and therefore unlawful. Accordingly, loyalty rebates, which have the effect of providing incentives to customers to continue purchasing from the same seller, are treated with suspicion in Europe. Where these rebates are likely to produce the same effect as an exclusive-supply contract they are deemed to constitute an abuse of dominant position. American courts, by contrast, appear more tolerant of single-product (or loyalty) rebates. American courts also relate loyalty rebates to exclusive-supply agreements. But the Rule of Reason governs exclusive-supply agreements in the United States, and agreements that collectively involve less than 30 or 40 percent of the market are generally viewed as per se lawful. Moreover, because American courts view rebates or discounts as a form of price-cutting, behavior that is at the core of competitive activity, they are especially reluctant to interfere with it. Only when the price of a good including rebate falls below average variable cost (and therefore into the range of predatory behavior) are US courts likely to overcome this reluctance. The leading US case on single-product rebates until quite recently was the Eighth Circuit’s Concord Boat Corp. v. Brunswick Corp.4 of 2000, but then the Ninth Circuit grappled with single-product rebates a decade later, and these decisions may turn out to be more important precedents.5 The Brunswick case involved a supplier of “stern drive engines” for recreational boats, holding a 75 percent or more market share at various times relevant to the case. Brunswick offered discounts of 3 percent, 2 percent, and 1 percent

single-product loyalty rebates

119

to boat builders, depending on the percentage of their requirements that they took from Brunswick. Initially, boat builders that took 80 percent of their requirements from Brunswick were entitled to the 3 percent discount, but later the top discount of 3 percent was offered to builders that took only 70 percent from the firm. (Additionally, the Third Circuit’s decision in ZF Meritor, LLC v. Eaton Corp.,6 discussed in chapter 6, involved a defendant that was employing targeted rebates in a successful attempt to secure the entire market for itself. Because the court decided the case on the grounds of the defendant’s exclusive-supply contracts, we have followed the court’s approach and treated Meritor as an exclusive-supply case.) The Eighth Circuit analyzed Brunswick’s discount system under Section 1 of the Sherman Act using the criteria applicable to exclusive-supply contracts: the foreclosure of rival suppliers from a substantial share of the market, the duration of any exclusive arrangement, and the height of entry barriers. In the Brunswick case, boat-builder-purchasers claimed that Brunswick’s use of loyalty rebates enabled it to acquire a monopoly position in the engine market, forcing them to pay monopoly prices for its engines. The court determined, however, that the plaintiffs had failed to produce substantial evidence of foreclosure because boat builders were free to purchase their engines from rival suppliers at any time and did in fact switch when rival suppliers offered greater discounts. The court also found that entry into the engine market was relatively easy, and as a result the defendant would have been unable to control prices through exclusive dealing. In this connection, because Brunswick’s discounted prices were significantly above its costs, its discounts would not have discouraged entry.7 The court applied a predatory pricing approach to its evaluation of the discount system under Section 2 of the Sherman Act. Because Brunswick’s discount system generated prices that were above its costs, the court viewed them as prima facie lawful. It explicitly rejected the relevance of cases dealing with bundled discounts to the resolution of controversies involving single products.8 Further, it accepted Brunswick’s business justification, which was that it offered its discounts in an attempt to sell its products.9 Finally, it again assessed the exclusive dealing issue, this time under Section 2. On this issue, the court concluded that the discount program was lawful because it left customers free to purchase up to 30 percent of their needs from rival sources without foregoing the discount, that customers were free to leave Brunswick for other suppliers at any time, and that they did so when rivals offered better discounts.10 As noted, a decade after the Eighth Circuit’s decision in Brunswick, the Ninth Circuit decided two cases involving single-product discounts. In both cases, Tyco Health Care Group was the defendant and both involved

chapter seven

120

two types of distribution agreements Tyco employed: market-share agreements with hospital customers and group-purchasing agreements negotiated by Tyco with group purchasing organizations (consortiums of healthcare providers). Tyco was the dominant seller in the pulse oximetry market because it entered the market early and held a patent on important early technology. After the expiration of its initial patent, Tyco developed further technological improvements and a new line of equipment and marketed its products using the described agreements to foster sales. Under the market share discount agreements, customers (hospitals or groups of hospitals) were entitled to purchase Tyco products at discounts off list prices if they committed to purchase a minimum percentage of their pulse oximetry requirements from Tyco. Under the agreements, the greater the percentage of the customer’s requirements purchased from Tyco, the greater the discount. Tyco’s sole-source agreements were negotiated with group purchasing organizations (consortiums of health-care providers, or “GPOs”). Under Tyco’s solesource agreements, a GPO agreed that it would not enter into a purchasing contract with any other vendor of pulse oximetry products, and Tyco offered a deeper discount. In the latter of the two cases (Allied Orthopedic) the court ruled that both the market-share and the sole-supplier agreements were reasonable restraints that did not obligate Tyco’s customers to purchase anything from Tyco and that the agreements merely provided discounts benefiting purchasers who wanted Tyco-brand equipment while permitting them to purchase less expensive generic equipment. Rival suppliers had only to offer a better product or a better price. In the court’s view the freedom of customers to opt out of Tyco products at any time made the arrangements reasonable and therefore lawful under Section 1. The plaintiffs’ Section 2 challenge was based on Tyco’s design changes that removed new product compatibility with generic complements. The court, however, considered the redesign lawful.11 In the earlier case (Masimo), the court upheld a jury determination that Tyco’s sole-source and marketing agreements were unlawful. Subsequently (in Allied Orthopedic), the court explained the earlier Masimo decision on two grounds: (1) Tyco’s sole-source agreements involved in Masimo contractually obligated GPO members to purchase a set percentage of their pulse oximetry requirements from Tyco, whereas the sole source agreements involved in Allied Orthopedic did not; and (2) Tyco’s R-Cal patent was still in effect during the time period involved in Masimo, so customers needing sensors to use with their installed base of monitors would have to buy them from Tyco.12 As a consequence, Tyco’s discount (on all purchases from Tyco) would be computed on a base that included necessary purchases.

single-product loyalty rebates

121

This would render it more difficult for independent suppliers of sensors and monitors to compete, as they would not only have to meet Tyco’s prices on competitive products but would also have to meet prices that included discounts on other products (replacement R-Cal sensors) for which the independent suppliers could not compete. The Ninth Circuit’s position on loyalty discounts appears to be nuanced. A short-term loyalty discount leaves customers free to purchase from other suppliers when price and quality factors make them attractive. In the court’s view, the loss of a customer’s cumulative discount when it buys from an alternative source does not normally affect the lawfulness of the arrangement. In some circumstance, however, it will. When a customer is compelled to buy products from a particular supplier (as was the case for customers needing replacements for R-Cal sensors during the period before the expiration of the patent), then the cumulative discount on all of the customer’s purchases is inflated by the discount on these necessary purchases. In those circumstances, loyalty discounts become problematic in the Ninth Circuit. As will be seen below, in the European Union all tailored volume discounts are problematic. In December 2009— one month before the Ninth Circuit’s decision in the second Tyco Health Care case— the Federal Trade Commission (FTC) brought a proceeding against Intel, charging it with using loyalty rebates as a means of securing exclusive-supply arrangements, charges that closely resembled Intel practices the European Commission condemned in a decision issued the preceding May and discussed below. Intel accepted a consent order barring those practices in August 2010.13 The Intel case involved the market for x86 microprocessors, a market that is an effective duopoly, shared between Intel (with 65 percent of that market) and AMD.14 In the view of the FTC, Intel’s behavior was designed to disadvantage its rival, AMD, in accessing customers and constituted monopoly maintenance. The FTC brought this action under Section 5 of the Federal Trade Commission Act, although its theory of unlawfulness drew heavily from the law governing monopolization. A fair reading of its complaint— and of its subsequently issued consent order— strongly suggests that the FTC was using Section 5 to expand its powers into the substantive area covered by Section 2 of the Sherman Act without conforming to the case law restraints that would have bound a court in a Section 2 case. The differences between the antitrust rules that courts apply and the FTC’s Intel decision are outlined below. The FTC viewed Intel’s discounts and rebates to several large computer manufacturers as impeding sales by its rival AMD. The evolving US law on single-product discounts generally requires a showing that the dis-

chapter seven

122

counting seller has offered prices (when adjusted for the rebate) that are below its costs. This below-cost approach has been taken in Brunswick and Tyco Healthcare, both involving single-product rebates, and was taken in Nicsand, an exclusive-supply case, where a monopolist secured the acceptance of exclusive-distribution arrangements by offering lower prices than its single rival was willing to offer. The below-cost approach comes from predatory pricing cases where the most commonly used criterion for cost is “average variable cost.” In its initial complaint, however, the FTC charged Intel with granting discounts that effectively brought its prices to levels that were “below cost,” defined as “average variable cost plus an appropriate level of contribution towards sunk costs.”15 In other words, the FTC was not charging Intel for pricing at levels below average variable cost. This was not, however, a rejection by the FTC of all mainline analysis. Rather, the FTC’s definition of cost in its complaint appears to match Baumol’s “average incremental cost” as used by the European Commission (LAIC).16 The consent order forbids Intel to use discounts or rebates that are conditioned on a customer achieving a sales target, but that order mentions below-cost sales only in conjunction with the sale of a bundle of a microprocessor and a chipset.17 Here cost is defined to mean Intel’s “product cost of sales,” as that term is used in Intel’s ordinary course of business minus depreciation. This cost standard appears more like AAC than LAIC, but it clearly differs sharply from the addition of a share of sunk costs suggested as appropriate in the complaint. Does this difference in the cost definition and its application reflect a retreat by the FTC from a more aggressive earlier position? No, not in conjunction with single-product rebates, because the sweeping prohibition of all target-based rebates in the consent order renders the cost definition immaterial.18 It was only with bundled discounts that a determination of cost remained necessary at the time of the order.19 Because Intel settled with the FTC, there was no adjudicative determination of either the facts or the law. We know the FTC charged Intel with offering rebates that brought its adjusted prices to levels that were less than its long-run incremental cost. This was what the European Commission also charged. But we have no FTC findings of fact demonstrating that this was in fact the case. The evidence in the FTC record, however, included the decision of the European Commission, which complaint counsel introduced as evidence of the facts. The procedural posture of the case has meant that the FTC has not been required to show the market effects of the behavior it condemned. All of these factors (the introduction of the European Commission decision, the use of long-run incremental cost, the absence of finding about market effects) in the aggregate, suggest, but do not demonstrate, that the

single-product loyalty rebates

123

FTC’s Intel case was an attempt by that body to incorporate into US law the antitrust approach of the European Commission. The EU Experience The “Suction Effect”. The European Commission and a number of European

antitrust observers argue that loyalty rebates are anticompetitive because they make it increasingly difficult for rivals, including possible new entrants into the industry, to sell to the customers of a firm that is offering those rebates. Thus buyers accorded single-product loyalty rebates are increasingly tied to the seller as their total purchases approach the target amount. For example, consider the case of a seller, firm X, who offers widgets at a price of $10, but offers a $1 per unit rebate to buyers who buy 100,000 widgets over the course of a year. For buyers who expect to purchase 100,000 widgets during the year, this may be an attractive offer. They have an incentive to confine their purchases to that one seller. When such a buyer purchases the first widget, it likely considers competing offers from rival suppliers, who may also be offering discounts. When the buyer purchases its second widget, it will incur a slight cost should it decide to switch suppliers, say to firm Y. Assuming that firm Y was offering an identical target rebate, the buyer then would forfeit the $1 rebate on its first purchase that it would have received from firm X had it continued to deal with firm X until its purchases reached the target amount. If the buyer switches after purchasing its second widget from firm X, it would forfeit $2. Thus the cost of switching suppliers increases as the buyer’s purchases from firm X increase. The cost of switching increases gradually at first but grows rapidly later on. After the buyer has purchased 90,000 widgets, the cost of switching would be $90,000. As the hypothetical shows, as a buyer’s purchases increase, the seller is effectively offering that buyer a greater incentive to continue purchasing from that seller because he pays less and less for incremental units. The price for an incremental unit is effectively the discount price ($9) per unit less the rebate on past purchases. As the buyer’s purchases increase, the rebate on past purchases increases, eventually growing to the point where the per-unit price of incremental units is negative.20 Figure 7.1 illustrates the hypothetical discussed above, where the list price is $10, the rebate is $1, and the target is 100,000 units. The rapid decline in the effective unit price is apparent as the purchases approach the target amount. The effective unit price, factoring in the rebate, starts at $9 and then gradually declines as the rebates accumulate. When the buyer has purchased

chapter seven

FIGURE 7.1.

124

unit price of incremental units.

90,000 units, the effective unit price for the additional 10,000 that will take the buyer to the target is zero. At 95,000 units, the effective unit price for the remaining 5,000 units necessary to reach the target is negative ten dollars (–$10). The unit price continues to fall rapidly. This rapid fall in the unit price of additional purchases is called a “suction effect” in Europe because of the increasing incentive of the buyer to continue purchasing from the same supplier.21 At 100,000 units, the price for additional units climbs abruptly to $9 and holds steady thereafter. The potential significance of the mechanism generating the “suction effect” becomes clear when viewed from the standpoint of an alternative seller, perhaps an entrant. Assume that the purchaser has the characteristics previously described and that this purchaser is buying above the target amount at 105,000 units. Assume further that the entrant faces sharply declining costs with a minimum efficient scale of 10,000, at which point its costs match those of the incumbent. The entrant, however, has little chance of selling the 10,000 units to the purchaser. If the purchaser has already bought 95,000 units from the incumbent, it would lose $95,000 by purchasing the next 5,000 units from the entrant. After attaining the target amount of purchases (100,000 units), the purchaser would be free to buy additional units from others (including the entrant) at prices of $9 or below without losing money. If the purchaser in question is the only market for the entrant’s goods, and if the purchaser’s needs do not reach 110,000 units, the entrant would be incapable of attaining minimum efficient scale.22 This argument demonstrates how much foreclosure turns on the ability of the incumbent seller either to

single-product loyalty rebates

125

tailor targets to each buyer individually or to an average firm in a situation of small variation of sales among firms. The European Commission’s Discussion Paper and Guidance. The European

Commission initially analyzed the competitive significance of loyalty rebates with the suction effect and their competitive significance in a 2005 staff discussion paper. Subsequently, the Commission issued a guidance on its enforcement priorities that incorporated the core of its discussion paper analysis. We first consider the discussion paper’s analysis of fidelity rebates. Following an approach similar to one taken in the United States in 1996 by a US district court in Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc.23 in evaluating bundled discounts, the discussion paper employed an equally efficient competitor test for identifying the kinds of loyalty rebates that posed foreclosure problems. Thus if the rebate brought the dominant seller’s effective price to a buyer below the seller’s cost, then the rebates would be deemed to be capable of a foreclosure effect. The paper frames the issue in terms of an “entrant” challenging the dominant company and employs a comparison between “a required share” and a “commercially viable share” of the market in question to develop a standard governing the lawfulness of loyalty rebates. For a given discount percentage, the paper first asks: “how big a share of customers’ requirements on the average the entrant must minimally capture so that the effective price is at least as high as the average total cost of the dominant company.”24 This is the “required share,” the average share of all purchasers that an equally efficient entrant must capture in order to sell at a price that is not below its own costs. In terms of the diagram above, this means simply how far back toward the y axis must the entrant’s quantity be able to reach so that matching the discount of the incumbent firm does not lower the effective price below the incumbent firm’s cost. Next, the paper’s analysis employs the concept of a “commercially viable share,” which is “the share of customers’ requirements an efficient entrant can reasonably be expected to capture.” This term draws from the paper’s discussion of single branding, where it argues that purchasers often buy a substantial part of their needs of a product from the dominant producer because their own customers demand the dominant firm’s brand or because capacity constraints on rival suppliers prevent them from supplying more. But it must be noted that the concept of “commercially viable share” is then posited to stem from two utterly different proximate sources: one related to demand and the other to supply. A concrete example can illustrate that the latter is likely only derivative from the former. Many computer manufacturers will buy a portion of their microprocessors from Intel because of the

chapter seven

126

attraction of the Intel brand to ultimate consumers, and, to the extent that this factor is important, it might change only slowly and unpredictably over time. The demand for Intel processors at a given time dampens market penetration by firms such as AMD that will doubtless have production capacity geared to this reality. But an Intel rival should be expected to experience only short-run capacity constraints— ones that will be alleviated as expected demand warrants; so demand for the output of particular suppliers appears to be the ultimate driver of the “commercially viable share.” The “commercially viable share” identifies the amount of purchaser requirements that is open to alternative suppliers. The “equally efficient” entrant whose situation is proxied by the price-cost margin of the incumbent, as previously explained, cannot compete unless a sufficiently large part of the incumbent sales can be displaced so that loss-inducing implicit prices need not be matched. Foreclosure is therefore established by comparing the required share with the commercially viable share. If the required share is greater than the commercially viable share— it is not feasible to reduce the incumbent’s sales sufficiently to avoid short-run losses with the rebate at a specified level— then that rebate is deemed to produce a foreclosure effect. And there is a final consideration suggested earlier. Even if a sufficiently large volume of sales is not blocked by brand preference (the required share is less than the commercially viable share) the scale of production of the entering firm might be too low for economies of scale to be realized. The Commission here has moved from an approach to culpability for foreclosure that is based entirely on calculations that the incumbent could estimate from its own data to a calculation that requires estimates of the costs and sales of another firm. The identification of a commercially viable share marks a sharp departure in the discussion of loyalty rebates for a single product. By separating customer purchases into two categories, the commercially viable share (or contestable amount) that is open to the dominant firm’s rivals and the required share that the customer feels compelled to purchase from the dominant firm (the compelled amount), a critical ambiguity has arisen. The single product is not really a single product after all. There is no longer an assumption of complete displacement of one supplier by another but rather an assumption that the purchaser needs some of the dominant firm’s “brand.” This may be realistic, but without exploring why complete displacement is not possible, the approach simply posits that it is not and also requires an assumption about the minimum amount needed from the dominant firm. How does one estimate how much of the market is really open to contestation? How would a potential defendant know whether its discount or

single-product loyalty rebates

127

rebate was excessive by the “equally efficient” competitor standard? The Commission offers no characterization of the two competing product varieties beyond assigning categories. A very similar problem arises in the following chapter where goods sold in bundles are divided into “monopoly” and “competitive” products for an examination of price-cost relations using a rebate attribution test. Dividing products into two groups is problematic for what is ultimately the same reason as the employment of “commercial viability”: each differentiated product has its own demand determinants— and especially substitution characteristics— and competitive analysis must explore the properties of those individual demands instead of simply assigning products to categories. The discussion paper gives a formula for computing the required share, which is R × P/(P – ATC). P is list price, ATC is average total cost, and R is the rebate stated in percentage terms, so that R × P would give the rebate in dollar or euro terms. The paper gives an example in which list price is 100, the percentage rebate is 5 percent, and average total cost is 75.25 Therefore, the numerator of the fraction is 5 (= 100 × 5%) and the denominator is 25 (= 100 – 75). The fraction provides a ratio of the rebate to profit. In the example, that ratio is 1:5 so the rebate is equal to 20 percent of the seller’s nominal profit (that is, the profit computed without the rebate). That fraction also indicates the point at which the seller’s sales to any particular buyer bring its effective price down to the level of its costs. That will be at the point at which the buyer has purchased 80 percent of the target amount. This implies that a rival, attempting to sell to that buyer, must penetrate to at least 20 percent of the buyer’s needs. This is the “required share.” If it were assumed that the smaller supplier could successfully compete for 40 percent of the total, the allowable rebate could be twice as high. The Commission’s guidance document, published in the official journal early in 2009,26 closely follows the discussion paper’s analysis. Like the discussion paper, the guidance uses the equally efficient competitor standard to evaluate fidelity rebates. It substitutes “contestable” for the discussion paper’s “commercially viable” share but with the same meaning. The guidance refines the discussion paper’s approach to cost, however. The discussion paper used average total cost as the standard for determining whether an equally efficient competitor would be able to meet the dominant seller’s effective price, that is, list price less rebate. The guidance paper substitutes two types of cost: average avoidable cost (AAC) and long-run average incremental cost (LAIC).27 If the price is below AAC, an existing firm cannot meet its out-of-pocket costs of production; if the price is below LAIC, that firm cannot profitably expand its production capacity. When the effective price is below AAC, the Commission will assume the rebate system is ca-

chapter seven

128

pable of foreclosing equally efficient competitors. When the effective price falls between AAC and LAIC, the Commission will investigate, inquiring, for example, as to whether rival suppliers have counterstrategies available to them. And when the effective price is above LAIC, the rebate system is not equated with foreclosure. An Examination of the Discussion Paper and the Guidance. The discussion paper’s required-share analysis focuses on the ability of an equally efficient rival firm to capture a commercially viable share of the business of a customer of the seller offering a rebate. The guidance also uses the measure of an equally efficient rival supplier. The use of an equally efficient competitor as a reference point suggests that the European Commission shares the objective of American antitrust law of encouraging aggressive, but nonpredatory, price competition. But the EU approach contains some ambiguity that applications of the guidance will reveal. Moreover, as the Intel case discussed below suggests, the EU objective may in fact be different from that of US policy. Both the discussion paper and the guidance attempt to identify the effects of loyalty rebates on the market. Yet the analysis in both seems to overemphasize the effects of loyalty rebates on particular customers. As the discussion paper puts it: “the question is whether the rebate system hinders them [that is, competitors of the dominant firm] from supplying commercially viable amounts to individual customers.” The discussion paper apparently seeks to bridge the gap between the market and a single customer by assuming the analyzed customer is an average or representative purchaser. Thus the discussion paper speaks of “how big a share of customers’ requirements on the average” the entrant would have to capture in order to avoid selling below cost.28 The guidance uses identical language.29 If the focus were directly on the market, it would be more straightforward to speak of how big an aggregate share of all customers’ requirements the entrant would need to capture. There would be no need to speak of customers’ requirements “on the average.” But if these documents determine market effects from generalizing effects on a representative purchaser, then they ignore variations in the apparent strategy of the dominant seller toward specific buyers and variations in the responses of particular buyers to the dominant seller’s rebates and the market-wide ramifications of those variations. The discussion paper seems ambiguous about how it employs the required share and commercially viable share to reach conclusions. The guidance is also unclear in its corresponding analysis of whether an equally efficient rival supplier would be able to offer an above-cost price on the contestable parts of customer demand. Sometimes the guidance speaks

single-product loyalty rebates

129

of “a customer’s purchase requirements” and sometimes it refers to customers in the aggregate. The situation of individual purchasers may vary for many reasons. Even those firms that buy only for resale to final customers may differ widely in their purchasing behavior. In particular, the contestable share may vary with geographic variations in the mix of final demand or by the structure of final demand faced by various types of resellers. Complications increase considerably when there are several major sellers; different firms may “dominate” in sales to particular buyers. The ambiguity of the approach can be explored with the discussion paper’s example (5 percent rebate, $100 list price, $75 ATC, and a 20 percent required share). First, if the required share is 20 percent on average, this does not tell us much about the actual sales success of the entrant, which may be supplying 20 percent of the entire market but capturing much higher percentages of the requirements of particular customers. An entrant that is supplying 20 percent of the market by supplying 20 percent of each customer’s needs is selling at cost because the entrant’s effective price must be reduced to neutralize the buyer’s loss of the dominant seller’s rebate. But the reasons the discussion paper gives for limiting the contestable part of demand (brand preferences of ultimate customers, and capacity constraints on alternative suppliers) may apply differentially to different customers so that the entrant may be supplying substantial amounts (well over the required shares) of the needs of particular customers. An entrant that is supplying 20 percent of the market by supplying 30 or 40 percent of the needs of some customers is earning substantial profits because the price reduction required in these specific situations to neutralize the dominant seller’s discount is substantially less. The discussion paper does not tell us how to draw conclusions from its analysis. The paper tells us that if the required share exceeds the commercially viable share in sales to a representative customer, foreclosure is likely. But in that circumstance, the entrant could not sell to that customer except at a loss, so the foreclosure of that customer has already occurred. Instead, the discussion paper may be referencing the required and commercially viable shares to the aggregate market (rather than to a representative customer). In that case, the analysis should be applied to the entire market. If the required share of the market is larger than the commercially viable share of the market open to an entrant, the foreclosure has already occurred. At all events, a serious problem arises if and when the Commission uses an analysis of a representative customer to draw inferences about aggregate market effects. The use of a representative buyer having average requirements as an analytical tool is not inconsistent with an ultimate focus upon the entire

chapter seven

130

market.30 Indeed, the discussion paper explicitly focuses on overall market effects. It does this by referencing the foreclosure effect of rebate systems to the foreclosure effect of tying arrangements, stating that “the potential negative effects will in general depend upon the size of the tied market share, and the guidance does apply a substantial share-like approach in examining overall market impact.”31 An alternative application of the discussion paper analysis is more complex and involves more judgment. The seller may provide rebates to some customers and not to others. And those rebates may vary. There is no way the market effects can be approximated by the summing of the effects on a single representative customer over similar firms because the extent and scope of the rebating itself turns on the dominant seller’s perception and judgments of differences among its customers. In these circumstances, the discussion paper says that the Commission “will investigate whether or not these selected buyers are of particular importance for the possibilities of entry or expansion of competitors.”32 Similar language is contained in the guidance.33 Additional concerns would presumably include economies of scale in production and the feasible rate of a challenging firm’s expansion. This part of the Commission’s foreclosure determination appears to involve more subjective factors and, accordingly, seems less predictable. The Commission’s discussion paper and guidance shift assumptions sharply from a simple tailored volume discount case. There the issue was the displacement of supply of a product by a substitute product from another source. In the more complex model, there is an assumption of market power by the dominant firm that renders a part of the purchaser’s requirements impervious to the offer made by the contending firm. For this reason, instead of a possibly level playing field at the beginning of each new period for rebate calculation, outsiders face an ongoing and inescapable disadvantage because they can never more than partially displace the rival firm’s offerings. The Commission’s assumptions mean that what may look from one perspective like a special version of the single-product discount case more closely resembles the multiproduct case to be examined in the following chapter. In those bundled discount and rebate situations, analysts typically rely on a simple assumption of a “monopoly” and a “competitive” good. But some brands of a nominally “competitive” product are protected by such high product differentiation demand barriers that they would pass a test for separate product markets based on the profitability of a certain price-cost margin for a sustained period of time. The Commission documents seem to deal with the degree of product competition between or among substitutes through a kind of ad hoc judgment. Observers will differ about the persua-

single-product loyalty rebates

131

siveness or applicability of the results in specific contexts. In particular, how should a dominant firm decide what part of the market should be regarded as “contestable” for purposes of discovering its own vulnerability to the “equally efficient competitor” test?” Prokent-Tomra. The decisions of the General Court and the Court of Justice

in Prokent-Tomra add a new gloss to the EU law on fidelity rebates.34 The Commission decision was issued in March 2006, three months after the release of the staff discussion paper. Its decision was affirmed by the General Court in September 2010 and by the Court of Justice in April 2012. The product involved was reverse vending machines, machines that collect drink containers for cleaning and reuse or for recycling, and the servicing and maintenance of those machines. Perhaps because the Commission decision followed the discussion paper by only a few months, the decision employs only a partially developed version of the discussion paper’s analytical framework. As in the discussion paper, the suction effect plays the key role, showing how the price that a rival supplier would have to meet declines rapidly as a customer’s purchases approach the target amount. Like the discussion paper, Tomra breaks demand into contestable and noncontestable segments.35 It focuses on how the contestable part of demand may be insufficient to permit a rival to meet or undercut the dominant seller’s price without selling below (the dominant seller’s) cost. The data the Commission relied upon to support its decision raises significant questions about its analysis. The Commission used a shortcut to estimate the noncontestable share; it was simply that part of national demand covered by Tomra’s exclusive-supply contracts. The data were limited and subject to wide fluctuations.36 Tomra held extremely high national market shares over the period in question, 1998– 2002, that typically varied between 70 percent and virtually the entire market.37 Nevertheless, the total share other firms held was sometimes quite substantial. Moreover, the large Tomra market shares and the generally much smaller share under exclusive-supply contracts suggest that its dominant position is explained by the attractiveness of its product. Most of Tomra’s sales were in the contestable part of the market. Tomra differs from the discussion paper analysis in the reasons it gives for part of demand being noncontestable. The discussion paper discusses strong brand preferences and capacity constraints on alternative suppliers. By contrast, the Tomra case explains the noncontestable part of demand as due to exclusive-supply contracts already entered into by Tomra’s customers. On that rationale, however, customers would be free to purchase from alternative sources after the current contract period ends. In Tomra, the

chapter seven

132

contracts generally did not exceed a period of three years.38 The Commission argued that the effective period of exclusion was longer because of the relatively long life of the machines and incentives for customers to bunch purchases in order to earn a discount.39 The Tomra analysis thus focuses on existing exclusive-supply contracts as a base from which rebates are used to expand into the contestable portions of demand. Notably, however, the Commission never demonstrated any specific anticompetitive outcomes from the “suction effect” in Tomra. In particular, there was no demonstration that the “suction effect” drove Tomra’s effective prices to any definite relation to its costs. Instead, as two Commission analysts argued, Tomra’s anticompetitive behavior can be inferred from market outcomes. Tomra’s overall EU market share over the period was 85– 90 percent, and there was “no successful entry into any of the relevant national markets during the time frame covered by the decision. On the contrary, some of the competitors left the market due to either insolvency or acquisition.” Moreover, there had been “several unsuccessful entries into the market, while entry is neither technically particularly difficult, nor exceedingly costly.”40 None of this seems to have mattered to the General Court or to the Court of Justice. In upholding the Commission on appeal, the General Court ruled that the foreclosure of 40 percent of the market by Tomra was “a considerable proportion (two-fifths) of total demand,”41 and that a foreclosure of a substantial part of the market cannot be justified by showing that the contestable part of the market is still sufficient to accommodate a limited number of competitors.”42 In upholding the General Court, the Court of Justice repeated these rulings.43 In ruling that exclusive-supply agreements by dominant producers involving a “substantial share” of the market were per se illegal, these courts were recognizing and applying the Hoffmann-La Roche precedent.44 In so doing the reviewing courts were also implicitly ruling that the Commission’s elaborate market analysis was irrelevant. The Court of Justice was quite explicit on this issue. The Commission’s guidance discusses economies of scale as among those market conditions relevant to a determination of anticompetitive foreclosure.45 The Court of Justice, however, ruled that a “minimum viable scale” test was not necessary because foreclosure of 40 percent of the market was sufficient to violate Article 102.46 The 40 percent foreclosure condemned here appears to be a level that is tolerable in US antitrust practice, so the Court’s rejection of a minimum viable scale test runs counter to current US antitrust thinking.47 This appears to be another instance in which a so-called effects-based analysis by the Commission is superseded by judicial decisions substituting a form-based approach. The Court of Justice, however, managed to obscure

single-product loyalty rebates

133

the extent to which it is unwilling to defer to the Commission’s guidance by observing that the Commission’s decision (made in 2006) predated the guidance (issued in 2009) and therefore was formally inapplicable to that decision.48 The Intel Case. In May 2009, the European Commission condemned the Intel

Corporation for granting rebates to several large computer manufacturers (Dell, Hewlett-Packard, Lenovo, Acer) on their purchases of Intel microprocessors. The Commission also condemned Intel’s payments to a large European retail chain for stocking its shelves with computers powered by Intel. The Commission employed an analysis that was similar to that of the discussion paper and the guidance for this case. Like those documents, contestable shares of buyers’ purchases were compared with required shares. As in the discussion paper and the guidance, the contestable share of a buyer’s purchases was the (percentage) amount the buyer was willing to obtain from an alternative source, if it was satisfied with the terms that source offered. As in the both the discussion paper and the guidance, the Commission looked to see whether an equally efficient supplier could supply its product at a price that matched the effective price of principal supplier and was not below cost. Although the Commission embraced the approach it had adopted in the guidance (and that had been earlier outlined in the discussion paper), it has impeded the efforts of observers to critically assess its application of this analysis in Intel because it has removed almost all information about sales, market shares, and contestable shares from the opinion it has made public. Since information about these matters can be gained from other sources, however, it is still possible to evaluate the Commission’s Intel decision. Even absent additional market information, the Intel opinion raises critical questions about the Commission’s method. First, the Commission calculated the required share for each customer and then compared it with the contested share for that customer. The discussion below points out that the method does not necessarily reflect the competitive impact of the Intel rebates because it appears not to include all of the purchasers of x86 microprocessors. Accordingly, even if the Intel rebates dissuaded the computer manufacturers to whom the rebates were offered (Dell, Hewlett-Packard, Lenovo, Acer) from purchasing AMD microprocessors, that alone would not have excluded AMD from the microprocessor market. Second, the Commission referred to AMD’s average share of the consumer market in Europe as having been around 33 percent during the period from 2003 to 2007.49 This suggests that AMD was unlikely to be driven from the consumer market,

chapter seven

134

and although the business market is not included in the estimated share, it appears that AMD’s overall share of the global x86 microprocessor market during the relevant time period was somewhere between 20 and 30 percent.50 A Critical Examination of the Commission’s Intel Case in Comparative Perspective. The Intel rebates that the Commission determined to be unlawful dif-

fered among Intel’s customers. The rebates granted to Dell applied to all of Dell’s microprocessor purchases.51 The rebates awarded to Hewlett-Packard were focused on processors used on HP’s desktop machines sold to corporate users.52 The rebates granted to Lenovo and Acer were keyed to their imposing delays on their marketing of machines powered by AMD processors, to restricting the geographic areas in which such machines would be sold, and to restricting their offerings to smaller enterprises. It will be recalled that both the discussion paper of 2005 and the guidance of 2009 employed an “as efficient competitor” analysis. This analytical approach underlies the US law dealing with predatory pricing. It was employed by the US district court in the Ortho litigation in 1996 for dealing with possible predatory pricing in bundled discounts.53 Later (and subsequent to the discussion paper but prior to the guidance) the Antitrust Modernization Commission in 2007 embraced an “as efficient competitor” analysis as a technique for dealing with bundled discounts and rebates.54 As explained in chapter 5, predatory pricing under US antitrust law is an offense under the monopolization and attempted monopolization clause of the Sherman Act as well as an offense under the Robinson-Patman Act. Because of the danger of inhibiting socially desirable price-cutting, however, the Supreme Court has been vigilant in limiting the offense to situations in which a seller with substantial market power is selling below cost in an attempt to drive rivals to exit the market (or to discipline them into maintaining supracompetitive prices), and there is a dangerous probability that the seller will be able to recoup its losses in the postpredatory period.55 Because recoupment will be impossible if rivals are free to compete away supracompetitive profits in the postpredatory period, the affected market must necessarily be protected by entry barriers. Professors Areeda and Turner, who designed this predatory pricing analysis, argued that a seller who prices at or above its marginal cost cannot drive out an equally efficient competitor.56 Although the European Commission has adopted an “as efficient competitor” standard for determining whether a seller’s rebates bring its effective prices to below-cost levels, it has not explored the market effects of the rebates in ways that most US courts would recognize. In Intel there was no determination that Intel would monopolize the market by driving out AMD, its only rival in the x86 microprocessor markets. The Commission did not

single-product loyalty rebates

135

examine how the market would be altered as a result of the Intel rebates. Indeed, the Commission appeared to acknowledge that machines equipped with AMD microprocessors were reaching all segments of the ultimate consumer market while AMD maintained a market share of more than 20 percent. The Commission focused on the market strength of the two largest computer manufacturers, Dell and Hewlett Packard, and their strategic importance in fostering consumer acceptance of alternatively sourced microprocessors.57 In particular, the Commission found that these two computer manufacturers could play an effective role in inducing corporate acceptance of machines equipped with AMD processors. It further found that Intel used its rebates and other payments to persuade these and other computer manufacturers to delay and/or limit their marketing of machines powered by AMD processors. As a result, consumers’ choices were restricted.58 In addition to attacking the rebates that Intel provided to the leading computer manufacturers, the Commission also determined that similar payments by Intel to a large European retailer conditioned on the retailer selling only computers powered by Intel microprocessors were also unlawful.59 Indeed, the focus of the Commission’s decision appears to be based not on an abuse of a dominant position equivalent to potential monopolization of the market by Intel, but on an abuse of a dominant position by restricting consumer choice. The Commission took a special view of consumer choice, but one that appears quite consistent with the overall character of EU-US differences. Thus when the Commission quoted Intel’s expert as stating that “European OC and server OEMs sold systems with AMD microprocessors in every segment and every price point during the SO period,”60 it chose not to dispute the truth of that statement but its relevance. According to the Commission, “the magnitude of the consumer harm associated with a lack of choice can be dependent on the remaining availability of close substitutes in the market.”61 Of course, there was a close substitute for AMD microprocessors available in the market: Intel microprocessors. Moreover, computers powered by AMD microprocessors were apparently widely available. The Commission concluded, however, that although the loss is likely to vary from one customer to another, Intel microprocessors were not adequate substitutes for AMD microprocessors for many customers. Therefore, even if the desired AMD microprocessor were available on computers of a different brand, that would not adequately remedy the loss of consumer choice.62 The Commission concluded that Intel’s actions “prevented important and genuinely different AMD-based products from ever being brought to the market in significant quantities.”63 This strong emphasis on consumer choice was

chapter seven

136

more explicit in the guidance64 than in the discussion paper that focuses on market structure and the ability of rivals and entrants to compete. The EU Intel case was followed in 2010 by the US FTC complaint and settlement with Intel discussed in the previous section. One major difference from the EU case is that the FTC complaint aims at the viability of price and product competition and not at contemporaneous product variety for the final purchaser. But a common feature of both cases is striking. In neither case was evidence produced of a clear danger that AMD would have been marginalized or driven from the market. Some commentators— both those in favor of the FTC’s action and those opposed— see the American action as a conscious attempt to move US intervention toward dominant firms closer to EU practice.65 But it must be stressed again that the FTC order was uncontested. Had Intel chosen to take the fight to the courts, the probable effect of Intel’s behavior on AMD and the overall market would likely have been subject to much closer examination. Further Analytical Refinements Commercially Viable Share v. Required Share. The paradigm case of an entrenched incumbent posits that the typical buyer must buy (and hence resell, either as is, or as an input or component) at least some of the dominant seller’s product. The buyer can purchase only from the dominant firm, but he cannot purchase only from the challenger. Leaving exclusive contracts and capacity constraints aside, this is a very different model from the one used by the US Eighth Circuit in the Brunswick case discussed earlier in the chapter. The EU approach simply assumes a minimum share for the “dominant” firm and therefore that the competing firm must bear an asymmetric burden in discounting for the rest of the buyer’s supply because the total value of a discount offered by the competing firm to the buyer is spread over only over the buyer’s residual purchases after it has already purchased from the dominant firm. This model parallels the argument developed in the next chapter about the bundled selling of a “monopoly” and a “competitive” good that has driven so much theoretical development in recent years. As will be argued further, it seems to solve by assumption problems of substitution among differentiated products that need to be carefully considered on a case-by-case basis. Differentiated products are, by definition, not perfect substitutes, but the reasons why vary sharply. Two very different cases illustrate the problem in the context of this discussion of sales to resellers: production inputs and final products. It is a frequent convention in industrial economics to assume that production inputs are purchased objectively, solely on grounds of func-

single-product loyalty rebates

137

tionality. But that convention suggests at least two further complications. One is that the performance of the input may unfold to professional buyers only though extended experience. This time-related factor brings a dynamic element to the analysis. Another complication may arise if final users have their own views about input quality that might vary from more expert opinion. For these and other reasons, the “contestable” share of such inputs may vary widely by buyer and may change substantially and quite quickly at the level of the entire market as well. Antitrust law in both the United States and the European Union has tried to formulate rules that provide safe harbors to firms striving to avoid antitrust liability; those rules usually can be self-applied as they generally incorporate information known to the firm, such as its prices and costs. Yet this potential for rapid change in the contestable share of inputs jeopardizes a dominant firm’s safe harbor calculations because, in the circumstances described, its own prices and costs must be augmented with possibly unknowable information on the contested share. The preceding case of a production input can be contrasted with another frequent pattern. Consumer products’ market positions sometimes rest on decades of promotion and brand acceptance. Hence, the assumption of a minimum share of the dominant firm’s variety of a product may well be a long-lived market parameter. In this context, the assumption of a “representative” buyer-reseller appears more realistic. Thus if all actual or potential competitors have the same production costs at every level of output, greater acceptance of the dominant firm’s product suggests a higher price-cost margin and hence a greater capacity for discounting (measured as a percentage of list price) than the competing firm would actually experience. In that circumstance the dominant firm would be the more efficient because it would be producing greater value for the same cost. Indeed, the dominant firm may be the more efficient in this sense, even when its costs are higher if its price-cost margin is greater. In any event, so long as the contestable share is not changing rapidly (as in the example in the preceding paragraph), the dominant firm would be able to calculate a safe harbor under the EU guidance that would nonetheless confront the competitor with the prospect of selling at a loss. Raising Rivals’ Costs. An alternative but complementary perspective on loy-

alty discounts found in the economics literature rests on the raising of rivals’ costs discussed in connection with exclusive-supply contracts in the previous chapter. This literature directly addresses the disadvantage visited upon the weaker firm or firms, and how much additional pricing power this would give to the dominant firm. The analysis does not turn on the weaker

chapter seven

138

firm or firms going of business, but simply a reduced ability to compete. The weakening of a nondominant firm’s position could result from a need to employ less cost-effective distribution channels and from a smaller scale of production. This would allow the dominant firm to provide a larger share of the total market at a higher price and possibly to share additional profits with distributors or input purchasers who buy only from the dominant firm. Brennan has suggested the use of this perspective and that any effective distribution foreclosure should be treated with a test similar to that used for mergers,66 but, as noted in the previous chapter, that approach leaves many questions open. From a consumer welfare point of view, the main question is presumably whether the output of the “product” goes up or down as a result of pricing innovations. And, given the complications, inter alia, of private branding for resale or, where the product is an input, changing product lines, this might be very difficult to determine even after the discounting scheme was firmly in place. Similarly, from a total welfare standpoint, any pricelimiting effect of relieving a distributional constraint through policy intervention would need to be weighed against possibility higher overall costs of production and distribution. Conclusion Although both jurisdictions now employ some version of predatory pricing analysis in their consideration of single-product discounts, this superficial similarity rests on a structure of considerable difference. US courts have generally found loyalty discounts to be procompetitive, although the Ninth Circuit’s explanation of its Masimo decision indicates movement in the direction of the European Commission’s guidance. In contrast, EU courts appear willing to ignore the complex models of the Commission and declare such discounting by dominant firms as abusive without considering specific competitive effects. In addition, the Commission’s approach accepts a more inclusive cost standard for computing predatory pricing and gives less attention to recoupment or lasting market impact than would be expected in the United States. Some observers have seen the FTC Intel case as an innovation because it appeared to embrace LAIC as a standard and therefore to move the United States in the direction the Commission favors, but the pricing conduct agreed in the settlement is difficult to interpret, and no adjudication was involved.

8

Bundled Discounts Introduction A seller that offers a lower price for a bundle of goods than it charges for the same goods when sold separately is engaged in a common and seemingly innocuous practice. Yet in some circumstances, this practice has raised antitrust concerns. The problem arises when a multiproduct seller offers a discount on a package of goods that an equally efficient rival, selling only one or a few components of the bundle, cannot meet. It is possible to construct a scenario in which a seller uses bundled discount pricing to drive rivals from the market and to acquire a monopoly for itself. So described, bundled discounts have sometimes been regarded as a species of predatory pricing. There are significant differences between bundling and predatory pricing. A seller engages in predatory pricing when it sells its goods at prices that are less than its costs, driving its rivals out of the market and thereby acquiring a monopoly for itself. In the United States predatory pricing by a single firm constitutes monopolization or attempted monopolization. The courts, greatly aided by scholars, have developed tests for identifying this behavior. As discussed in chapter 5 most courts now follow one or another version of the Areeda-Turner approach; a large seller (that has— or can obtain— the capacity to supply the entire market) must

chapter eight

140

be pricing its goods below some measure of cost before it can be deemed to be pricing predatorily. While Areeda and Turner favored marginal costs or its surrogate average variable cost, many have been persuaded that the most appropriate operational short-run standard is average avoidable cost: AAC. In addition, many commentators on both sides of the Atlantic would consider prices up to long-run average incremental cost (LAIC) as possibly predatory depending on the circumstances. One such circumstance is barriers to (re)entry into the market. In the predatory scenario, the successful predator recoups the losses it incurred during the predatory period from the monopoly revenues it earns subsequently. By contrast with predatory pricing, a seller employing bundling could reach the same end result without apparently selling any product below average total cost (or LAIC)— let alone average variable cost (or AAC)— a matter to be explored in some detail. In common with single-product discounting where there is a “noncontestable share”— in the bundling case there is a “monopoly” product that constitutes a “share” of the bundle— and, by contrast with exclusive dealing contracts, the element of time (as opposed to uncertainty) may be irrelevant to the exclusionary impact of the practice. The effect of bundled discounting on competition is both complicated and subject to great controversy. The difficulty of establishing the appropriate perspective in a given case is greatly increased by the common business practice of combining fixed volume and targeted rebates with a discount for bundling. Even if the criterion of not foreclosing an equally efficient firm is accepted as a critical element in the antitrust analysis, getting to that point often presents formidable theoretical and empirical challenges. Although bundled discounts have been the subject of US antitrust litigation in the past,1 the Third Circuit’s 2003 decision, in LePage’s Inc. v. 3M (Minnesota Mining & Mfg. Co.),2 generated widespread attention to this marketing technique and sharp differences of view among antitrust commentators. In that case, the court condemned 3M’s use of bundled discounts on a wide range of office products as unlawfully disadvantaging LePage’s, a competing supplier of transparent tape. In 2007 the Antitrust Modernization Commission (AMC) disapproved the Third Circuit’s analysis and provided its own.3 The Ninth Circuit thereafter developed its approach to an antitrust evaluation of bundled discounts, which only partially followed the AMC.4 Because bundled discounts involve single-firm conduct, their lawfulness in the United States depends on whether they meet the standards of either monopolization or attempted monopolization under Section 2 of the Sherman Act. Accordingly, only firms with a very large market share run a risk of antitrust liability for employing bundled discounts under this approach. Alternatively, bundled discounts could possibly be evaluated as

bundled discounts

141

ties.5 In that case, antitrust vulnerability could extend to firms with market shares not much more than 30 percent.6 The Troublesome Nature of Bundled Discounts The troublesome aspects of bundled discounts arise from their widespread use as marketing tools and the broad recognition that in the overwhelming number of cases, their use is procompetitive as a price-cutting tool. Legal scholars have labored to discover the appropriate analysis of bundling.7 As is the case with the sale of single products, varying buyer power adds great complexity to analyzing real situations.8 Bundled prices often result from bargaining between sellers trying to increase share and power buyers seeking better terms.9 The impact of price changes upstream on the welfare of final purchasers, in turn, depends importantly on the strength of competition in intermediate commerce. Most observers believe that power buyers pass at least some of their savings on to final purchasers.10 Nevertheless, the use of bundled discounts can, in some circumstances, drive customers away from more efficient producers toward less efficient producers who are marketing a wider range of products and also may raise final prices. They can also negatively affect market outcomes from the standpoint of the kind of consumer choice criterion that the European Union demonstrated in the Intel case. So described, bundled discounts bear a marked resemblance to predatory pricing. Yet bundled discounts do not easily fit judicial conceptions of predatory pricing. Neither the Supreme Court’s Brooke Group test11 for evaluating pricing challenged as predatory nor the Areeda-Turner test generally employed in predatory pricing cases by the lower courts (along with Brooke Group) work with bundled discounts, even though bundled discounts can generate exclusionary effects similar to predatory pricing. Under Brooke Group, pricing cannot be condemned as predatory unless it “falls below an appropriate measure of cost.”12 In the bundled discount cases that have reached the courts, the sellers have all priced their several products at above-average-variable-cost levels. So those prices would be deemed nonpredatory under that test. Moreover, both Brooke Group and Areeda-Turner include a likelihood that the seller will recoup its losses as an essential element in predatory pricing analysis, and that element of the analysis appears irrelevant where there are no losses to recoup. Under a profit “sacrifice” approach, recoupment of sacrificed profits could be calculated, but at the expense of complicating the analysis.13 Despite the difficulties, the basic logic of predatory pricing, the sacrifice of profits now for greater profits later once foreclosure has been accomplished, can be adapted for bundled discount situations.

chapter eight

142

In the mid-1990s these issues were faced in the US Ortho litigation.14 In that case Ortho Diagnostic Systems had complained that the bundled pricing of Abbott Laboratories was excluding it from marketing a competitive blood-screening product. The court was troubled by the exclusionary potential of bundled discounts combined with the uselessness of the AreedaTurner test in the situation in which the seller is offering bundled discounts at above-cost prices. The question . . . is whether a firm that enjoys a monopoly on one or more of a group of complementary products, but which faces competition on others, can price all of its products above average variable cost and yet still drive an equally efficient competitor out of the market. The answer to this question seems to be that it can, thereby casting into doubt reliance on the Areeda-Turner formula in package pricing cases where a party has market power over one of the components of the package. A hypothetical example illustrates the point. Assume for the sake of simplicity that the case involved the sale of two hair products, shampoo and conditioner, the latter made only by A and the former by both A and B. Assume as well that both must be used to wash one’s hair. Assume further that A’s average variable cost for conditioner is $2.50, that its average variable cost for shampoo is $1.50, and that B’s average variable cost for shampoo is $1.25. B therefore is the more efficient producer of shampoo. Finally, assume that A prices conditioner and shampoo at $5 and $3, respectively, if bought separately, but at $3 and $2.25 if bought as part of a package. Absent the package pricing, A’s price for both products is $8. B therefore must price its shampoo at or below $3 in order to compete effectively with A, given that the customer will be paying A $5 for conditioner irrespective of which shampoo supplier it chooses. With package pricing, the customer can purchase both products from A for $5.25, a price above the sum of A’s average variable cost for both products. In order for B to compete, however, it must persuade the customer to buy B’s shampoo while purchasing its conditioner from A for $5. In order to do that, B cannot charge more than $0.25 for shampoo, as the customer otherwise will find A’s package cheaper than buying conditioner from A and shampoo from B. On these assumptions, A would force B out of the shampoo market, notwithstanding that B is the more efficient producer of shampoo, without pricing either of A’s products below average variable cost.15

bundled discounts

143

The Ortho court’s analysis, which we review below, is keyed to the defendant’s possessing a monopoly over one of several products that are being marketed in a package of fixed composition based on complementarily in use. This is one kind of bundled selling, but the most common form is a discount based on total sales of a supplier’s product by value, either overall or in certain categories. Bundled Discounts under US Antitrust Law The US Antitrust Landscape before LePage’s. While the Third Circuit’s Le-

Page’s decision focused attention on antitrust treatment of bundled discounts, that decision was not the first time the US courts faced bundled discounts in an antitrust context. Antedating the Third Circuit’s LePage’s decision were two bundled discount decisions, the older of which, SmithKline Corp. v. Eli Lilly,16 was decided by the Third Circuit in 1978. Eli Lilly’s use of a bundled rebate in its marketing of pharmaceuticals was held to have constituted monopolization. In the view of the court, Lilly used its effective monopoly over several cephalosporin drugs to divert sales on a generic cephalosporin (cefazolin) from a rival to itself. Lilly did this through a bundled discount arrangement that made it economically advantageous for hospitals to opt for Lilly’s version of the generic drug rather than for the rival’s. As the court saw the matter, by making it difficult for the rival to market its version of the generic, Lilly strengthened its monopoly over cephalosporin drugs and its highly profitable cephalothin version in particular. Lilly’s patents over all the cephalosporin drugs other than cefazolin, in which it was competing with SmithKline, strengthened that monopoly. The other case, Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc.,17 was decided by the federal court for the Southern District of New York in 1996. Ortho also involved the marketing of drugs. In Ortho, Abbott Laboratories was selling test products used in screening blood supplies for the presence of viruses. Ortho, a competitor of Abbott, objected to Abbott’s use of package pricing, which had the effect of discouraging customers from buying a competing product from Ortho. As observed above, the Ortho court understood the exclusionary effects of bundled discounts as traceable to the defendant’s ability to package a monopoly product with competitive products. More precisely, the court seems to have assumed that the exclusionary effect of bundled discount pricing results from a kind of cross-subsidization: the monopolist is using its monopoly profits to subsidize an attractive package price, that is, the low package price is equivalent to offering buyers a low price on the contested product, and those low prices

chapter eight

144

for the contested product are subsidized by revenues from sales of the monopolized product. In fact, how the low price on the contested product is financed is not the issue; whether the price covers cost is the competition policy question. Using this analysis, the Ortho court was able to grant judgment to the defendant by adopting a rule that condemned bundled-discount pricing only where “(a) the monopolist has priced below its average variable cost or (b) the plaintiff is at least as efficient a producer of the competitive product as the defendant, but that the defendant’s pricing makes it unprofitable for the plaintiff to continue to produce.”18 In formulating this test, the court seems to be treating bundled discounts as a species of predatory pricing and adapting the Areeda-Turner test for predatory pricing as a tool for evaluating the lawfulness of bundled discounts. The focus of the court’s adaptation is the same as that of the Areeda-Turner test: does the defendant’s pricing threaten to exclude equally efficient producers from the market? In that case, the defendant was selling all its products above cost, so it passed clause (a) of the court’s test. Because the plaintiff remained profitable despite the defendant’s bundled pricing, the court was able to conclude that the plaintiff’s survival was not jeopardized. As a result, the plaintiff failed to make a case under clause (b) of the court’s new test. Ortho is frequently cited as an important decision in the evolution of standards for evaluating bundled discounts. LePage’s and Its Aftermath. In LePage’s the Third Circuit condemned 3M

for selling a bundle consisting of a range of office and other products to a group of large retailers, on which it offered rebates conditioned on the buyer’s aggregate purchasers meeting target (dollar volume) amounts. In so ruling, that court followed its SmithKline decision of two decades earlier. In both LePage’s and SmithKline, the court condemned the defendants’ bundled pricing as violations of the monopolization clause of Section 2. To so rule, of course, the court had to determine that the defendants possessed monopolies. In both cases, the court inferred that the defendants possessed monopolies from their commanding market shares. From one perspective, it is not clear that there were substantial entry barriers protecting transparent tape, the “Scotch” brand of which was the monopolized product. The technology for producing transparent tape is straightforward. LePage’s was able to enter the market and had attained 14.44 percent of the total transparent tape market19 before losing share to 3M’s intensified competition. From another viewpoint, however, Scotch brand tape could be seen as a separate product from the essentially generic tape LePage’s sold because the strength of 3M’s brand reduced what oth-

bundled discounts

145

erwise would have been high cross-elasticity between physically similar products. It might be thought that 3M’s alleged damage to LePage’s through a bundled discount was inadvertent. After all, if 3M’s cost of production was lower (as LePage’s economist conceded20), then it could simply have underbid LePage’s in the generic market. But an idea from Barry Nalebuff,21 emphasized by Benjamin Klein and Andres Lerner,22 suggests that lowering the price on the monopolized good, here “Scotch” tape, is more profitable for the firm controlling both a monopoly and a competitive good than lowering the price of the competitive good directly because the expansion of sales of the higher mark-up product yields higher net revenue for the firm. But in the present case there is another possible motive for bundled discounting and any associated weakening of LePage’s. The “monopoly” and “competitive” good here are substitutes,23 and any price shading of the competitive good would have a negative impact on the sales of the monopoly good. This is the same logic that underlies the “unilateral effects” concern in merger analysis.24 The Third Circuit reached its result in LePage’s without requiring the plaintiff to meet the standards set forth in Ortho. The plaintiff was not required to show that the defendant was selling its goods at below cost prices. Nor was the plaintiff required to show that it was at least as efficient as the defendant. (A fortiori, the plaintiff was not required to show that the defendant’s pricing could have excluded an equally efficient rival from the market, a superior test, later adopted by the Antitrust Monopolization Commission.25) Critics complained that the Third Circuit had created a new broad and open-ended antitrust offense without providing limits or even guidance as to the scope of that offense. Subsequent to the LePage’s decision, the AMC examined the antitrust problems arising from bundled discounts. It proposed that in evaluating the lawfulness of bundled discounts, the discount on all products of the bundle be aggregated and then applied to the single product in which the discounter competes with an independent rival.26 If the aggregate discount applied to the discounter’s sales of that single product brings the sales price below its incremental cost, then the discount should be subject to further scrutiny. In proposing to aggregate the discount, the Commission was following the approaches of all of the prior court decisions: SmithKline, Ortho, and LePage’s. But the Commission’s analysis moves beyond previous court opinions in several ways. First, LePage’s did not require that the price, adjusted by the aggregated discount, fall below the seller’s costs. Second, the Commission’s approach was an improvement over Ortho because part of the Ortho test inquired into the effect of the discount on the plaintiff firm’s

chapter eight

146

ability to compete. Such a test, however, could not be employed generally without impeding the procompetitive use of bundled discounts, since most companies lack knowledge of the cost structures of their competitors. The Commission reformulated the test to require only that the aggregated discount be applied to the defendant’s own cost structure. The Ortho court wanted to prevent a firm employing bundled discounting from driving “an equally efficient competitor out of the market,”27 but it made the mistake of adopting a test of illegality that referenced the rival’s costs rather than the discounter’s own costs. So long as the seller’s discounted prices (adjusted by aggregating its discount) exceed its own costs in producing the identical competitive product, the discounter cannot exclude an equally efficient competitor— at least if the competitor is well established in the market and has access to finance at the same cost as the discounter. The Commission did not consider an additional attribution problem: products offered by both the bundler (A) and the plaintiff (B) beyond one monopolized good and one or more competitively produced goods. The aggregated discount approach presented so far has the curious feature that the broader the line of competitive products offered by B in a bundle that competes against A’s bundle, which includes all of B’s offerings but also one or more monopolized goods, the lower the impact of the discount offered by A on B’s competitiveness. This suggests that B’s competitiveness problem can be mitigated by cooperation among the sellers of competitive products to offer a competing bundle.28 One approach to analyzing the pricing of a bundle consisting of some monopoly and some competitive products would simply assume that each competitive good should absorb its share of the dominant firm’s discount. The accumulated discount across the bundle could be prorated and assigned to the calculation of the effective price of the goods in the bundle that are produced under competitive conditions by someone— not necessarily the plaintiff.29 Otherwise, the calculated price discount for any one seller of less than the complete set of competitive products would too easily suggest foreclosure and could seriously overdeter and discourage price competition.30 One must also consider the fact that goods are not simply sold under “monopoly” or “competition,” and the monopolist’s offerings of the nominally competitive goods likely sell at higher prices than those of other sellers due to greater market acceptance (assuming similar production cost). At what price premium does the offering of a dominant firm become a “monopoly” product instead of a competitive product? The broad conceptual approaches do not tell us. A classification that may seem relatively clear in pharmaceuticals where patent protection provides a bright (if sometimes misleading) line will fail elsewhere. If the aggregated discount is to be ap-

bundled discounts

147

plied only to competitively produced goods, one must know what those goods are. This classification issue again suggests the ambiguity of “an equally efficient” competitor of differentiated goods. In particular, it implies that using the monopoly firm’s prices for nominally competitive products may give an unduly positive picture of price-cost margins of other firms as the rebate-adjusted prices are calculated. The dominant firm’s rivals likely have smaller price-cost margins prior to the reduction in price they must make to compete with the discounting firm. An additional serious problem arises in the typical case in which a firm produces more than one good: the assignment of joint costs. There are various ways of handling the issue, none of which is entirely satisfactory.31 At the end of the day, however, an “equally efficient producer” is almost never simply a source of supply that realizes sufficient scale economies in the production of one good. Economies of scope in production, distribution, promotion, and management must be carefully considered. As Oliver Williamson has argued so effectively, the economics textbook firm, often modeled as little more than a production function for a single good, seriously misleads.32 Indeed, the failure of economies of scale estimates at the plant level to capture firm scale and scope economies was an important weakness of the Harvard empirical tradition. The foregoing discussion suggests that there are really two possible elements to the disadvantage that a producer of a single (“workably”) competitive good, B, faces in bundled competition even if cost disadvantages of scale and scope in production and distribution are minimal. One is a coordination problem and the other is a market power problem. If all goods sold in a discounted bundle were competitive, then the only challenge B faces is to participate with other firms to compose a fully competing bundle. But this task necessarily fails to achieve competitive parity with another bundle when the latter bundle contains one or more goods with nonduplicable appeal to purchasers. This is market power, and the competitive advantage to those offering the nonduplicable (or less duplicable) goods in their bundle grows as those goods loom larger in the total value upon which the discount schemes are based. Moreover, because the degree of substitutability among goods within and across bundles varies widely and cannot be estimated with certainty, real cases will inevitably involve contestable assumptions about this issue. And this implies frequently great uncertainly for potential plaintiffs and for defendants. The AMC recommended three steps in bundling analysis. First, there should be an examination of whether the aggregated discount reduces prices on the competitive product below the seller’s incremental costs. Second, is the discounter likely to recoup those short-term sacrifices of profit? Third,

chapter eight

148

is the bundled discount program likely to exert an adverse effect on competition. The first two concerns mimic requirements of much predatory pricing analysis, and the third examines the impact of the bundled discount on the competitive structure of the market. Unfortunately, the AMC approach may lead to serious errors involving both false positives (type one errors) and false negatives (type two errors). Subsequent literature has demonstrated that the use of bundling for price discrimination and some other legal purposes can generate calculations of below cost pricing that have no antitrust significance. On the other hand, using above average variable cost pricing as a safe harbor could also be inadequate if bundling is part of a monopoly firm’s strategy of predation or exclusion, a problem already considered in connection with single-product discounts. Dennis Carlton, Patrick Greenlee, and Michael Waldman have suggested an alternative set of tests for vertical arrangements that can be used in conjunction with— or as an alternative to— the AMC three-pronged test where complete bundle on bundle competition is not possible.33 In addition to the possession of market power in at least one good, they suggest economies of scale, the demonstration of an increase in price for the previously “competitive” product purchased alone, and that competitive constraints have weakened either from actual or prospective exit or because rivals’ marginal costs have risen.34 The Ninth Circuit, in Cascade Health Systems,35 followed the AMC in rejecting the approach of the Third Circuit in LePage’s. It rejected the view— which it found in LePage’s— that all bundled discounts by a monopolist are anticompetitive with respect to its competitors who do not manufacture an equally diverse product line.36 The Ninth Circuit also criticized LePage’s for creating an open-ended antitrust offense whose boundaries were unclear.37 Yet that court chose to follow only the first part of the bundled discount test the AMC recommended, rejecting the second and third parts.38 The Ninth Circuit rejected a need to establish a likelihood of recoupment on the ground that a multiproduct discounter may not suffer any losses,39 since the discounted prices may all be above cost; only when the combined discounts are allocated to the single contested product is a sale deemed to be below cost. But because the below-cost sale is a legally constructed one, the discounter does not itself necessarily incur any actual loss. In such cases, there is no loss to recoup. But a distinction between profits foregone and actual losses has little meaning in economics. For example, the incumbent in a multiproduct firm engaging in simple single-product predatory pricing may experience no losses at the firm level either. The general principle is that any profit sacrificed today can only be justified if the associated strategy yields more than compensating higher profits in the future, suitably discounted,

bundled discounts

149

but, as discussed in chapter 5, this “profit sacrifice” test would be administratively problematic if it were adopted as part of an antitrust regime. That court also rejected the Commission’s recommendation that the plaintiff prove an adverse effect on competition, on the ground that such a requirement is redundant because such an effect would have to be proved under antitrust standing requirements.40 The court’s approach here is problematic. Under the court’s approach, we could similarly collapse almost all other antitrust offenses into the plaintiff’s showing of standing, and we would radically simplify the definitions of all antitrust offenses. But this technique cannot be used in an antitrust case brought by the government, because the government always has standing. Antitrust law ought not to define offenses differently depending on the identity of the plaintiff. It should embrace rules that are the same for all litigants. The Ninth Circuit’s inadequate approach to standing highlights that many of the problems of bundled discount cases derive from the failure of the courts to articulate standards for assessing the competitive consequences of such pricing. The plaintiff in LePage’s, for example, had to establish standing, and as part of that requirement should have had to prove anticompetitive effects. Yet it is not at all clear that 3M’s bundled discount pricing in that case actually produced anticompetitive effects. And without anticompetitive effects, there should have been no antitrust violation. The requirement that plaintiffs establish standing (and thus anticompetitive effects) did not produce the results that it should have in the 3M case. The Ninth Circuit’s approach— abandoning an inquiry into the anticompetitive effects of the bundled discounts as an explicit part of the offense— does little to remedy this judicial failure. Rather, the Ninth Circuit’s approach encourages courts to let the behavior’s competitive effects remain obscure, buried in the plaintiff’s task of showing standing. Indeed, despite that court’s criticism of LePage’s, the sole difference between its decisional criterion and that of LePage’s is the requirement that the defendant’s prices (reduced by the adjusted aggregated discount) fall below the defendant’s average variable cost. The impact on market structure and probable dynamics, while perhaps important in the court’s mind, is muted in its opinion. In LePage’s, 3M sometimes targeted a small number of large office supply stores and discount department stores as the recipients of its bundled rebates. Outside of these particular customers, 3M was apparently selling at its generally prevailing (undiscounted) prices. It is likely, therefore, that 3M’s rebates to these large purchasers was an effort to secure patronage though price reductions. So understood, this behavior would increase 3M’s sales, would enlarge 3M’s profits, and would benefit 3M’s customers (by reducing their supply costs), and (assuming competition among resellers) it would

150

chapter eight

also drive those customers into offering lower prices to final purchasers. So described, 3M’s behavior appears procompetitive. On this analysis, it would enhance both consumer and aggregate welfare. 3M may well have had multiple motives for offering bundled discounts. Reclaiming market share from LePage’s was certainly a benefit, but so too was greater market penetration for its other products. The goal of completely marginalizing LePage’s or driving it from the market seems unlikely. As noted, the manufacture of tape is rather straightforward. Economies of scale in tape manufacturing and distribution were never estimated as part of the case, but the principal barrier to entry into the entire transparent tape market appears to be product differentiation through brand acceptance rather than objective product characteristics. Indeed, Einer Elhauge implies as much by treating the case as one about a monopolized good (“Scotch” tape) and private label tape— without comment about whether they actually differ except in the eyes of the purchaser.41 Elhauge claims only that LePage’s was weakened in scale by 3M’s offensive in private label tape, the manufacture of which likely has low entry barriers. Overall, although the bundled rebate schemes it employed did win back market share from LePage’s, and hence had the effect of reducing that firm as a more serious rival in the future, nothing in the case suggests that 3M was likely to recoup any short-term sacrifice of profits, and there is nothing in the record to suggest that profits were sacrificed. Bundled Discounts and Rebates: The European Union The Commission has devoted a portion of its guidance on abusive practices under Article 102 to multiproduct rebates. The guidance, however, does not present its position with clarity. The critical paragraph is set out below: In theory, it would be ideal if the effect of the rebate could be assessed by examining whether the incremental revenue covers the incremental costs for each product in the dominant undertaking’s bundle. However, in practice assessing the incremental revenue is complex. Therefore, in its enforcement practice the Commission will in most situations use the incremental price as a good proxy. If the incremental price that customers pay for each of the dominant undertaking’s products in the bundle remains above the LRAIC [LAIC in our designation] of the dominant undertaking from including that product in the bundle, the Commission will normally not intervene since an equally efficient competitor with only one product should in principle be able to compete profitably against

bundled discounts

151

the bundle. Enforcement action may, however, be warranted if the incremental price is below the LRAIC, because in such a case even an equally efficient competitor may be prevented from expanding or entering.42

In the simplest interpretation of this language: Does the incremental revenue from each product cover the costs of adding that product to the bundle? If the answer is not in the affirmative, predation can be inferred because otherwise the seller is incurring needless losses. That straightforward interpretation fails, however, because the Commission is almost certainly dealing with a more complex interaction between price and cost, one in which incremental revenue is adjusted by the accrued rebates to which a purchaser is entitled if it purchases the products in question from the seller offering the bundle. Restated, the Commission is adjusting the price for the products in the same manner that it adjusts price when it evaluates single-product rebates. In the case of single-product rebates, the target is frequently set in terms of the number of units a purchaser must buy to be entitled to the discount. In a bundled rebate situation, the target almost necessarily has to be set in terms of aggregate value of purchases from the seller, either overall or within certain product categories. This allows the buyer at least some freedom to select the items and amounts of each, so long as his aggregate purchases meet the preset target, but it also allows for third-degree price discrimination in the tailoring of the discount offer. When the Commission adjusts price to reflect the contingent rebate to which the buyer will be entitled if he meets the target, that adjusted price may fall rapidly as the target is approached. This is the “suction effect” discussed in the previous chapter. This approach appears to be equivalent to the allocation of the entire rebate to the particular product in which the bundle seller is competing with a singleproduct supplier, as recommended by the US AMC in the first step in its approach to bundled rebates. The complexities of treatment of components of the bundle produced under varying conditions of competition were noted in the discussion of US policy; they pose the same challenges in the European Union. But the weaknesses of the EU position go beyond these issues. The problems with the Commission’s approach only begin with its failure to address more clearly the attribution problem. It appears to neglect the overall market significance of bundled discounting. The bundle can still result in lowered prices for purchasers. Even if rival suppliers cannot match the dominant seller’s adjusted prices to specific buyers, those rivals may still provide competition with the dominant firm in the overall market. Whether a dominant firm’s bundling strategy is likely to weaken rivals to the detri-

chapter eight

152

ment of final purchasers (or total welfare) must be explored directly. Otherwise, a seemingly “predatory” bundling strategy will resemble a simple use of selective low prices to certain buyers for market penetration purposes: so long as a sufficient level of competition remains, welfare will improve using either a consumer or total welfare standard. Leaving this issue aside, the question remains whether or not there should be a “safe harbor” at some certain level of cost. The Commission appears close to accepting a version of average total cost as exonerating, while the Bush DOJ’s suggestion of average variable cost or average avoidable coast as such a standard did not survive the change of administration. The third step of the AMC’s approach to bundled rebates focuses upon their overall market effect. The Third Circuit did not deal with market effects, and the Ninth Circuit asserted that market effects would necessarily be considered under the rubric of standing, an approach we criticized earlier, although the Commission’s guidance superficially appears to make the same mistake as the Third Circuit. Its paragraph assesses overall market impact using an approach similar to the US “substantial share” standard as articulated in Tampa. The issue should not be whether rivals can match the effective price generated by a bundled rebate to specific buyers, but whether the rebates reduce overall competition in the market. Its substantial share analysis attempts to do this. In the Intel case, however, the Commission, in challenging Intel’s rebates, did not elect to address their actual impact on the market as economists would generally analyze it. Instead, the Commission employed a specialized construction of competition, the furtherance of which does not serve consumer welfare as ordinarily understood. American and European Law on Exclusivity and Rebates: A Comparison Under US antitrust law, exclusive-supply arrangements are assessed under the Rule of Reason, and the party challenging the lawfulness of such an arrangement bears the burden of proving them unreasonable, which means impairing consumer (or possibly aggregate) welfare. The case law suggests that exclusive-supply arrangements involving less than 40 percent of the market are likely to be upheld because these arrangements carry the potential of reducing a variety of distribution costs. Exclusive-supply arrangements involving higher percentages of the market, while not per se unlawful, are more suspect because their efficiencies are increasingly likely to be offset by negative market power effects as the distribution opportunities of rival suppliers are squeezed and they incur concomitantly higher distribution costs. At very high market-share percentages, Section 2’s monopoliza-

bundled discounts

153

tion or attempted monopolization clauses may be triggered, barring such exclusive contracts. Under EU competition law, exclusive-supply arrangements are evaluated under both Articles 101 and 102. Exclusive-supply agreements are virtually outlawed by Article 102, which presumes that exclusive-supply arrangements are abuses of dominant position. This presumption can be overcome, however, if the supplier can justify the exclusive arrangements on efficiency grounds. It can do this by showing that the exclusive arrangement meets the conditions of Article 101(3), which is effectively incorporated into the case law construing Article 102. Since these constraints apply only to dominant firms and, as a practical matter, a market share of 40 percent or more is requisite for dominance, the EU law under Article 102 appears to generate results similar to the US law under Section 1 of the Sherman Act and Section 3 of the Clayton Act.43 But Article 102 is not the only provision of EU competition law that governs exclusive-supply contracts. As observed above, exclusive-supply contracts also fall within the scope of Article 101. The Commission has issued a bloc exemption for such contracts, but that exemption applies only where both seller and buyer occupy less than 30 percent of any relevant market affected by their agreement. The reader will observe that this 30 percent share standard is substantially less than the 40– 50 percent share that would make exclusive-supply agreements vulnerable to challenge under the US antitrust laws. With the exception of the FTC’s recent proceeding against Intel and the Ninth Circuit’s Masimo decision, pure single-product rebates have not raised significant antitrust issues in the United States. As the Commission’s decision in its own Intel case and the decisions of the General Court and the Court of Justice in Prokent-Tomra illustrate, however, single-product rebates are vulnerable to antitrust attack in the European Union. US law on bundled rebates has moved toward an “as efficient competitor” test as a subpart of an analytical framework designed to apply predatory pricing standards to the evaluation of bundled discounts. This approach, first employed in the Ortho litigation, was refined by the AMC. There it was used in conjunction with a discount attribution rule that imputes the discount on the bundle sold to a customer to the seller’s price on the particular competitive product. According to the AMC, if the attributed discount brings the price below the seller’s “incremental cost”44 for the product sold, then the first step of the AMC framework is satisfied. In that case an equally efficient competitor would be unable to compete in sales of the contested product to the customer without selling below its (incremental) cost. The second step asks if the apparently sacrificed profits are likely to be recouped as a result of

chapter eight

154

damage to rival firms whose prospects are dimmed by the bundled discount. We have previously noted the Ninth Circuit’s rejection of the AMC’s second step on the grounds that when bundled discounts do not generate actual losses, there is no need for recoupment. The economist’s concept of an opportunity cost incurred when a firm sacrifices a profit opportunity has not found favor in US antitrust law because of the perceived vulnerability of that concept to type one errors, a matter that we discuss below. But we have also suggested, following Baumol, that opportunity costs incurred only by the predator are relevant only in a determination of the predator’s “sacrifice,” and are irrelevant on the issue of whether an equally efficient rival can be driven from the market. The third step in the AMC analysis examines the market impact of the bundled discount: whether the market structure has become less competitive. The purpose of the exercise, regardless of precisely how it is performed, is the same as that of predatory pricing. Operationally, does the behavior of the dominant firm portend higher future prices? The US courts that have so far found the use of bundled discounts to constitute antitrust violations have focused upon their use by a monopolist to prevent rivals from effectively competing with it in sales to the buyers to whom the discounts were offered. The rationale, however, is unclear. Monopolization involves the use of monopoly power to attain or to maintain a monopoly. In what way does a firm offering a bundle use its monopoly power? Perhaps the firm’s power over the monopoly product is being used as a means for enticing the buyers to take the bundle. How could the firm’s power be used in this way? Perhaps it is by selling the monopoly product in the bundle at a price less than the monopoly price.45 (Note that if the seller is offering the monopoly product at an effectively discounted price in the bundle and no prices rise elsewhere, then this is welfare-enhancing behavior, whether measured from a perspective of aggregate welfare or consumer welfare.) In the LePage’s case, the court treated branded tape as a unique monopoly product and found a violation of Section 2 of the Sherman Act when sold in a bundle and discounted along with all other products in the bundle. Whereas in SmithKline (in the Court’s understanding) the defendant used the power of the monopolized product to coerce buyers to purchase a competitive “contested” product, in LePage’s, how the court understood 3M to have wrongfully used the monopolized product is difficult to understand. Because the case involved discounting all the products in a bundle that included the monopoly product, and there is no indication that the monopoly price was in any way inflated for the bundle, the court’s theory of monopolization is unclear. 3M’s customers certainly wanted the substantial discount the package purchases generated. It is not clear from the court’s opinion how great a role the monopoly product played in the customers’ decisions

bundled discounts

155

to deal exclusively with 3M on all the products in the bundle. To what extent would the customers have purchased from 3M exclusively if the bundle had not included Scotch tape? If the proportion of the bundle composed of Scotch tape were relatively small, it would seem contrived to rest a monopolization determination on the presence of that product in the bundle. In such a case, the plaintiff should be challenged to establish a connection between the defendant’s employment of the monopolized product in bundled pricing and its own injury and between that practice and a reduction of competition in the general market. The court sidestepped the issue of the role of the monopoly product in generating injury to the plaintiff as well as generating effects on the general market. Rather, the court ruled that bundled pricing was impermissible when the seller possessed a monopoly and when the bundled pricing impeded a rival’s sales to those buyers to whom the bundled prices were offered. The court’s theory of liability thus lies not in a misuse of monopoly power; rather, liability arises under a rationale of monopoly maintenance, broadly construed. Under this rationale, firms possessing large market shares are discouraged or forbidden from employing bundled pricing. Normally, the antitrust laws encourage a monopolist to compete, and to compete aggressively. The really interesting question that case raised, accordingly, is the extent to which a firm possessing a large market share should be free to employ bundled discounts as a marketing tool. Both the European Commission and the AMC employ an “as efficient competitor” test to identify the market that is open to an entrant selling at a price that is not below its costs. Both the European Commission and the AMC adjust price to reflect the discount available to the buyer. Under that test the discounting seller needs only to ensure that its (adjusted) prices are not below its own costs. Although the “as efficient competitor” test is drawn from predatory pricing analysis and uses the terminology of predatory pricing, the European Commission uses the test in a slightly different way than the AMC has used it. The EU framework initially focuses on foreclosure of a representative buyer and addresses market impact in the second step of its analysis. As we have pointed out above, the strength of the EU analysis depends upon its ability to assess overall market effects. The US analysis asks whether the adjusted price on the customer’s incremental purchases are below cost as a step into an examination of the impact on market structure. The reader will recall that the EU analysis of what is nominally singleproduct discounting is tied to the so-called suction effect, the loyalty impact a conditional rebate generates. The other operative concepts (the required share and the commercially viable share) in the EU framework draw their significance from their relation to the suction effect. As a customer’s

chapter eight

156

purchases from the dominant seller increase, the customer’s incentives to continue obtaining its supplies from that seller grow stronger. At the point when the effective price (that is, (list price − accumulated discount)/future purchases) falls to the level of its cost, a rival is effectively barred from competing for the customer’s patronage. For a rival to be able to compete for that patronage, the customer must be open to rival sources for a larger part of its demand than remains when the effective price to a customer falls to cost. The customer’s commercially viable share must be larger than its required share. Because the entire EU analysis just described is keyed to the suction effect, it is vulnerable to the deficiencies of the suction-effect concept. The suction effect arises in two dimensions. The first dimension is the suction effect in the current sales period. That suction effect disappears when the target purchases have been made. At the beginning of the next sales period, rivals are free to offer their own competitive sales terms, prices, and rebates. The second dimension is the long-term incorporation of the suction effect into the ongoing relation between the dominant supplier and the customer under consideration. Insofar as the customer is seen as compelled to take a major part of its supplies from the dominant seller, then only the remaining part of his purchases (the residual after the compelled amount) will be open to the competition of rival suppliers. If such compulsion is posited, then the suction effect is incorporated in the buyer’s continuing relation to the dominant supplier, and it will be appropriate to follow the Commission’s framework comparing the required share with the commercially viable share. But this is the nub of the issue. Why is the customer compelled to purchase from the dominant seller? The Commission gives us two possible reasons. One is that the customer prefers the dominant seller’s product, quite probably because its own customers prefer it. The other reason is that rivals are capacity constrained, but this could not be a long-run argument because capacity can typically be expanded.46 Moreover, as pointed out in the discussion above, while temporary capacity constraints may limit a rival’s ability to supply the entire market in the short run, they are far less likely to limit a rival’s ability to supply a particular customer who is open to the rival’s product. Thus the application of the Commission’s analysis beyond the current period depends on the brand preference of the customer. The customer must prefer the dominant supplier’s brand so much that its residual (noncompelled) demand is less than the required share. The customer’s brand preference itself “forecloses” rival suppliers from a part of the market; the rebate extends that foreclosure into an area of sales that would otherwise be “contestable.” The Commission’s compelled share approach is a noteworthy innova-

bundled discounts

157

tion that attempts to bring the antitrust analysis of single-product rebates into the framework of bundled rebates. What will be the response of American courts when the “compelled share” analysis is brought before them? American courts are likely to take such a proffered analysis seriously, but they are also likely to be more skeptical than the European Commission that customer brand preferences can be impervious to sustained competitive assault. The Goals of Competition Policy Once Again While the United States now maintains an almost exclusive focus on either consumer welfare or total welfare,47 the European Union has historically taken a quite different view of the appropriate goal of competition policy by pursuing what we have called a rivalry standard. This standard is also very familiar in the United States. In the heyday of the Harvard School and earlier, it was believed that little was sacrificed by keeping markets deconcentrated. The demise of the rivalry standard in America resulted from a growing belief that the price paid in economic efficiency had been grossly underestimated. Moreover, many US adherents sought the rivalry standard as an imperfect means for the achievement of some measure of economic welfare all along, rather than as a way of achieving other social or political objectives. In Europe a version of the rivalry standard remains well embedded and coexists in ongoing tension with the welfare standards. As a recent account of this difference in terms of bundling policy makes clear, the divergent perspectives can be traced to ordoliberalism and the view of market power as an aberration that sharply divides dominant firms from other market participants. As we have already argued, however, this approach is facially suspect because all firms that sell differentiated products have some degree of power over price and often some supernormal profits as well. Thus dominance lacks credibility as a distinct state. Its closest US analogue, “monopoly power,” is now treated as an important but somewhat arbitrary idea for purposes of directing the necessarily limited resources of public policy to situations in which damage to consumer or total welfare is most likely. This view appears to be gaining strength in Europe, but there is still a powerful body of opinion of the traditional kind. The older European position would reject both the application of an equally efficient competitor test and the broader— and, in our view, more appropriate— consideration of the situation of the overall market. It would instead favor some version of the residual form-based concern with inappropriate practices. This is sometimes called a “consumer sovereignty” standard.

158

chapter eight

This consumer sovereignty standard is based on the view that dominant firms engage in “impediment competition” that is not “normal” and that other firms should be protected from such competition.48 As applied to bundling: the dominant company will not respond to preferences on the parameters of the bundle  . . . but will attempt to reduce consumer sovereignty by coercing or inducing the intermediaries to reject competing bundles . . . therefore denying the final consumers the opportunity to influence the market selection process according to their own preferences.49

Viewed most sympathetically, this position might be construed as a kind of path dependent limitation of consumer choice. And there is no doubt that cumulative market effects do often limit final choices— usually on the basis of quality-corrected price trumping variety. In our view, however, the failure of the consumer sovereignty approach to present an operational means of weighting price, quality, and variety that can substitute for the forces of competition (as ordinarily understood) is fatal as a useful policy prescription.50 And even those sympathetic to a high level of market participation understand that in most of the economy there is a conflict between efficient production or distribution and maximizing the number of sellers and their market access. Predictably, the problem has been considered most carefully in Germany. David Gerber recounts the legislative struggle to distinguish “performance competition” from “abuse” and the attempts to establish balancing principles. Extensive consideration by academics and legislators failed to establish persuasive distinguishing features of behavior that were economically justified from those that were not.51 The failure to make the consumer sovereignty standard coherently operational in conventional economic terms understandably drives those committed to it toward form-based policy. How Will US and EU Law Develop in Response to the Problem of Bundled Discounts? Because of the potential procompetitive and generally welfare-improving character of bundled pricing, American courts and legal commentators generally recognize that most bundled pricing should be treated as lawful. The disagreements arise over where the dividing line between lawful and unlawful should be drawn. Should the antitrust laws ever prohibit a profitmaximizing seller from engaging in bundled pricing where no product fails

bundled discounts

159

to cover its cost and is welfare-enhancing in the short run? If so, in what circumstances should the prohibition come? Phillip Areeda and Herbert Hovenkamp, for example, would make legality turn on whether the defendant can establish “an adequate business justification.”52 Posner has not offered an analysis specifically keyed to bundled discounts, although his broad approach to exclusionary practice is cast in terms of allocating the burden of proof focused on the question of excluding an equally efficient competitor.53 When courts are uncertain about the consequences of a defendant’s behavior, they are prone to treat it as prima facie “exclusionary” and to require the defendant to justify its behavior. That is what the US Supreme Court did in the Aspen Skiing case,54 for example, and that technique follows the path of burden allocation most antitrust litigation employs. The plaintiff must establish the anticompetitive character of the defendant’s behavior. Then the burden of going forward falls on the defendant to establish a business justification for the behavior. Then the plaintiff bears the burden of disproving that justification. Problems arise, however, when the court is uncertain how to evaluate the behavior in question. Courts tend to treat behavior they do not understand as prima facie anticompetitive and to find the defendant’s justification as inadequate. Often defendants are unable to articulate a business justification the court is able to understand. Thus the initial categorization of the behavior in question as prima facie anticompetitive or exclusionary is likely to determine the ultimate result. In recent years, the Supreme Court has become increasingly sensitive to type one errors (or false positives) by the courts. Type one errors occur when a court mistakenly evaluates behavior as anticompetitive. Judge Frank Easterbrook has argued55 that type one errors are likely to produce more social harm than type two errors (false negatives), because type one errors are maintained through the judicial system itself, whereas the market may well correct type two errors. In light of the judicial practice requiring defendants to justify behavior the courts do not understand, there is a substantial risk of judicial error in evaluating bundled discounts. LePage’s/3M provides an example. Both the majority and the dissent understood the potential for 3M’s bundled discounts to divert customers from LePage’s to 3M. They understood the exclusionary potential of the aggregated discount in the contested tape market. Yet they evaluated the antitrust legality differently. Accordingly, because of the potential of bundled discounts to increase welfare and the difficulties of the courts in evaluating this practice, we suggest that bundled discounts be presumed lawful. We would require the plaintiff in a bundled discount case to prove that the defendant possesses a monopoly, that the bundled pricing is not profit-maximizing absent suc-

chapter eight

160

cessful foreclosure, that the aggregated discount in the bundle when prorated across all competitive products in the bundle (not necessarily sold by the plaintiff) causes price of the contested product to fall below the seller’s LAIC, and that the effective bundled price level is not likely to be sustained after the exit of the plaintiff’s product from the market. We can foresee a general US acceptance of something close to the standard just outlined. We hope, but do not confidently predict, that the EU Commission and the courts will abandon what appears to be a bias toward rivalry— based on some amalgam of a concern about fairness to weaker producers and the desire to maximize the variety of offerings to consumers— and more fully embrace a version of the efficient competitor test enunciated in the discussion paper and guidance. This means, for example, that a case such as Intel would need to be far more carefully geared to demonstrating, not that AMD did not provide substantial sales of chips to all major manufacturers at all times, but that Intel’s behavior was threatening to foreclose the overall market.

9

Intellectual Property, the Two Microsoft Decisions, and Antitrust in Dynamic Industries Modern economies have created various forms of intellectual property in order to remedy the deficiencies of the traditional property regime, which provided insufficient stimulus to invention, artistic creativity, and to fostering product reputations. These particular deficiencies are addressed by three of the most common forms of intellectual property: patents, copyrights, and trademarks. In addition, intellectual property can be protected as a trade secret. While trademarks and trade secrets play only minor roles in the interface between intellectual property law and competition law and policy, the other two means of protection loom very large. Patent law protects levels of inventiveness beyond the capabilities of ordinary professionals in a given field. In technical terms, anything less would be “obvious” and thus would fail the requirement of nonobviousness in section 103 of the US Patent Act.1 Europe uses a similar standard.2 There is a limited subset of patentable inventions known as “pioneer” inventions,3 which, in view of their technology-changing aspects, are accorded an especially wide scope of protection by US courts. The term of patents is now consistent between the United States and the European Union (and the rest of the world) at twenty years. Copyright laws in both the United States and the European Union now extend to the end of the creator’s life plus seventy years, but they offer generally weak protection

chapter nine

162

against rival products and are designed for a market of competition among differentiated products. This is evident from copyright’s traditional function of protecting literary and artistic expression where the protected products are generally exposed to myriad competing expressions. The works themselves (that is, the author’s expressions) are protected against copying, but not the ideas underlying them. As a result, rivals are generally free to compete against copyrighted works by offering their own versions.4 Trade secret law generally provides relatively weak intellectual property protection. It protects only against unlawful appropriation of confidential information such as theft or in breach of contract.5 The common thread across all forms of intellectual property is a measure of exclusivity. Indeed, it is exclusivity that defines the scope of the intellectual property. In this chapter, we first address the extent to which antitrust law is used to override intellectual property rights in the United States and the European Union by imposing duties on rights holders to share their intellectual property with their competitors and others, or conversely, the extent to which antitrust law respects the exclusive rights characterizing intellectual property. The chapter then considers whether so-called dynamic industries, those based on intellectual property, need special competition policy rules. This task of investigating legal demands on intellectual property holders is complicated by the fact that in both the United States and the European Union, antitrust law sometimes imposes a duty to deal outside of the intellectual-property context. Both jurisdictions have recognized an “essential facilities” doctrine under which a property owner has been required to make its property available for use by rivals. As we discuss below, the essential facilities doctrine is in a state of retrenchment in the United States, while seemingly expanding in the European Union. Indeed, the locus of EU expansion is intellectual property. But a duty to deal has also been imposed by courts and other authorities in both the United States and the European Union that have either ignored the essential facilities doctrine or have explicitly rejected it as a rationale. Antitrust Impositions of Duties to Deal or Otherwise to Share Property Rights Outside of the Intellectual Property Context The US Experience. The two jurisdictions take different approaches to the duty of a business firm to deal with others. The US position is that business firms generally possess absolute discretion over the choice of their trading partners. The classic articulation of this position is found in United States v. Colgate 6 of 1919 in which a producer let its customers know that if they

intellectual property and antitrust in dynamic industries

163

resold its products below the price suggested by the producer, the producer would not resupply them: In the absence of any purpose to create or maintain monopoly, the act does not restrict the long recognized right of a trader or manufacturer engaged in any entirely private business, freely to exercise his own independent discretion as to the parties with whom he will deal; and, of course, he may announce in advance the circumstances under which he will refuse to deal.

At the time, resale price maintenance agreements were per se illegal, and by the advance announcement, the manufacturer was securing the effects of a prohibited resale price agreement. Yet the Court found no objection to the manufacturer’s skirting the prohibition on resale price agreements in this way. These circumstances underscore the strong US policy behind the right of a business firm to refuse to deal. As the quoted language from Colgate acknowledges, however, there are exceptions to the right to refuse to deal. These involve “a purpose to create or maintain a monopoly.” Indeed, one of those exceptions is illustrated by the decision in United States v. Terminal RR,7 decided in 1912, just seven years before Colgate. The Terminal RR case concerned the terminal facilities on the Mississippi River at Saint Louis where the eastern and western railroads met. There were twenty-four railroads whose lines terminated at Saint Louis. But a consortium of fourteen railroads acquired the Terminal Railroad Co., which owned the two railroad bridges and the only railroad ferry, and thus the means by which the eastern and western railroads could interconnect. The US Supreme Court treated the arrangement as an unlawful monopoly possessed by the fourteen owners of the terminal railroad company in violation of Section 2 of the Sherman Act and the underlying contract as an unreasonable restraint in violation of Section 1. As a remedy for these violations, the Court required the parties to submit a plan for the reorganization of the terminal railroad company providing for the admission of any railroad wishing to join and for “just and reasonable terms” of usage by any railroad not electing to join. The remedy compelled the fourteen defendant owners of the terminal railroad company to deal with the remaining railroads. In the context of the actual market restraint, the Court’s remedy made sense. It was impractical for each railroad to own or build its own bridge, and indeed it would be socially wasteful for each of the several to build or to acquire one when the existing bridges possessed the capacity to carry all of the railroad traffic. In this situation, joint ownership reduced the costs of each of the rival railroads

chapter nine

164

and maximized welfare. The Terminal RR case is widely understood to have been the origin of the essential facilities doctrine, although the Court did not employ that phrase in its opinion. Later courts have articulated the standards for applying the essential facilities doctrine in a number of ways, all of which relate in some way to the original bridge scenario. The doctrine is premised on the defendant or defendants possessing a monopoly over the facilities, the facilities being essential for rival(s) to compete, the capacity of the facilities being sufficient to accommodate the needs of the plaintiff and other claimants, and the practical inability of the plaintiff and other claimants to duplicate the facilities.8 In the original situation, the capacity of the bridges to sustain other traffic seemed to make the joint use of the facilities an efficient use of resources, while the exclusion of the rivals emphasized the wasteful and welfarereducing aspect of the initial ownership agreement. Ownership of the facilities by less than all of the potential users, all of whom were competitors, emphasized the exclusionary character of the ownership agreement. Yet judicial articulations of the doctrine have not always incorporated analogues to all those aspects of the St. Louis Terminal agreement. While requiring that the owner(s) of the facilities be able to accommodate the usage the plaintiffs demanded, they do not always emphasize the decreasing-averagecost aspect of increased usage. Moreover, judicial articulations of the doctrine properly do not require that the ownership be shared among a group of conspirators; a single owner is sufficient. Although earlier articulations had not always insisted on the existence of a competitive relationship between the owner of the facilities and those seeking access, recent articulations affirm the importance of that competitive relationship.9 There have been a number of situations in which business firms have been required to deal with rivals beyond the essential facilities doctrine. The most well known of these is Aspen Skiing Co. v. Aspen Highlands Skiing Corp.,10 where the Court affirmed an order requiring the Aspen Skiing Co. to enter into joint marketing of a four-mountain ski pass with its rival. It that case, the Court treated Aspen Skiing, the owner and operator of skiing facilities on three mountains in Aspen, Colorado, as a monopolist 11 and treated its refusal to engage in joint marketing with Highlands, the owner and operator of skiing facilities on the one other mountain in Aspen as a “predatory” act. The rationale for this label lay in Aspen’s earlier participation in the joint marketing of all mountain ski passes, when three independent companies (including Aspen) operated three competing ski facilities, and its later refusal to engage in joint marketing when it had acquired three

intellectual property and antitrust in dynamic industries

165

of the four mountains. The Court referred to this as “a decision by a monopolist to make an important change in the character of the market,” that is, a decision to depart from a practice that originated in a more competitive market to a practice that only benefited the dominant firm. In the Court’s view, the predatory nature of Aspen’s behavior was confirmed when Aspen refused to sell daily lift tickets in bulk to Highlands for resale to the latter’s customers and when Aspen failed to offer any efficiency justification for this conduct. In its decision, the Supreme Court declined to pass on the applicability of the essential facilities doctrine.12 The rights of business firms to select their customers were similarly limited in Lorain Journal Co. v. United States (an authority relied on by the Aspen Court)13 and in Otter Tail Power Co. v. United States.14 In the former case, the only daily newspaper in the relevant area was not permitted to refuse advertising from those who patronized a competing radio station, and in the latter case, a power company that controlled all of the high voltage lines in the affected area was not permitted to refuse to carry electricity to a municipal system that wanted to replace the power company in the retail distribution of electricity. In both cases the Court held that the defendants’ refusals to deal violated the attempt-to-monopolize clause of the Sherman Act’s Section 2. More recently, in Verizon Communications, Inc. v. Law Offices of Curtis V. Trinko,15 the Court limited the scope of Aspen Skiing as a precedent. In Verizon Communications, the plaintiff law firm charged that Verizon, the incumbent local telecommunications carrier, had violated Section 2 of the Sherman Act by failing to provide access to operations support systems to competing local exchange carriers as it was obligated to do under the Telecommunications Act of 1996. In its decision, Justice Antonin Scalia, writing for the Court, ruled that the antitrust laws imposed no duty on Verizon to deal with a rival. He distinguished Aspen Skiing as involving “the unilateral termination of a voluntary (and thus presumably profitable) course of dealing,” suggesting “a willingness to forsake short-term profits to achieve an anticompetitive end.” Scalia further stated that Aspen Skiing “is at or near the outer boundary of § 2 liability.” Adverting to the essential facilities doctrine, he stated, “We have never recognized such a doctrine . . . we find no need either to recognize it or to repudiate it here.” In summary, US antitrust law recognizes a broad right of a business firm to choose those with whom it wishes to deal or not to deal, with narrow and shrinking exceptions. Aspen Skiing imposed a duty on a monopolist ski resort to cooperate with its rivals where it failed to justify a departure from a pattern of presumably profitable behavior that originated in competitive

chapter nine

166

circumstances, but that exception appears to be confined to those facts. The Supreme Court has denied that it has ever recognized the essential facilities doctrine and has declined to approve or to repudiate it. The EU Experience. In Europe, the Court of Justice has sometimes treated

a refusal to deal by a dominant firm as an abuse of dominant position. In United Brands v. Commission of the European Communities,16 when a large banana supplier refused to continue supplying a distributor that had made overtures to a competing supplier, the Court ruled that the supplier had abused its dominant position. In its Istituto Chemioterapico Italiano SpA and Commercial Solvents Corporation v. Commission decision,17 the Court took the same approach toward Commercial Solvents, a producer of aminobutanol, a raw material used in the production of the antituberculosis drug ethambutol. Commercial Solvents had been selling aminobutanol to Zoja, the principal Italian producer of ethambutol. In the early 1970s, however, Commercial Solvents decided to produce ethambutol directly and no longer to supply aminobutanol to others. The Court ruled that Commercial Solvents held a dominant position in the market for aminobutanol, a position it abused when it decided to reserve its entire production for its own needs. In both United Brands and Commercial Solvents, the dominant firm had previously been dealing with a downstream firm, and the abuse consisted of terminating that previous supply arrangement. Drawing from these cases and the intellectual-property cases described below, the Commission has recently articulated an “essential facilities” doctrine under which a dominant firm may be required to supply downstream firms that depend upon the dominant firm for a needed input into their production processes. Although neither the General Court nor the Court of Justice have explicitly recognized this doctrine, the advocate general has employed it in his presentations to the Court. As the decisions of the Court of Justice attest, there appears to be an essential facilities doctrine in fact, and one much broader than the US antitrust doctrine of the same name. The Antitrust/Intellectual Property Interface: Tying, Exclusivity, and Duties to Share The Court of Justice, in its 1988 Volvo decision, considered the interplay between intellectual property and competition law.18 A critical issue in that case was whether Volvo, the auto manufacturer, held a “dominant position” because it owned rights under United Kingdom design-protection law over certain body panels. The ECJ declined to answer this basic question directly.19 Instead, the ECJ upheld Volvo’s right to deny permission to a third

intellectual property and antitrust in dynamic industries

167

party to import panels covered by its design right on the ground that its refusal to license would not constitute an “abuse” regardless of its holding of a dominant position. To force such a rights owner to grant a license, even in return for a reasonable royalty, would deprive the owner of “the substance of his exclusive right.”20 Accordingly, the refusal to grant such an exclusive license could not “in itself” constitute an abuse of a dominant position.21 In so ruling, however, the ECJ also indicated that the exercise of such an exclusive right could constitute abuse if it involved a range of behaviors, including “the arbitrary refusal to supply spare parts to independent repairers, the fixing of prices for spare parts at an unfair level or a decision no longer to produce spare parts for a particular model even though many cars of that model are still in circulation.” Thus although a dominant firm’s refusal to license intellectual property rights does not “in itself” constitute abuse, its “exercise” of those rights can constitute abuse, and the ECJ provided a nonexhaustive list of ways in which a refusal to license could constitute an abuse. In American antitrust parlance, Volvo was using intellectual property to control the aftermarket. The use of intellectual property (almost always patent or copyright) to control a market for an associated (complementary) market has a long and complex history under US law. The legal issues involve tying and monopolization under the US antitrust laws and patent and copyright misuse under the patent and copyright laws. The focus of the EU cases has been on the extent to which competition law imposes a duty on intellectual-property rights holders to share their exclusive rights with competitors. The Antitrust/ Intellectual Property Interface in the United States: A Focus on Tying and a Gradual Absorption of Economic Concepts. The interaction of

US antitrust law and related policies with intellectual property has multiple dimensions and has undergone a complex evolution. Most of these interactions involve tying behavior, which can raise issues under the antitrust laws governing tying arrangements (Sherman Act Section 1, Clayton Act Section 3), monopolization (Sherman Act Section 2), and the patent misuse doctrine. In Henry v. A. B. Dick Co. of 191222 the Court upheld the right of the patentee of a mimeograph machine to authorize its use only with stencil paper, ink, and other supplies the patentee made. Two years later Congress enacted the Clayton Act, whose Section 3, forbids tying arrangements whose effects “may be to substantially lessen competition or tend to create a monopoly.” In 1917 Motion Picture Patents Co. v. Universal Film Mfg. Co.23 overruled the A. B. Dick Co. decision. In Motion Picture Patents, the owner of a

chapter nine

168

patent on a mechanism for feeding film into motion picture projectors conditioned the use of the patented device with film produced or approved by the patentee. The Court, relying on the “policy” it saw as embodied in the recently enacted Clayton Act, incorporated that policy into its application of patent law, thus creating what came to be known as the “patent misuse” doctrine. In later cases the Court gave further substance to that doctrine, ruling that the use of a patent on a tying product constituted misuse, making the patent unenforceable as a matter of patent law.24 The apogee of the misuse doctrine occurred in 1944 in the Mercoid cases.25 In these cases, the Court was dealing with patents on heating systems that consisted of a combination of several unpatented component parts— the patent was on the system as a whole. The two patented heating systems were sold with two specially designed components (a combination stoker switch and a combination furnace control) that had no other use than in the patented heating systems. Under then existing patent law, the sale by a third party of such devices would constitute contributory infringement. Moreover, because the tied products could only be used in the heating system, it was clear that they were economic complements and were used in a fixed ratio to the heating system. The patentee thus has no ability to raise the price of the tied product.26 Nevertheless, the Court determined that the patentee was using the patent as a tying product to sell the specially made (but unpatented) devices as tied products. This behavior constituted patent misuse, rendering the patents unenforceable. Congress disapproved of the Mercoid decisions in its 1952 revision of the Patent Act. Section 271 of that legislation limits the scope of the patent misuse doctrine, specifically allowing a patentee to sell specially made components and defining equivalent behavior by third parties as contributory infringement, thus overruling the Mercoid decisions as precedents. This part of Section 271 gives implicit recognition to the single monopoly profit theorem. Patent misuse involving tying was limited to tied products that were staple articles of commerce.27 Under a 1988 amendment to Section 271, tying involving staple products can constitute misuse only if the patentee possesses market power in the market for the tying product. Although Congress was limiting the power of the courts to find misuse in the mid-twentieth century, the courts, during the same period, were expanding their power to invalidate tying arrangements involving both patents and copyrights by presuming that those forms of intellectual property conferred market power.28 Under the governing law, tying imposed by a firm possessing market power in the market for the tying product was per se illegal. In 2006 the Court brought the patent misuse doctrine and antitrust law into

intellectual property and antitrust in dynamic industries

169

harmony by drawing on the 1988 amendment to Section 271 to erase the presumption of market power that antitrust law had attached to patents.29 The rule making tying per se illegal when entered into by a firm possessing market power over the tying product has generated problems in aftermarkets. The most prominent decision is Eastman Kodak Co. v. Image Technical Services, Inc. of 1992.30 Kodak initially raised no objections to independent servicing organizations performing maintenance and repair work on its high-speed copiers and micrographic equipment. Later, however, Kodak sought to service the machines exclusively and denied parts to independent servicing organizations. In an antitrust action brought by those organizations, the Court ruled that because Kodak controlled 100 percent of the parts compatible with its machines, the plaintiffs had established a prima facie case of unlawful tying and of monopolization. Kodak had contended that because its equipment occupied only about a 20 percent share of the equipment market, it lacked power in the markets for parts and servicing, as price increases in these aftermarkets would be offset by consequent price reductions compelled by competition in the equipment market. The Court, however, ruled that Kodak had the burden of proof on this issue and remanded the case for trial of Kodak’s defense. Although some antitrust observers first thought that Kodak would severely limit the power of producers of physically differentiated products to control the aftermarkets, the ruling was based on Kodak’s failure to prove that customers generally knew from the beginning about its policy of refusing to sell parts to independent servicing organizations.31 If it had provided such proof, then it would have established that competition in the equipment market constrained its power in the parts and servicing markets. On remand in the Kodak case, Kodak sought to defend its refusal to make parts available to independent servicing organizations on the ground that some of the parts were patented. The Ninth Circuit ruled that a patentee presumptively is entitled to the exclusive rights conferred by patent law but can lose these rights if it is asserting them pretextually. The court then affirmed the judgment against Kodak on the grounds that the jury must have determined that Kodak asserted its patent right pretextually when it decided as it did. Nevertheless, in the federal circuit Xerox won a decisive victory in a 2000 case involving similar issues. Xerox controlled its copier machines’ aftermarkets by refusing to sell patented replacement parts to independent servicing organizations and by refusing to make its copyrighted manuals available to them. The plaintiffs contended these behaviors constituted monopolization, but the federal circuit ruled that Xerox was lawfully exercising the exclusive rights it possessed under the patent and copyright laws.32

chapter nine

170

The reader will note the similarities between the Xerox case of 2000 and the A. B. Dick case of 1912, despite the complex array of legal developments that took place in the intervening years. In both cases, the patentee was able to use its patent to control the aftermarket. In the years separating those cases, the courts initially expanded the misuse doctrine, but Congress halted this and made clear that a firm in Xerox’s position would not be vulnerable to a charge of patent misuse for refusing to sell patented goods. And the Court’s incorporation of the misuse standards into antitrust law in its Independent Ink decision of 2006 33 shows that such a firm would not be violating the antitrust laws. The leading edge of the US antitrust/intellectual-property interface is found in the facts giving rise to the US Microsoft case of 2001.34 Although computer software in both the United States and Europe has traditionally been protected primarily by copyright law, this usually modest barrier to follow on competition is reinforced by trade secrets and the great advantage of network effects. Network effects provide a growing advantage to the market leader, and trade secret protection guards the application programming interfaces (APIs) on which the network effects depend. The overall result is essentially patent-like in its protective impact for Windows. In that case, Microsoft had contended that copyright law protected the restrictions it imposed on original equipment manufacturers from liability under Sherman Act Section 2. The restrictions in issue forbade: “(1) removing any desktop icons, folders, or ‘Start’ menu entries; (2) altering the initial boot sequence; and (3) otherwise altering the appearance of the Windows desktop.” The DC Circuit viewed these restrictions as prima facie anticompetitive because they impeded the manufacturers from distributing non-Microsoft browsers and thus impeded Netscape from reaching the critical mass of users potentially necessary for it to generate platform competition for the Windows operating system. In those instances where the court believed the restriction protected a substantial copyright interest 35 of Microsoft, however, it approved of the restriction. Thus it agreed that a replacement of the Windows desktop with a user interface designed by the manufacturer or Netscape would be “a drastic alteration” of the copyrighted desktop that “outweighs the marginal anticompetitive effect of prohibiting the OEMs [original equipment manufacturers] from substituting a different interface automatically upon completion of the initial boot process.” The court also agreed in principle that it would be lawful for Microsoft to bar alterations to Windows that would undermine its “value . . . as a stable and consistent platform,” but ruled that Microsoft had not shown that the stability and consistency of its platform were threatened.

intellectual property and antitrust in dynamic industries

171

The court also ruled that it was lawful for Microsoft to integrate the browser into the operating system so long as others could separate them. Moreover, it would be lawful to integrate the two products even if others could not disintegrate them if that were necessary to achieve an efficiency goal. Accordingly, Microsoft’s exclusion of the browser from the add/remove program was unlawful because that exclusion was not required for efficiency reasons. The court further ruled that Microsoft’s commingling browser and operating system code in the same files was unlawful because it prevented third parties from separating the two products, and there was no efficiency reason shown for the commingling. But Microsoft’s action in causing Windows to override the user’s choice of a default browser in a number of situations in which the default browser would be incompatible with the Windows operating system with which it was being employed was ruled lawful because there were objective reasons for the override. Thus copyright trumped antitrust law where substantial copyright interests were involved. The most dramatic ruling in the Microsoft case, however, involved the court’s decision to evaluate the integration of the browser with the operating system for Sherman Act Section 1 purposes under the Rule of Reason. This required the court to create an exception to the per se rule against tying for platform software. As noted elsewhere, the court did this in the interest of fostering innovation. This ruling was not based on intellectual property considerations but solely on the court’s understanding that innovations involving platform software generally involve the additions of new functionalities that have previously been available as separate products. Yet although the court decided the tying issue under Section 1 favorably to Microsoft, it decided that the same behavior violated Section 2. The Section 1 and Section 2 issues were different, because the focus of the Section 1 issue was on competition in the market for the tied product (browsers), whereas the focus of the Section 2 issue was on Microsoft’s “maintaining” its platform monopoly. By integrating its own browser into the Windows operating system, Microsoft made it harder for Netscape to secure the level of usage that would be necessary for it to develop as an alternative platform. The EU Microsoft Decision and Its Case Law Background: What Is the Extent of the Obligation to Share Intellectual Property? The EU Microsoft case of

2007 36 arose out of complaints to the European Commission by Microsoft’s competitors about its methods of doing business. The first part of that decision dealt with a complaint by Sun Microsystems. Sun produced server software that competed with Microsoft’s server software. The point of contention concerned the degree of integration attainable by Sun servers

chapter nine

172

in Windows work-group networks. Work groups are commonly composed of a server or group of servers and a larger number of so-called client PCs. Most of the client PCs in small- and medium-size work groups employ (and have employed) Windows operating systems. Although Microsoft was late in entering the server operating-systems market, by the 1990s increasing numbers of servers were using the Windows NT server operating system. As a result of Windows’ growing share of the server-software market, many work groups employed Windows server operating systems principally or even exclusively. Nonetheless, many work groups in which most servers employed Windows operating systems also included servers using rival operating systems (like Sun’s). The introduction of Windows 2000 server operating systems changed the competitive context. Whereas previously rival servers could operate in a work group in the same way as Windows servers, now they could not. At the same time, Windows servers with the new software acquired capabilities beyond those possessed by Windows NT or rival server software.37 The result was that the high degree of integration in the Windows 2000 system heightened the advantages of Windows 2000 server operating systems over rival server operating systems. Rival systems (such as Sun’s) could operate in a work-group system network (as so-called member servers but not as domain controllers) and communicate with PCs, but they lacked the capabilities of communicating with other servers to the degree possessed by the Windows 2000 server software. The increased capabilities of the Windows 2000 servers over their predecessor Windows NT servers and the rival servers increased their attractiveness to buyers. If Microsoft had merely increased the capacities of its server operating systems, the increased attractiveness of its server software vis-à-vis rival software would have been legally unproblematic. This would merely have been the market at work fostering competition in quality. But the increased capabilities of the new system turned on intercommunication among servers so that tasks could be allocated among them and so that they could draw on each other for assistance. This interaction and cooperation among servers combined to confer increased power and abilities upon Windows work groups employing the Windows 2000 server software. The ability to write server software enabling servers to communicate among themselves was not the exclusive property of Microsoft. Sun and other Microsoft rivals were capable of writing similar programs.38 Sun could market server operating system software with characteristics similar to Windows 2000 that provided similar intercommunication among its own servers. Such a move by Sun would challenge Microsoft with competition at the work-group level. Competition within work groups in which suppliers of

intellectual property and antitrust in dynamic industries

173

server operating systems vied for sales of those operating systems for use by individual machines, however, would be handicapped because providers of alternative server software would not have access to the protocols through which Windows 2000 servers communicated with each other. On the assumption of competition among suppliers of server operating systems at the work-group level (and suppliers of each brand producing server operating systems that communicated with other operating systems of the same brand), operators of work groups needing additional server operating systems would probably opt for additional systems of the same brand as the dominant brand in their own work group for the reasons just discussed. The Commission found that Microsoft held a 60 percent share of the server operating system market (composed of servers priced at $25,000 or less). A likely inference from that finding is that most work groups were Windows work groups, but not necessarily Windows 2000 work groups. Windows 2000 work groups were likely to grow by adding additional Windows server operating systems in order to take advantage of the new capabilities of Windows 2000 server operating systems to communicate among themselves. If 60 percent of the work groups were in fact Windows 2000 work groups, Windows 2000 would be likely to represent at least 60 percent of new sales. The Commission claimed that Windows had a “network advantage.” The extent of these network effects and the ramifications of these network effects for antitrust policy are discussed below. The design of the Windows 2000 software raised highly charged issues about the proper boundaries for private competitive activity. Antitrust law, of course, has always been concerned with these boundaries, but in this case the boundaries were especially complex as they impinged upon definitions of property; the scope that society wishes to extend to inventive activity; the role that society wishes to accord to incentives for innovation; the proper extent of intellectual-property rights; and, ultimately the allocation of decision making between the market and government. In the first part of the General Court’s Microsoft decision, the Court upheld the Commission’s order requiring Microsoft to share sufficient information with its rivals to permit them to produce server software that would operate indistinguishably from Windows 2000 server software. The second part of the General Court’s Microsoft decision considered the lawfulness of Microsoft’s integration of the Windows Media Player into its Windows operating system. The Commission had ordered Microsoft to produce versions of its Windows operating system in which the Windows Media Player was not bundled. Again, the General Court upheld the Commission’s order.39 This part of the decision is more closely bound to prior

chapter nine

174

case law on tying and technological tying. It bears broad similarities to the American case law, and its consideration provides background for a discussion and assessment of technological tying. The Case Law Background to the EU Microsoft Decisions. The competition law

of the European Community began to develop deep inroads into intellectual property with the Magill case in 1995, which appeared to treat copyright as an essential facility. Thereafter, the European law has continued in that direction, culminating in the General Court’s Microsoft decision in 2007, where the differences between the European and American antitrust laws reached their zenith. In Magill, the Court of Justice imposed a duty to license copyrighted materials on the copyright owners. Magill involved a publisher who wanted to produce a periodical that listed program schedules for the several television stations serving Ireland. Although at that time each television station published its own programming guide, there was no comprehensive guide, and the publisher (Magill) was seeking to fill this void. Magill was thwarted, however, by the refusal of the television stations to grant permission under the copyright law to publish the schedules. (The European Commission and the courts assumed that Irish copyright law gave the television stations the right to control publication of their schedules.) Magill complained to the European Commission. The Commission ruled that the television stations held dominant positions in the market and that their refusal to license their schedules to Magill was an abuse of those positions within the meaning of what is now Article 102. Both the General Court and the Court of Justice upheld the Commission.40 The Court of Justice ruled that the mere ownership of an intellectual property right cannot confer a dominant position.41 It also reaffirmed the statement in its earlier Volvo decision42 that the refusal to license intellectual property, even by a dominant firm, cannot alone constitute an abuse, but it also repeated Volvo’s other statement that the “exercise” of an exclusive right could, in “exceptional circumstances” involve an abuse.43 The ECJ, however, ruled that because the television stations were the only source of the information needed to publish a programming guide, they held a de facto monopoly over that information, enabling them to prevent competition in the market for programming guides. As a result, the stations were held to possess a dominant position. The ECJ found the existence of “exceptional circumstances” giving rise to abuse lay in three factors: (1) the stations’ unwillingness to license prevented the appearance of a new product (that is, a weekly television guide) that the stations themselves did not provide and for which there was a po-

intellectual property and antitrust in dynamic industries

175

tential consumer demand, (2) their refusal to license was not objectively justified, and (3) the stations reserved to themselves the secondary market of weekly television guides by excluding all competition in that market.44 Within three years of the Magill decision, the Commission and the courts faced two more claims involving what amounted to an essential facilities doctrine. In 1997 the General Court decided Tierce Ladbroke, a case involving a Belgian bookmaking company’s attempt to secure rights to televised pictures and commentary on French horse racing.45 The French horse racing association had granted a German company exclusive rights to exploit the televised pictures and commentary in much of Germany and Austria. When Ladbroke, the Belgian bookmaking company, was denied access to these materials, it asked the Commission for relief. Ladbroke contended that it was entitled to access to these materials under Magill. When the Commission denied relief, Ladbroke appealed to the General Court, which also denied the requested relief. In its discussion, the General Court employed language suggesting that requirements of Magill could be met not only by showing that access was necessary to provide a new product, but those requirements could be met, as well, by a showing that access was “essential” for the claimant’s business, and no substitutes were available. The refusal to supply the applicant could not fall within the prohibition laid down by Article 86 [102] unless it concerned a product or service which was either essential for the exercise of the activity in question, in that there was no real or potential substitute, or was a new product whose introduction might be prevented, despite specific, constant and regular potential demand on the part of consumers.46

This interpretation appeared to gain support from the decision of the Court of Justice the following year in Bronner,47 a case in which the plaintiff based its claim upon an “essential facilities” doctrine that it also found in Magill. Bronner, the publisher of a small-circulation newspaper in Austria, wanted to use the home-delivery system established by Mediaprint, whose two papers held over 46 percent of the Austrian daily newspaper market, as a means for the delivery of its own papers. Although Bronner based its claim to use Mediaprint’s delivery system upon the essential facilities doctrine, the ECJ carefully avoided ruling upon whether the standards governing refusals to deal in intellectual property governed refusals to deal in ordinary property. It rejected Bronner’s claim on the ground that even if those standards applied, Bronner failed to meet them. It said that the standards of Magill would require Bronner to show: (1) that the refusal of the home-delivery service was

chapter nine

176

likely to eliminate all competition in the daily-newspaper market; (2) that the refusal was not objectively justified; and (3) the service was indispensable to Bronner’s carrying on of its business, since there were no substitutes for the home-delivery service.48 On indispensability, the ECJ ruled that Mediaprint’s delivery service was not indispensable, because there were substitutes, such as distribution through the mail and through shops and kiosks.49 These aspects of indispensability, it will be observed, were not found in the General Court’s earlier decision in Ladbroke. Because Bronner was unable to show that there were no substitutes for Mediaprint’s home-delivery system, the ECJ ruled against it. In addition, the ECJ also added a significant economic dimension to this indispensability element. It ruled that in assessing indispensability, economic viability should be keyed to the dimensions of the actual delivery service rather than to the size of the claimant’s business; a service would not be deemed economically indispensable unless it were established that a second delivery system with a circulation comparable to the existing newspaper delivery system (that is, with a circulation comparable to Mediaprint’s) were shown to be economically infeasible.50 The decisions of the Court of Justice in Magill and Bronner (and the General Court’s decision in Ladbroke) were followed by IMS Health.51 IMS was the only firm in Germany collecting data on a regional basis about the use of pharmaceutical products. In connection with data-collection activities, IMS had developed a classification system keyed to small geographical units, referred to in the case as a “brick structure.” The pharmaceutical companies, for whom IMS was collecting usage data, keyed their own operations to this classification system. Rival companies could not enter the datacollection market (or so they claimed) without using the same classification system. IMS, however, asserting its rights under copyright law, refused to license its system to those rivals. The European Commission intervened, ordering IMS to license its rivals to use its system. Under the Commission’s view, IMS’s copyright conferred a dominant position upon it, and its refusal to license its rivals was an abuse of that position. Thus its refusal to license fell under Article 102’s condemnation of abuses of dominant positions. The Court of Justice ruled in the IMS case that three conditions would have to be satisfied before a refusal by a copyright holder in the position of IMS to license its system could constitute abuse: (1) the firm requesting the license intends to use the system to offer new products or services not offered by the copyright owner for which there is a potential consumer demand, (2) the refusal is not objectively justified, and (3) the refusal reserves to the copyright owner the market for the supply of data on sales of pharmaceutical products by eliminating all competition in that market.52 The ECJ explicitly ruled that the three conditions were cumulative.53

intellectual property and antitrust in dynamic industries

177

Although the Court of Justice set forth the standards for determining the application of Article 102 in the context of the IMS facts, it remanded the case to the national court to apply those standards. At this point, many observers saw IMS as holding a strong position because IMS’s rivals wanted to use the IMS system themselves, not to employ it as an input for the development of a new product. Hence the refusal by IMS to license its brick structure would not prevent the emergence of any new product.54 The conditions set forth in IMS resemble the conditions set forth in Magill and differ from the Bronner conditions. Perhaps that is because the focus in IMS and Magill is upon licensing intellectual property to potential new entrants, whereas the focus in Bronner is upon making a service available to a rival already in the market. Thus the first condition in IMS and Magill has to do with one or more firms entering the market with a new product, whereas Bronner concerns not a possible new product but the claim of a rival already in the market and already offering its own version of an existing product to assistance from a larger rival. The third condition in IMS and Magill has to do with the firms reserving to themselves an entire market. Bronner’s third condition is cast in the language of essential facilities: the service concerned (in Bronner’s case, a delivery service) is indispensable to the continued operation of the rival firm. Despite their differences in phrasing, the Magill/ IMS test is quite similar to the Bronner test. The third condition of the Magill and IMS decisions refers to eliminating all competition in a second market, as does the first condition of the Bronner test. The second condition of all three cases is the same: that the refusal is not objectively justified. The real difference is that the first condition of the Magill and IMS test is that the refusal to deal prevented the emergence of a new product, while the third condition of the Bronner test is that the refusal to deal concerned a product that was “indispensable” to Bronner’s continuing in business. Bronner thus focuses on the continuance in business of an existing firm that markets an existing product. The superficial tension between the Magill/ IMS test and the Bronner test is explained by the different concerns of the two tests: the enrichment of the market with a new product, or the mere survival of one or more existing firms marketing a product already in the market. Since the Bronner opinion describes the situation in Magill as the refusal to supply an element that was indispensable to the publication of the proposed new television guide, the ECJ, in Bronner, gave the appearance of equating the emergence of a new product with the continuance of an existing business. Nonetheless, the ECJ’s ruling in Bronner suggested that it would be harder for the owner of an existing business to qualify for relief than for a potential supplier of a new product. Bronner differed from the other cases, also, in that it was the only

chapter nine

178

decision by the Court of Justice in the series in which intellectual property was not involved. Even the General Court’s Ladbroke decision involved licensing rights to intellectual property. The ECJ’s decision in IMS Health to reemphasize the new product criterion in an intellectual-property context suggested to some observers that the earlier Magill formulation would prevail, at least in the intellectual-property sphere. But as pointed out above, there may be no inconsistency between the rulings in Magill and Bronner, because they apply to different situations. The General Court’s Ladbroke decision accepted both formulations as describing then-existing law, on the rationale previously described. In any event the ECJ, in its IMS decision, incorporated Bronner’s indispensability criterion into the new product condition: the refusal to license concerns intellectual property that is indispensable for the offering of a new product.55 As shown below, the Commission and the General Court reinterpreted these precedents in the Microsoft case in such a way as to cast an immense cloud of uncertainty over the EU law governing refusals to deal. In the Microsoft case, the Commission and the General Court faced two major issues: (1) The extent to which a firm possessing a dominant share of the serversoftware market is required to supply information to its competitors that would enable them to produce new server software competitive with that of the dominant firm, and (2) the extent to which a firm possessing a dominant share in the personal computer operating systems market is free to integrate new functionalities into its operating system. The first issue is a novel one in the antitrust context. The second issue was previously the subject of antitrust litigation in the United States. What Does the Server-Software Part of the Microsoft Decision Do to the European Precedents? In its Microsoft decision, the General Court radically

reinterpreted the European precedents. Thus the first condition set forth in Magill and IMS that makes a dominant firm’s refusal to license an abuse is that the refusal prevents the appearance of a new product. Microsoft had contended that providing rival firms access to its protocols would merely enable them to upgrade their server software. The rivals would then be able to market higher quality server software in competition with Microsoft, but they would not be offering a new product. The General Court responded that the rivals in fact would be offering a new product, because their version of server software would probably embody some characteristics that differed from the Microsoft version. In so ruling, the General Court gave new content to the first condition as articulated in IMS and Magill. A new product, within the meaning of the case law, now can be merely a differentiated variant of a preexisting product.56 To be sure, European commentators had

intellectual property and antitrust in dynamic industries

179

earlier flagged the “new product” condition as one laden with ambiguity. One commentator pointed out that in using the open-ended “new product” establishing this condition, the General Court avoided using concepts having established meanings like “substitutes.”57 In exploiting this openness, the General Court has created a potentially dangerous ambiguity. The new product condition was providing assurance that dominant firms would be required to license their intellectual property only in exceptional circumstances. As the new product condition weakens, so does the criterion of “exceptional” circumstances. Since virtually all products (other than commodities) come in a range of differentiated versions, the General Court ruling means that every rival firm in an industry composed of differentiated products is selling something that when introduced was in some sense a “new product.” Thus the ruling effectively abolishes the “new product” requirement. Just as the Lerner index of “monopoly power” has a range of very high to trivial because it simply reflects a product’s price inelasticity, so now does “new product” in Europe apparently become limitlessly ambiguous for exactly the same reason. Because the General Court asserted that a new product can be a differentiated version of an old product, a new product need not embody significantly different characteristics from those of rival products. Moreover, Magill had suggested that the new product (a TV guide) would serve a theretofore unserved demand. The Microsoft decision has changed that. It is no longer a requirement that a new product serve a previously unserved demand. Read literally, the Microsoft decision is a precedent under which the Commission can order large firms to license their intellectual property to their competitors. Although the holding dealt with interoperability protocols, the General Court’s opinion explicitly applies to intellectual property generally. And since a market share of less than 50 percent can confer dominance, the ruling is widely applicable. Although the Microsoft decision appears to be a major departure from prior rulings, such a drastic change in the legal landscape may have been deemed necessary to confront a major competitive threat. The following section identifies the behavior that gave rise to the General Court’s Microsoft decision, as well as the continuing patterns into which that behavior falls. Product Integration: Tying Doctrine and the Incorporation of the Windows Media Player into the Operating System. The General Court’s ruling that the in-

tegration of the Windows Media Player into the Windows operating system constituted an illegal tie appears consistent with the prior EU case law on tying arrangements. The language of Article 102 targets tying by a firm possessing a dominant position,58 and a series of precedents have treated such

chapter nine

180

behavior as an unlawful abuse of a dominant position. The Commission and the General Court explained that normally tying by a dominant firm can be summarily condemned because the tie would be presumed to foreclose competitors from marketing their products in the tied-product market.59 In this case, however, because independent sources for the tied product were available on the Internet at no charge, the Commission decided to examine the market effects of the tie before condemning it, an approach of which the General Court approved.60 The case also differed from traditional tying cases in another way. Here the Media Player was integrated into the Windows operating system, creating a so-called technological tie. It appears to be the first case in which the European courts have separated the component parts of an integrated product to find an illegal tie. In determining whether the integration of the media functionality into the operating system should be seen as a tie, both the Commission and the General Court relied upon a consumer demand test,61 a test that had originated approximately twenty years earlier in the US Supreme Court’s decision in Jefferson Parish Hospital v. Hyde.62 Both the Commission and the General Court concluded that the media player was a product separate from the Windows operating system, because media players, including some versions of the Windows Media Player itself, were separately available on the market.63 Accordingly, there were two products (the media player and the operating system), and Microsoft’s action in integrating the media player into the operating system constituted a tie. The General Court chose to approach the issue involving Windows 2000 server software as one of interoperability. When the Court dealt with the media player, it treated that issue as one involving tying. Underlying both issues, however, was the deeper (and more complex) issue of product design. Comparing the European, American, and other Approaches: The Potential for Convergence The Media Player. There are significant similarities between the approaches the European Commission and the General Court took in their Microsoft decision and those the DC Circuit took earlier. Like the DC Circuit on the Section 1 tying issue, the European Commission decided not to apply a per se rule. Rather, the Commission ruled that an inquiry into the competitive effects of the tie was warranted. The rationale for this ruling, however, was somewhat different from that underlying its American counterpart. The European Commission decided against applying a per se rule to the tie because alternative non-Microsoft products were available and were used by com-

intellectual property and antitrust in dynamic industries

181

puter users employing the Windows operating system. The US court decided not to apply the otherwise applicable per se rule because it feared that applying the per se rule would deter innovation, which often takes the form of adding a new functionality to the operating system. The European decision thus rested on a conventional static demand analysis while the DC Circuit was based on a dynamic efficiency rationale. Nonetheless, both rulings are broadly consistent in that they each focused on the competitive effects in the market for the tied product. The rulings of the two jurisdictions on Microsoft’s technological tying are both similar and different. Despite creating an exception to the per se rule for platform software, the DC Circuit ruled that Microsoft’s integration of its browser into the Windows operating system in ways that could not be undone was a violation of Section 2. The circuit court was able to reach this result— which superficially appears to conflict with the policy underlying its Section 1 ruling— because the Section 2 issue was concerned with monopoly maintenance: the preservation of Microsoft’s monopoly over operating systems. The Section 1 issue involved preserving competition in the browser market. The remedial decree, accordingly, is designed to end the irreversible integration that was held to have constituted monopolization. Under the decree any OEM or end user that desires to use a non-Microsoft browser (or any non-Microsoft middleware) is entitled to do so. Indeed, although the government attacked only the integration of the browser into the operating system, the decree extends to all middleware that carries the potential of evolving into a platform, specifically including the Windows Media Player (WMP). Accordingly, in both jurisdictions the law governing Microsoft’s integration of the WMP is broadly similar. In neither jurisdiction can Microsoft require consumers to accept the integration of the WMP into the operating system. In both jurisdictions consumers are entitled to choose their media player. Indeed, the laws in the United States and in the European Union as they apply generally to technological ties by dominant firms, although superficially conflicting, actually bear broad resemblances. Thus despite the effort of the European Commission to study the effects of the tie rather than to apply a per se rule, the Commission concluded that the technological tie was unlawful because of network effects. The approach of the DC Circuit on the tying issue was not dissimilar. Although the US court refused to apply the per se rule for fear of discouraging innovation, that court, as the following discussion shows, was apparently comfortable with a legal framework in which the tie could be condemned as an attempted monopolization of the browser (the tied product) market. In the Microsoft case, the government lost on the government’s claim of attempted monopolization only because

chapter nine

182

it failed to prove a separate browser market and failed to prove that entry barriers protected that market.64 A Closer Look at the Ruling of the DC Circuit. The DC Circuit could, with-

out inconsistency, take the position that the Rule of Reason should govern the integration of new functionalities into the operating system in order to foster innovation while simultaneously ruling that it should be condemned when the effects of such an integration threatened to preserve the existing operating system monopoly. Indeed, the Section 1 ruling merely placed the burden of proof on the government to prove anticompetitive effects. For procedural reasons, the government was impeded from carrying that burden.65 It is unclear whether the government could have carried that burden in any event. Traditionally, tying law has focused on competition in the market for the tied (here, browser) market. Under the court’s ruling the tie was in effect presumed lawful. To rebut that presumption, the government would have had to establish— under traditional approaches to tying under Section 1— that negative effects in the browser market were sufficient to offset the advantages of innovation, and success would have depended on the proof standards.66 But if the court had allowed the government to carry its burden by showing anticompetitive effects in the market for the tying product, then the government would have been able to employ its Section 2 case in aid of its position on the Section 1 tying issue. An even more interesting question is whether the DC Circuit’s ruling on the Section 1 tying issue would have precluded it from ruling that Microsoft had attempted to monopolize the browser market if the district court’s findings had not been deficient on that issue. In the Supreme Court’s 1966 Grinnell decision,67 the Court gave a definition of monopolization. In that case, the Court defined monopolization to include both a monopoly and the acquisition or the maintenance of a monopoly through behavior other than “growth or development as a consequence of a superior product, business acumen, or historic accident.” Other cases have ruled that tying behavior can constitute monopolization.68 In such cases, a defendant’s use of a monopolized product as a tying product can involve “leveraging” of that monopoly. Clearly, leveraging a monopoly in one product in order to gain a monopoly in another would involve monopoly acquisition and thus would meet one prong of Grinnell’s definition of monopolization. In the Microsoft case, the government could not have charged Microsoft with monopolization of the browser market because Netscape, the rival browser, still appeared to control a major market share of that market.69 Instead, the government contended that Microsoft was attempting to monopolize the browser market; it lost on that issue because it failed to prove the existence

intellectual property and antitrust in dynamic industries

183

of a browser market and its protection by barriers to entry.70 Because the government prevailed on its charge of monopoly maintenance and failed to muster the requisite evidence that would have supported its attempted monopolization allegation, legal issues connected with the acquisition prong of the Grinnell test did not have to be resolved. Accordingly, we do not know whether the DC Circuit’s ruling on the Section 1 issue would have affected the ability of that court to decide an attempted monopolization issue, had it been supported by findings identifying the market and findings of entry barriers. We can approach that question analytically, however. The decision to consider approval of the “tie” of a new functionality to the operating system necessarily sets limits to the use of the attempt-to-monopolize clause, since otherwise the attempt clause could negate the entire purpose of the innovation-fostering rationale. Conversely, however, the attempt clause could be used to effectively set limits on the circumstances in which a new functionality could be added in an unremovable way to the operating system. Adding a new functionality to the operating system can be understood as a technological tie, but one done in a fixed one-to-one ratio. According to the single monopoly-profit theorem, an operating-systems monopolist could not increase its profits by tying a media player or other functionality to the operating system because buyers in the aggregate would be unwilling to pay for the combination a price that exceeded the monopoly price for the operating system plus the market price for the functionality. But some functionalities (like browsers and media players) are sometimes given away. A market in which the product is given away is unusual but not anomalous. The traditional newspaper market is the paradigm case in which the product, while usually not given away, is subsidized. And the newspaper is the model on which a multiplicity of web-based services are based. A producer who gives away a functionality like a media player may be relying on revenue to be gained from charges imposed on music distributors, advertisers, or users, just as a newspaper publisher who sells its paper at a subsidized price is relying on revenues from advertisers. Significant revenue from any of those sources would make the media player market independently important economically. This, in turn, would mean that Microsoft would be using (or leveraging) its monopoly in operating systems to acquire (or attempt to acquire) a monopoly in an independent market. Conversely, maybe the operating systems monopolist is merely competing, using its media player to improve its operating system monopoly to stay ahead of potential competitors. This analysis can be understood as an elaboration of the DC Circuit’s approach. That court’s premise was that the incorporation of new functionalities into platform software should be encouraged when that incorporation was efficient. On this analysis, the Rule-of-Reason approach to the

chapter nine

184

tie of the operating system to the new functionality would govern where it was more efficient for the functionality to be used as a part of the operating system and the functionality lacked independent significance as a revenuegenerating device in its own right. If the US courts followed this reasoning, they would be close to the position of the European Commission and the General Court not only in result (Microsoft forbidden to integrate the WMP into the operating system) but also in condemning that kind of integration regardless of whether Microsoft had maintained its platform monopoly, whenever the tied market had independent economic significance. Examined from this perspective, the divergence of the antitrust law between the two jurisdictions on the technological tie issue would be narrower than superficial appearances suggest. Server Software. The European approach to server software might strike an outside observer as one designed to foster rivalry in ordinary production, ignoring the role of rivalry in stimulating extraordinary technological advance. Another way of describing the European approach would be as one focused on static, as opposed to dynamic, efficiency. Accordingly, American antitrust observers are likely to wonder how the next technological advance will be engendered, now that market participants are aware that the European competition-law authorities can deprive them of the economic benefits of their innovations. Such a view would be misleading, however, because the European authorities are indeed concerned with innovation, and both the Commission and the General Court have weighed the effects of their decisions on innovation. They believe that although Microsoft’s incentives to innovate may be diminished, this effect will be outweighed by “follow on” innovation by other firms. Because Microsoft did not make its server protocol information available to its rivals, the rivals were effectively barred from developing innovations that exploited server interconnectivity, “following on” to Microsoft’s original innovation. Once Microsoft’s hold over industry development has been broken, other firms would be free to innovate. And, in the view of the Commission and the General Court, on balance the total amount of innovation would exceed the level of innovation in the industry prior to the Commission’s ruling.71 Network Effects. Network effects are those that increase the value of a prod-

uct as the number of its users increase, operating on the demand side of a market as an analogue to scale economies on the supply side.72 They were critical to the rulings of both the Commission and the General Court. A product generating network effects achieves a significant edge over its rivals when it attracts a larger usage base. When this occurs, its value relative to

intellectual property and antitrust in dynamic industries

185

rival products increases, and its lead over rival products grows rapidly, giving it an advantage that is increasingly difficult for the rival products to overcome. The product that acquires this advantage may not be technically the highest quality product; a product of inferior quality may acquire a practical monopoly.73 Network effects are double edged. When a product produces them, it generates an advantage for the first mover, privileging the producer with the largest market share. Microsoft’s operating system generated a virtuous cycle as software applications producers offered more for its operating system and its operating system became more attractive to consumers. These mutually reinforcing effects generated its near monopoly in personal computer operating systems. Thus network effects produce entry barriers,74 making it difficult for producers of rival operating systems to challenge Microsoft’s dominance. Overall, network effects make competition among rivals (say, rival producers of operating systems) a competition for dominance. As proponents of Schumpeterian competition75 would say, competition in this circumstance is competition “for” the market, rather than within the market. Although network effects are inherent in some products, they have sometimes been controlled either by voluntary industry standards (for example, phonograph record and compact disc specifications) or public intervention (wireless network interfaces in much of the world). The network effects that produced the Windows operating system monopoly were dependent upon Microsoft’s control over the Windows application programming interfaces (APIs). Microsoft was the beneficiary of the network effects those APIs generated. If the APIs had been set by an industry standard-setting organization, then the network effects would have attached to the industry-set APIs rather than those unique to Windows. Windows would then have been competing with other operating systems (such as IBM’s OS2 Warp or Linux) primarily on the technical characteristics of each operating system. The European Commission’s decision to require Microsoft to disclose its server communication protocols to its competitors in a form that would enable non-Microsoft server software to interoperate “seamlessly” with Microsoft server software appears to replicate the effects of an industry standard, although the Commission denied that it was requiring the disclosure of source code.76 By contrast, the US Microsoft final judgment, although also requiring the disclosure of server/client protocols, was designed to prevent rivals from “cloning” the Windows operating system77 and did not require disclosure of protocols governing server-to-server communications. In so limiting the extent of disclosure, the US court relied in part on the negative effects such a requirement would have on Microsoft’s incentives to innovate.

chapter nine

186

Because network effects can transform competition into competition for dominance (or into competition “for” the market), that transformed competition carries the prospect of large rewards, thus stimulating the search for highly innovative solutions. Because Microsoft was— and is— competing “for” the market where the rewards are immense, it has a huge incentive to maintain that dominance by generating technologically superior operating systems. Thus network effects act both as entry barriers and as incentives to innovation. The Commission and the General Court appear to have taken the view that the role of network effects is generally anticompetitive. They see network effects not as inherent in the way the market operates for certain classes of products but as introducing a major distortion of the competitive process that competition authorities should suppress so far as possible. In the Microsoft case, the European Commission was concerned about network effects two of Microsoft’s products generated. First, the Commission believed that the network effects from Microsoft’s Windows operating system explained the demand for its server software because users of the server software wanted the enhanced compatibility with operating systems that came with its server software. The critical network element, however, was at the second stage. Users were often influenced to select Microsoft server software because of the enhanced communicative abilities of that  software among servers in the same work group. But once Microsoft server software gained that initial advantage, its own network effects took over. Then the enhanced communicative capabilities of Microsoft 2000 server software vis-à-vis rival server software practically insured Microsoft’s eventual dominance within Microsoft work groups, that is, within work groups in which Microsoft server software played a significant role. Even if other firms could market server software that was able to communicate within the same brand, Microsoft’s initial advantage multiplied by network effects could not be overcome. In other (non-Microsoft) work groups, a drift away from increased encroachment by Microsoft server software would seem to depend upon the extent to which Microsoft’s rivals provided software with interserver communicative abilities similar to, or exceeding, those of Windows 2000. The Commission took a similar position with respect to the WMP. As an integral part of the Windows operating system, the WMP benefited from the network effects the operating system generated.78 Although the WMP was attractive in its own right as it was the only media player that freely provided all the functionalities currently available in a media player, the Commission saw the technologies embodied in the WMP as representing a power imbalance between the WMP and its rivals. Because other media players

intellectual property and antitrust in dynamic industries

187

licensed some or all of their technologies from others (including Microsoft), the Commission doubted that they could be relied upon to constrain Microsoft’s behavior. US law recognizes that network effects may produce a single winner among competing products. Because of the structure of US law, it accepts that market result so long it does not involve monopolization or attempted monopolization. Because these are relatively well-defined categories, there is less flexibility in US law to interfere with the results of network effects than there is under the more elastic EU concept of abuse of a dominant position. Nonetheless, as explained above, the leveraging of network effects generated in the operating systems market to foster the use of the media player could have supported an independent attempted monopolization case if the existence of an independent media player market were established.79 The server-software matter is somewhat more complex. The European Commission (backed by the General Court) wanted Microsoft to share information about its protocols with its rivals. In this circumstance, Microsoft was benefiting from network effects as they developed within each work group, but those network effects are generated by the server protocols that were developed as part of the enhanced server interaction. Here Microsoft anticipated the success of its product and relied on those network effects when it invested in that product, and it seems that the innovation (enhanced server interaction) illustrates the positive side of network effects. Accordingly, and as more fully developed below, it appears that US antitrust law would take a position diametrically opposite to that the European authorities embraced. Nevertheless, there are a number places that US law embraces policies supportive of the European rulings on interoperability and that could lend support to a ruling by a US court requiring disclosure of the server protocols on antitrust grounds. These can be found in copyright law, in the consent decree, in patent remedial law, and in the copyright misuse doctrine. Other US Policies Supporting Interoperability as a Social Goal. US law provides

several analogues to the obligation European competition law imposes to share protocol information with rivals. First, US copyright law can be read as evidencing a public policy favoring interoperability, and the antitrust decree terminating the Microsoft antitrust litigation imposed an obligation on Microsoft to disclose protocols to its rivals in the software application market. Under US copyright law the protocols receive no protection, and, indeed, copyright law incorporates policies favoring interoperability. Access code (the portion of machine code that functions like a gate to a device or

chapter nine

188

system, admitting or excluding external software) falls under the scope of a “system” as used in Section 102(b) with the result that it is unprotectable. Indeed, the unprotectability of access code is consistent with Congress’s decision to extend protection to computer programs as “literary works,”80 and thus by implication to provide no protection for their utilitarian aspects. Indeed, that is the thrust of decisions like Altai,81 which denied protection to efficiency-enhancing aspects of computer programming. The mere unprotectability of access code, of course, does not make it practically available to others. The firm employing the access code may choose not to reveal it, protecting it (or attempting to protect it) as a trade secret. Indeed, in a number of cases, rivals seeking an access code had to decompile a software program in order to identify the access code. This, in turn, raised the issue of copyright infringement when all or substantial amounts of a software program had to be copied and decompiled in order to discover that code. The courts, however, have found that copying source code as a step in identifying the access code was protected under the fair use doctrine.82 The policy supporting the lawfulness of reverse engineering of access code for the purposes of achieving interoperability was endorsed by Congress in the Digital Millennium Copyright Act of 1998, where an exception to a general prohibition on circumventing technology controlling access was provided for reverse engineering keyed to achieving interoperability.83 In short, several strands of copyright law reflect policies favoring interoperability, effectively recognizing interoperability as in the public interest. Second, although US copyright law recognizes interoperability as positive and denies protection to access code, copyright law itself does not require affirmative disclosure of that code. Mandatory disclosure similar to that required under the General Court order, however, was in fact ordered by a US antitrust court in the Microsoft litigation, where the decree terminating the contest between Microsoft and the US Justice Department contained a provision requiring mandatory disclosure of application program interfaces and protocols governing client-server communications.84 The terms of the consent decree were then imposed by the court on several nonsettling plaintiff states and were upheld on appeal by the Commonwealth of Massachusetts.85 The rationale for these obligations was partially to ensure that Microsoft could not use its control over the Windows operating system to confer a technical advantage on its own application software (such as Word and Excel) in competing with rival software (such as WordPerfect and Quattro Pro) as well as to foster competition with the operating system from middleware. The disclosure of the APIs was directed to software loaded on PCs, and the disclosure of the client-server protocols was directed to server-

intellectual property and antitrust in dynamic industries

189

based software PCs used. The aim was to preserve competition in the downstream markets. One aspect of this order could affect competition in the operating systems market, however. To the extent that middleware is potentially capable of developing into a competing platform (the theory underlying the antitrust case) the disclosure of the APIs preserves this (perhaps remote) possibility. But the disclosure obligation under the US judgment differs from the obligation the European decree imposed requiring Microsoft to share server-to-server protocol information with rivals who are capable of immediately using that information to compete in the same (server) market. Moreover, since Microsoft had agreed to these provisions, they do not say much about the authority of a US antitrust court to override trade-secret law (which varies little among the states) over the objections of a rights holder. Third, other aspects of intellectual property law provide some support for the US legal system reaching results similar to those reached in Europe. The Supreme Court’s decision in eBay Inc. v. MercExchange, LLC86 held that injunctive relief is not always appropriate in patent infringement cases. To the extent that injunctive relief is not obtainable, however, the patent is protected not under a property rule but under a liability rule. Others can use the patent, subject only to their paying compensation to the patentee, and the patent in effect becomes subject to compulsory licensing. Only the amount of compensation to the patentee is in issue. eBay carries ramifications, not only for patent remedies but for copyright remedies as well. To the extent that eBay would sanction only a remedy in damages, that result would be analytically similar to the ruling of the General Court that Microsoft’s protocols were to be made available to its competitors, subject to a reasonable royalty. The difference between Microsoft’s claims that its protocols were protected by patent, copyright, and trade-secret law and the position of the European Commission and the General Court that it was obliged to provide protocol information to rivals in the server market can be reduced to a difference between whether that intellectual-property protection is accorded under a property or liability rule. eBay shows the weakness of that distinction under US law. Fourth, under US patent and copyright law, anticompetitive use of an intellectual property right can be treated as misuse, with the result that the right is not enforceable as long as the misuse continues.87 The copyright misuse doctrine makes the copyright unenforceable whenever the right holder employs his copyright to impose a restraint beyond a mere license to use. This can involve use of the copyright to impose a tie, or, even to impose an exclusive-supply contract.88 Microsoft’s tie of the WMP to its Windows operating system (were it not transformed by the provisions of the consent

chapter nine

190

decree) could constitute copyright misuse. Indeed, Microsoft’s tie of its Internet Explorer browser to the Windows operating system appears to have been vulnerable to the misuse doctrine. It is possible, however, that a contemporary court would modify the misuse doctrine in order to bring it into line with the ruling of the DC Court of Appeals in the Microsoft antitrust litigation under which platform-related ties challenged under Section 1 are evaluated under the Rule of Reason. Under such an approach the addition of new functionalities to an operating system would no longer be treated as misuse but presumptively a proper use. Such a ruling would be a move toward harmonization of the US antitrust and copyright laws. The impact of Section 2 on the addition of new functionalities to the operating system is discussed below. Special Rules for the “New Economy”? Scholars have distinguished between drastic innovation and incremental innovation, especially in the industries of the so-called new economy. In an influential paper David Evans and Richard Schmalensee have argued that a class of newer industries can be identified as possessing characteristics that differ from their traditional counterparts. These industries tend (1) to have high-fixed (sunk) costs and low marginal production costs; (2) to make more intensive use of labor and less intensive use of tangible capital; (3) to generate network effects; (4) to be involved in competition to create or replace an existing product through drastic innovation, which often involves winnertake-all races; (5) to provide industry leaders with substantial amounts of supra competitive profits (viewed ex post); and (6) to have firms whose fortunes are tied to success in the creation of intellectual property.89 Evans and Schmalensee argue that these industries are distinguished by “Schumpeterian” competition in which aggregate welfare is furthered, not by static efficiency or even through the fostering of incremental innovation. Welfare is instead furthered through drastic innovation in which an existing technology is replaced by an entirely new technology, often one that initially is unfamiliar or unanticipated. Each of the industry characteristics identified above reflects this process of creative destruction. Thus the high fixed costs are the result of deploying large amounts of human capital necessary for the research and development that produces a new technology, the winner-take-all result reflects the technological revolution as well as the network effects common to these industries, and the high profits of the dominant firms are necessary to compensate those firms for their high research and development expenses, and

intellectual property and antitrust in dynamic industries

191

risk. These authors argue that antitrust rules should be adjusted to accommodate these conditions. The US position on antitrust treatment of “new economy” industries is in a state of flux. The Antitrust Modernization Commission identified the same characteristics of industries of the new economy as those earlier identified by Evans and Schmalensee and expressed the view that existing antitrust analysis is capable of responding to the special circumstances these industries present.90 The Commission, however, refrained from endorsing some aspects of the creative destruction paradigm Evans and Schmalensee discussed. It left unmentioned the concept of competition “for” the market, as distinguished from competition within the market, as well as the prospect of serial monopolies. The US Department of Justice during the later years of the George Bush administration appears to have been significantly influenced by the Evans and Schmalensee approach. Thus Thomas O. Barnett, former assistant attorney general in charge of the antitrust division, identified several types of efficiency: static efficiency concerned with minimizing deadweight loss, incremental dynamic efficiency to reduce production costs using existing technology, and “leapfrog” dynamic efficiency that generates gains from entirely new ways of producing products or services.91 What Barnett calls “leapfrog” dynamic efficiency refers to the “drastic” innovation just outlined. Barnett concludes that leapfrog dynamic efficiency is so valuable that he appears willing to bend the normal antitrust rules in the interest of fostering it. Thus in discussing Apple’s tie of the iPod to its iTunes music distribution,92 Barnett concluded that society’s welfare was well served by that tying arrangement, despite its apparent conflict with the governing per se rule. Barnett argued that Apple’s music distribution system solved the problem of how to create a consumer-friendly, yet legal and profitable, system of downloading music from the Internet. Apple’s incentive to provide an innovative music distribution system was generated by its ability to capture economic rewards through the tie, an arrangement that fostered Apple’s ability to sell its high mark-up iPod. According to Barnett, whatever the static efficiency of the iPod tie, the arrangement is justified by dynamic efficiency that produces enhanced competitive results. The DC Circuit’s exception to the per se rule for platform software— which it announced in the US Microsoft antitrust suit— is also supported by dynamic efficiency concerns. Adding new functionality to platform software fosters competition among platforms and contributes to the generation of serial monopolies. Although Microsoft has so far been the only monopolist

chapter nine

192

of personal computer operating systems, it must fight to retain that monopoly by constantly improving its product, incorporating new and desirable functionalities into its operating system to keep its edge. The approach of the Obama Justice Department toward industries of the new economy appears to differ from that of the Bush administration and from that of the European Union. Carl Shapiro, who served as deputy assistant attorney general for economics in the antitrust division early in the Obama administration, has rejected the thesis that the new economy requires altered behavioral standards.93 Yet Shapiro expressed sensitivity about the market effects Evans and Schmalensee identified and would incorporate that sensitivity into his antitrust analysis. Thus, for example, he agreed that dynamic analysis is more important than static analysis,94 and that successful firms must recover research and development expenditures, and thus prices must significantly exceed marginal cost.95 He also expressed concern about the anticompetitive potential of network effects and switching costs, which may cause the market to tip toward one supplier, entrenching it permanently in a dominant position.96 Shapiro criticized the Microsoft remedy, contending that it failed to restore a competitive threat to Microsoft’s dominance on a par with the “nascent” threat to its platform monopoly the Netscape browser and Java posed.97 The European antitrust authorities clearly do not believe in according especially favorable antitrust treatment to industries bearing typical new economy characteristics.98 Indeed, the European authorities in Microsoft appear to take network effects as an impediment to competition and see a role for competition policy to neutralize network effects. The European approach appears to be based on the assumption that “follow on” innovation by Microsoft’s rivals will more than compensate for the disincentives to Microsoft’s own innovation efforts. Consequently, European authorities regarded the behavior patterns Evans and Schmalensee identified (especially tying) not as welfare enhancing but rather as welfare reducing. US courts recognize network effects as entry barriers but acknowledge their positive aspects as well.99 As is implicit in this discussion, there are three basic patterns of network effects: (1) they can be voluntarily governed by an industry standard, (2) they can arise from the market’s acceptance of a particular product as in the case of Windows, or (3) they can become common property by government order. Both the Bush and Obama administrators have been tolerant of network effects dominant products generate, but it is likely that US enforcers are also aware of the possibilities for mitigating the anticompetitive potential of network effects by broadening product competition (when feasible) through increasing use of shared industry standards. Yet voluntary industry standards arise only when the configuration of industry

intellectual property and antitrust in dynamic industries

193

participants is propitious. Sometimes the very existence of a new product, such as the Windows 2000 server software with its enhanced server-to-server communication abilities, arises simultaneously with communication protocols governing its operation. In this case, there is no way to reduce the network effects of the product without divesting the innovator of the fruits of its invention. Yet if European antitrust authorities can be criticized for eroding innovation incentives, the US authorities may be faulted for failing to distinguish innovation that is truly revolutionary from innovation that is more commonplace. Indeed, the correct distinction between industries may lie with revolutionary and incremental innovation rather than the “old” and “new” economies. The precise formulation of the former distinction is critically important if different antitrust rules are advocated. For example, Barnett claimed that Apple’s iTunes exemplified drastic innovation. Granted, Apple solved a distribution problem that had eluded others. But was it “drastic” innovation, meriting special antitrust treatment? If so, how did it differ from other successful innovations? The Obama administration appears to favor a focus on individualized firm behavior rather than categorical assessments, thus reducing the need for specifying the boundaries of “drastic” or “revolutionary” innovation taking into account both the special incentives and perils of “new industry” participants and the anticompetitive potential of network effects. This approach is also consistent with agency use of available information to assess the magnitude of innovation attaching to suspect behavior. The logic of Shapiro’s AMC testimony also suggests that government will try to mitigate the anticompetitive elements of network effects where possible. Both the US and European authorities would profit from factually intensive case-by-case analysis, although such an approach will sometimes yield different results. Factually intensive assessment suggests improvements in the handling of the WMP issue in both jurisdictions. The European Commission required Microsoft to offer a version of Windows without the WMP, although it allowed Microsoft to market a bundle of Windows and the WMP.100 In the United States, the integration of the WMP into the operating system is governed by the Microsoft final judgment that requires that both OEMs and users be able to substitute non-Microsoft middleware for any Microsoft middleware product such as WMP. This is essentially the result in Europe as well, since users of the Windows version lacking the WMP would be free to incorporate a media player from a non-Microsoft source. But since the non-WMP version has had few takers, this restriction does not prevent the WMP from exploiting Windows network distribution effects. Both jurisdictions sought to deny Microsoft the benefit of network ef-

chapter nine

194

fects for its WMP, the United States through the Microsoft final judgment and the European Union through the General Court decision, but have effectively been stymied by how to accomplish this end. In context, the simple option would have been to bar the distribution of Windows with the WMP attached and to let consumers download their choice of media players from the web. (Since at the time of the Commission decision, only the WMP freely carried all functionalities available in a media player, the likelihood was that the WMP would be the media player of choice, but this time without benefiting from the network distribution effects the Windows operating system generated.) Conclusions The EU competition law approach to matters of product design (implicating essential facility doctrines, associated obligations to license rivals, and technological tying) appears radically different from US antitrust law, but there are many resemblances beneath the surface. EU developments on the obligation of intellectual property rights holders to license competitors have probably produced the sharpest difference between the two antitrust systems in their approach to Microsoft’s server software. This is where the European Commission and the General Court have expanded their version of the essential facilities doctrine to, in American eyes, the point of deterring innovation. Yet the concept of essential facilities, upon which these rulings are based, was derived from US law. As this is written, the two jurisdictions are on opposite tracks in their approach to that doctrine. The European authorities have been expanding the concept and the corresponding duties to license rivals, while US courts have been narrowing it. Indeed, the European expansion of the essential facilities doctrine in Microsoft came at almost the same time that the US Supreme Court was scaling back that doctrine in its Trinko decision. In Trinko, the US Supreme Court limited the scope of a duty to deal with rivals, indicating that such a duty cannot extend beyond situations like Aspen and has even questioned the existence of an essential facilities doctrine. In short, the two jurisdictions are employing similar doctrines, but they are developing these doctrines differently and are reaching very different results. Both jurisdictions have stepped back from a per se approach to tying in the context of a device like the WMP. The Europeans did not invoke the per se rule because rival products are available for downloading. In the United States the DC Circuit had adopted a Rule of Reason in the Microsoft case to evaluate the incorporation of new functionalities into existing platforms

intellectual property and antitrust in dynamic industries

195

(like the incorporation of the Internet Explorer browser into the Windows operating system). The European authorities have applied a consumer demand test, one that originated in the US Supreme Court’s Jefferson Parish decision, to find that the WMP is a separate product from the operating system and therefore that the integration constituted a tie. In the United States, the DC Circuit essentially accepted the consumer demand test but avoided its implications by applying a Rule of Reason. Further complicating an analysis of the differences between European and US law on product integration is the issue of whether special rules are needed to evaluate behavior in industries of the so-called new economy. European authorities appear to be taking a mostly negative view of characteristic behavior of new economy industries, viewing network effects (ubiquitous in the new economy) as impediments to legitimate competition. The DC Circuit’s creation of an exception to the per se rule governing tying agreement could be understood as a move toward accommodating new economy industries. While the Bush antitrust division appeared favorable to this view, the Obama antitrust division has provided mixed signals, indicating through statements of prominent officials that special rules are unnecessary while also indicating that the special characteristics of new economy firms will be taken into account in enforcement evaluation.

10

A Summing Up One close observer of EU and US developments in competition policy drew the following conclusion in 2012: Although some substantive and procedural differences remain, for example in the analysis of some dominant firm conduct, they result largely from statutes and court decisions, and we are committed to minimizing their impact. Based on the recent record, I expect that remaining differences in US and EC competition policies will continue to diminish.1

Our view is considerably more guarded. This chapter first summarizes our previous conclusions about transatlantic differences in antitrust and then offers some observations about how differences may diminish or persist in the future. The chapter concludes with implications for transatlantic cooperation in competition policy. Recent years have seen EU competition policy move markedly away from “form-based” toward “effects-based” application. At the same time, in some respects, US policy has become more noninterventionist, and although the Obama administration’s rhetoric seemed to signal some reversal, rather little change was evident in the first term. Thus while some convergence can be discerned, the picture is complex, and there are substantial remaining differences. They were

chapter ten

198

examined in detail in the previous chapters and will only be briefly summarized here. Merger Policy If this book had been written a dozen years ago, differing merger standards would have loomed as an important cross-Atlantic difference and source of tension. The apparent EU treatment of efficiencies in the GE/ Honeywell case as a negative rather than a positive factor in evaluation was widely criticized. Further, there were many irregularities in the pursuit of the case, which led the authors of one study to conclude that “the case team was the victim of a self-confirming bias or possibly may have pursued a different objective from that assigned by the merger regulation.”2 And the controversy came after several defeats of the Commission in the courts. All of this generated a complete overhaul of merger procedures. As noted in chapter 3, however, the nonhorizontal guidelines develop a number of scenarios for possible enforcement activity that could lead to differences with the United States. In this area, as in others, some observers have argued that the United States lags Europe by not taking a sufficient amount of “post-Chicago” learning on board.3 In general “post Chicago” analyses tend to tilt somewhat in the direction of rivalry protection, and this is the main source of tension between the United States and the European Union overall. This difference between Chicago and post-Chicago antitrust analysis and its impact on current differences between US and EU policies is further examined below. There have been no important transatlantic disputes about mergers since the GE/ Honeywell case.4 They may well arise again, but the great similarity between the EU Merger Guidelines and the 2010 iteration of their US counterparts suggest a high level of similarity in evaluating horizontal mergers. It remains to be seen if the European Union’s more expansive view of nonhorizontal mergers expressed in its 2008 guidelines leads to greater stringency in enforcement. Price Discrimination A fundamental difference endures over price discrimination. In fact, much of transatlantic difference on vertical issues and exclusion can be traced to sharply different basic postures on price discrimination: the United States embraces price discrimination as a competitive tool unless it is found predatory under the US predatory pricing standards, while TFEU Article 102 casts doubt on the legitimacy of the very practice when employed by firms with a substantial market share. This difference drives differing emphases in ex-

a summing up

199

clusive dealing and manifests itself strongly in matters involving loyalty and bundled discounts as well as in predatory pricing. At the time of the original Treaty of Rome, both jurisdictions viewed price discrimination with suspicion. In the years since, a large gap has emerged. As economic thinking increasingly emphasized price discrimination as an important means of price competition, US courts and enforcement agencies returned to the major concern underlying the original language of Section 2 of the Clayton Act: predatory pricing. Price-cutting against direct (“primary-line”) competitors is now evaluated using the same standards as Section 2 of the Sherman Act, while Robinson-Patman concerns about differential advantages across competing purchaser-resellers has largely disappeared from enforcement (although some private action persists). The contrast with Europe is stark. The original language of the Clayton Act forbids price discrimination only when it serves “to lessen competition or tend to create a monopoly” and could therefore accommodate an evolving understanding that price discrimination typically increases competition instead of lessening it. The additional language that Congress added in the Robinson-Patman Amendments is more confining and underlay the FTC’s broad challenges to price discrimination from the 1940s through the 1960s. Yet even the Robinson-Patman language is cast in terms of the effect of the discrimination on competition and thus provides an opening for the courts to construe its prohibitions narrowly. Thus the Robinson-Patman Act no longer plays a major role in US antitrust law: the government no longer enforces it, and the courts have contained its impact by construing it narrowly. In short, the US authorities treat the Robinson-Patman Act as an aberration that reflects the understandings and values of a bygone age and that conflicts with the dominant welfare-increasing policy of the Sherman Act. In sharp contrast, the EU authorities appear confident that the suppression of price discrimination furthers the public good. In this they are confirmed by the provisions of Article 102(c), which are very explicit about the immediate impact of price discrimination on customers: dominant sellers may not apply “dissimilar conditions to equivalent transactions with other trading parties, thereby placing them at a competitive disadvantage.” The focus of this language is on protecting customer-resellers; the language comes close to a pejorative description of price discrimination, and it seems to allow for exoneration only through a finding that transactions are not “equivalent.” Moreover, the courts carry out this hostility toward price discrimination in other ways that are not compelled by any language in the Treaty. The Court of Justice has applied Article 102(c) against discrimination whose effects primarily impact the rivals of the discriminating seller (primary-line effects) despite the focus of that provision on pro-

chapter ten

200

tecting customers (secondary-line effects), and the ECJ has also used the Article 102 general provisions against discrimination generating primaryline effects. Predatory Pricing The Areeda-Turner rule, first presented in 1975, has become the focal point of US policy toward predatory pricing. The rule originally highlighted marginal cost as a bright line because it marked the pricing level below which a firm would not even be covering its out-of-pocket costs. But because even pricing below marginal cost or its more available proxy, average variable cost, was known to occur for many innocuous reasons, they believed it was necessary to establish a probability of recoupment to warrant condemnation as predatory. Almost all of the US lower federal courts have accepted some variant of the Areeda-Turner approach, including the use of average variable cost as a “surrogate” for marginal cost. Subsequent scholarship has suggested average avoidable cost as a superior index of the same basic idea, and the Justice Department and some lower courts have followed suit. The Supreme Court has required for predation that the seller’s prices fall below otherwise undefined “incremental cost” and that a probability of recoupment be shown. The situation in the European Union is very different. Average avoidable cost has been accepted as a standard below which predatory pricing is simply assumed without the need to show probable recoupment. Furthermore, pricing falling in the range between AAC and average total cost is regarded as problematic (as has been the case in some US circuits).5 While the EU guidance implies that a probability of recoupment will accompany most predation cases, the Court of Justice has pronounced otherwise. The US Supreme Court has repeatedly insisted that it is unwilling to treat above-cost pricing as predatory, and, although it has refused to define “the relevant measure of cost” for predation purposes, it has also repeatedly referred to the relevant cost as “incremental cost.” The most expansive approach to the price-cost standard was probably in the FTC’s Intel case, where the FTC complaint used as the relevant cost “average variable cost plus an appropriate level of contribution towards sunk costs.” This measure of cost is similar to what this book has termed long-run average incremental cost, LAIC. A transatlantic reconciliation between the currently very different positions might be achieved if the United States were to recognize more explicitly the possibility of competition-impairing price predation above AAC, and if the EU linked all predatory pricing attacks to enduring impacts

a summing up

201

on competition in the affected market. Such reconciliation is not an immediate prospect. Exclusive Dealing Most exclusive dealing in the European Union by firms possessing a market share of less than 30 percent has been block-exempt under Article 101 since 1999. Larger, dominant firms are subject to attack under Article 102, because such exclusive arrangements are considered abuses under Article 102’s general clause and because they almost invariably grant at least a de facto discount for exclusivity and thus involve price discrimination that could trigger the prohibition of Article 102(c). In the United States, the historical focus on exclusion has been directed largely to the share of the total market covered by exclusive-supply contracts, but the practical result is largely similar.6 In recent years only firms with a share above 40– 50 percent are attacked under Section 3 of the Clayton Act or Section 1 of the Sherman Act and even larger market shares are generally required under Section 2. Although the prevalent US position is that exclusive-supply contracts are recognized as generally efficiency enhancing, their foreclosure effects become more important as the share of the market subject to exclusive arrangements increases. Single-Product Discounts US courts have generally found loyalty discounts to be procompetitive. To test anticompetitive effect, the predatory pricing approach is most often used: was the price below an appropriate measure of cost, typically some version of average variable or escapable cost, and was recoupment likely? Because the effects of loyalty discounts are often similar to exclusive-supply contracts, it would be expected that when loyalty discounts generate de facto exclusive-supply relationships, they would run afoul of the substantial-share limitation (at around 40– 50 percent of the market) applicable to exclusivesupply contracts in the United States. At these market shares, the supplier would effectively bear the burden of proving the procompetitive effects of the arrangement. The FTC’s proceeding against Intel should be seen in this light. The FTC’s complaint drew upon the predatory pricing model just described, was cast in terms of the price-cost relation, and used an unusual definition of cost similar to LAIC. Because that case was settled, and the settlement employed a cost definition closer to the more common average variable cost, we do not know how the FTC’s original approach would have fared on judicial review.

chapter ten

202

The prevailing approach to loyalty discounts in Europe differs in a number of ways from the US approach. Single-product discounts by dominant firms are seen as abuses either because they injure directly competing firms or competing resellers. Jurisprudence on Article 102 has accepted average avoidable cost as a level below which a dominant firm’s effective price may not drop and has developed an elaborate model of how it can be applied in situations where part of the dominant’s sales are not “contestable.” Unfortunately, the model is fraught with ambiguity. Moreover, the ECJ has rejected its application as unnecessary, since the case law treats exclusivesupply contracts by dominant firms as abuses. A fortiori the ECJ has not seriously considered competitive effects in deciding such cases. Many commentators have interpreted the recent US Intel case as a movement in the direction of the European Union. Although the actual settlement reads differently, perhaps due to bargaining, the original complaint attempted to use LAIC instead of AAC as a cost standard. The use of LAIC might be seen as an innovation that would radically change the US approach to predatory pricing in dynamic industries, which typically have huge sunk cost and sometimes negligible avoidable cost. Many experienced observers, however, believe that Intel would have prevailed had it chosen to contest the complaint. At all events, the case raises new questions about what, if any, price standards make sense for such industries or whether violations of Section 2 should be grounded in other considerations. Bundled Discounts Like single-product discounts, bundled discounts are generally tolerated in the United States as procompetitive price reductions. The major recent exception was the Third Circuit’s 3M decision that treated these discounts as monopolization. The 3M case, however, has been widely criticized and explicitly rejected by the Antitrust Modernization Commission (AMC) and the Ninth Circuit. Bundled discounts become problematic from an antitrust standpoint when they exclude one or more rivals from the market, and the exclusion significantly alters the market structure, reducing the level of competition. In such cases, courts would be called on to determine whether the social benefit of the low prices is offset by the removal of the rivals from the market. Again bundled discounts are viewed differently in the European Union. Bundled discounts are usually geared to the specific circumstances of specific buyers and thus reflect the same opaque combination of third-degree price discrimination and bilateral-monopoly bargaining that raise red flags

a summing up

203

in Europe about exclusive-supply contracts— and typically single-product discounting as well— because different resellers are treated differently. Various attribution schemes for dealing with “monopoly” products sold in bundles that might lead to exclusion have been proposed in both the United States and the European Union. Their application, however, is bedeviled by the lack of clarity between “monopoly” and “competitive” products in real situations as well as the conceptual weaknesses of any single price standard. Thus, for example, the court in the 3M case treated 3M’s large market share in Scotch tape as a monopoly and 3M’s discounts as monopolization. Yet 3M’s discounts were applicable to a wide range of products, and it seems inappropriate to treat those discounts as monopolization when they applied to heterogenous bundles in which the constituent products varied widely in strength of brand preference. Intellectual Property The interface between intellectual property law— particularly that pertaining to patents and copyrights— and competition policy has a complex history in the United States. In recent years, however, cases involving patents and copyrights have been decided in ways that give a high degree of deference to the intellectual property holder. These cases recognize the right of patentees and copyright holders to refuse to deal with aftermarket rivals who needed the covered product in order to compete. This fits with the long tradition in US law of allowing maximum freedom to firms in choosing their own trading partners. The older view that intellectual property laws conflicted with the antitrust laws has given way to a reconciliation that recognizes both sets of laws as contributing to general economic betterment. The recent EU experience has been very different because enforcement there has increasingly found bases for requiring intellectual property holders to share their intellectual property with competitors. Several cases over the last twenty years have revealed a pattern of treating intellectual property as an “essential facility”— a doctrine first developed in the United States in 1909 to justify the forced sharing of railroad infrastructure. Although the doctrine originated in the US Supreme Court, that court has never explicitly accepted it as a general principle. In sharp contrast, the Court of Justice has proclaimed variations on several conditions that can be used to justify a condemnation of a dominant firm for failing to share intellectual property with other firms. These include the essentiality of the intellectual property for the marketing of a new product, the absence of business justification for denial, and the elimination of all competition in a market. In the EU Microsoft case,

chapter ten

204

this menu appeared to extend to the facilitation of an increased variety of offerings of an already existing product. Hence, there now appear no obvious standards in the EU case law for limiting the scope for compulsory sharing of intellectual property. Dynamic Industries The antitrust treatment of dynamic knowledge-based industries (sometimes called industries of the “new economy”) is evolving differently in the two jurisdictions. These industries rely heavily on intellectual property and sometimes involve winner-take-all competition for the market. Differences between the two jurisdictions in their approaches to intellectual property necessarily affect the broader issue of how these knowledge-based industries fare under competition law. The different approaches of the United States and the European Union to platform software provide a particular example. In the US Microsoft case, the DC Court of Appeals created an exception to the per se rule governing tying arrangements for platform software in order to foster innovation. By contrast, in the EU Microsoft case, the Commission and the General Court imposed a forced sharing obligation upon Microsoft with respect to the communications protocols used to communicate between servers. This reduced the incentives of Microsoft, which had developed an advance in server communications (as well as the incentives of other firms to make fundamental innovations), whereas in the US case the preexisting law was modified in order to foster innovation. The EU approach thus favors follow on competition over the US emphasis on initial innovation. Prior to the DC Circuit’s decision in Microsoft, some observers wondered7 if US antitrust law could develop sufficient flexibility to permit knowledge-based industries like Microsoft to flourish. In particular, would the free distribution of Microsoft’s browser be treated as predatory pricing and would the integration of the browser into the operating system be condemned as an illegal tie? In fact, the courts construed the antitrust laws in ways favorable to these knowledge-based industries. Moreover, the network effects the Windows operating system generated provided Microsoft with the protection necessary for a knowledge-based product to recoup its investment in research and development. Thus in this instance network effects fostered the development of knowledge-based products (like Windows) by bolstering copyright and trade-secret protection to a level of protection analogous to what patents provide. There is no consensus in the United States about how the antitrust laws should apply to these industries, but leading antitrust theoreticians, judges, and officials have recognized their unique characteristics. In its Microsoft decision, the European Commission expressly rejected

a summing up

205

the contention that dynamic knowledge-based industries should be treated differently from traditional industries under EU competition law.8 There has been no European analogue to the DC Circuit’s decision creating an exemption from the per se rule on tying for platform software. Moreover, there is an expanding essential facilities doctrine in the European Union that has been increasingly used to compel companies to share their intellectual property with their competitors, thus shrinking the incentives to engage in so-called leapfrog innovation, the kind of innovation associated with knowledge-based industries. On the surface, it appears that the two jurisdictions are on different tracks with respect to knowledge-based industries. The EU treatment of network effects as an impediment to competition that should be opposed wherever feasible does not necessarily suggest a neglect of dynamic efficiency concerns because the prevention of networkprotected bastions could in many circumstances actually accelerate technological change, as the antitrust-required dissolution of the old AT&T revealed. But can competition policy be generally used to effectively control such developments? The United States seems to have rejected the option of attempting to manage network effects with interventionist policy; it appears to accept any market-generated network advantages so long as they do not involve monopolization or attempted monopolization. The European Union’s attempt to mold the development of the server software market presents an instance of a sharp contrast in policy that shows no signs of abating. Explaining the Differences How can all of these differences be explained? First, there is an array of cultural and legal differences that we noted in the first chapter. Historical differences in ideology and doctrine, in institutions, and in interests have often produced different legal outcomes in quite similar competitive circumstances. These factors persist, and they continue to influence substantive policies, the choices of proof standards, the confidence of officials in predicting the future, and their choices of methodology. Second, a major doctrinal difference continues over the very goals of competition policy. To what extent does (or should) competition policy take a pure efficiency approach to its underlying goal? A pure efficiency approach would seek the maximization of aggregate social welfare. Is the goal of competition policy the total economic welfare of society or the welfare of consumers or something else? Writing in 1978, Robert Bork, in his influential Antitrust Paradox, advocated the maximization of “consumer welfare” as the basic and only goal of antitrust law. Yet Bork defined “consumer welfare” as what economists call “total welfare.” This usage confused many,

chapter ten

206

and some US court cases are murky. For decades the courts have been using the language of consumer welfare, equating or otherwise associating it with efficiency and often referencing Bork in the process. In these ambiguous circumstances, developing a persuasive argument that the US policy is substantially open to the total welfare standard is possible, but it would be an uphill fight. Previous chapters of this book have identified flashes of ambiguity in US law and policy on this issue but little more. For example, the 2010 US Merger Guidelines, in footnote 15, states: “The Agencies also may consider the effects of cognizable efficiencies with no short-term, direct effect on prices in the relevant market.” This means that the guidelines do not insist on an immediate price decrease, but they do not explicitly countenance an increase either. On the other hand, the general acceptance of the consumer welfare and efficiency language means that there is widespread agreement that adherence to ideas of rivalry or fairness within the US system seems to be slight and probably declining. The situation in Europe is very different. The total welfare standard is explicitly rejected by the language of Article 101 that insists on consumers receiving a share of any resulting benefit from the transaction in question.9 The Merger Regulation similarly requires that the efficiencies or other value generated by a merger be “to consumers’ advantage.” Consumer welfare is repeatedly invoked as underlying the administration of EU competition law. The guidance on Article 82, for example, invokes consumer welfare fourteen times. Philip Lowe, the European Commission’s former director general for competition, has spoken eloquently of the virtues of the consumer welfare approach to competition law and appears to draw no distinction between that approach and “efficiency.”10 But the basic EU law also makes adherence to an economist’s conception of a consumer welfare standard very difficult or impossible. In particular, concern for the disadvantaging of some resellers blunts price competition and virtually assures that consumer surplus will not be maximized. Price discrimination looms large in real world price competition, and its generally beneficial effects are not confined to “nondominant” firms. Therefore, the Commission’s rhetorical embrace of a consumer welfare standard appears to conflict not just with previous legal precedent but with the very language of the law. The Intel case further suggests the ambiguity of current standards in the European Union. As documented in chapter 7, the Commission appeared to regard the absence of AMD offerings in sufficient profusion as a problem independent of that firm’s viability— or even its sales to most major manufacturers. And this should not surprise. The Commission’s guidance explains its concern with consumer welfare “whether in the form of higher

a summing up

207

price levels than would have otherwise prevailed or in some other form such as limiting quality or reducing consumer choice.” It appears that no one is quite sure just what this broad goal, sometimes called a “consumer sovereignty standard,” means,11 but it clearly goes beyond the consumer welfare standard as generally understood. All competition in other than atomistically competitive markets can reduce consumer choice, and the EU standard as stated does not limit policy intervention in any clear way. The current EU standard has doubtless been construed with an emphasis on variety in part because that provides a nominally consumer-oriented basis for protecting firms that might otherwise be driven from the field by stronger competitors. In short, a de facto rivalry standard continues to contest a consumer surplus standard for primacy in the European Union. There are, of course, minority US voices that also effectively give support to a rivalry standard. One such American view suggests that FTC Section 5 could be used as a “gap filler” when conduct would not be susceptible to Section 2 enforcement but may injure consumers, through, for example, reduction of consumer choice.12 Similarly, the chairman of the American Antitrust Institute declared in 2008 that the vagueness of “abuse” in the European Union and “fairness” under Section 5 of the FTC Act was really a strength because “they are not restricted to a narrow efficiency based meaning.”13 In particular, the Institute envisions a convergence between the United States and the European Union largely by construing Section 5 in ways that credit “unfairness” to market participants other than the final customer.14 We would regard this type of convergence as a retrograde development; it falls outside what the antitrust scholar and former FTC commissioner William Kovacic calls the “double helix” of Chicago and Harvard, where the former has driven the substantive emphasis on efficiency and the latter an emphasis on administrable rules.15 Moreover, we think any attempt by the FTC to move US antitrust substantially in such directions would soon be thwarted by the courts. Third, some of the real or apparent differences between antitrust approaches in the United States and the European Union grow out of unresolved internal ambiguity or disagreement within each of the two jurisdictions. Both jurisdictions ban predatory pricing, and their standards for determining what pricing should be deemed predatory, although broadly similar, differ somewhat between the two jurisdictions. Within each jurisdiction the “as efficient competitor” is employed as a tool in determining whether the pricing in question is to be treated as predatory. But within each of the two jurisdictions, there is confusion or conflict about the role of the “as efficient competitor” test in evaluating behavior that does not fall within the scope of the rules prohibiting predatory pricing.

chapter ten

208

In the United States, the initial step in evaluating pricing for predation is determining whether the alleged predator has been selling its goods at prices that are below its own costs. If it has sold its goods at prices that exceed its own costs, then it passes the as efficient competitor test since if the firm’s prices exceed its own costs, they necessarily exceed the costs of a competitor that is as efficient as itself. An unresolved question in US antitrust law involves the circumstances under which a firm whose prices exceed its costs (and thus those of an equally efficient competitor) can nonetheless violate the antitrust laws. The Third Circuit takes the position that even though a firm passes the equally efficient competitor test, its conduct still must pass a Rule-of-Reason evaluation. The Antitrust Modernization Commission has attempted to incorporate predatory pricing standards into the evaluation of bundled discounting, by adopting a “discount attribution rule.” The discount attribution rule is designed to deal with a simple situation in which a firm is selling a bundle of goods and in which one product in the bundle is competing with a product of a rival. In these circumstances, the aggregate discount on the bundle is attributed to the competitive product. If, after that attribution, the seller’s prices on the competitive product exceed its costs, the firm is deemed to have been acting lawfully. But if, after the attribution, the firm’s price on the competitive product is less than its cost, then its behavior is suspect and further analysis is required. Again, there is tension between the “substantial share” test that nominally governs the lawfulness of exclusive-supply arrangements and the as efficient competitor test. The substantial share test is meant to prevent a firm from foreclosing distribution channels to competitors. But if the rivals can compete for the distribution channels, then perhaps the appropriate test is not substantial share but the as efficient competitor test. In the European Union the same tension is present. The Commission has incorporated the equally efficient competitor into its guidance on Article 102. Yet Article 102 itself condemns price discrimination and tying as abuses, and it is unclear how much leeway the courts and the Commission possess to reshape these provisions through interpretation. Currently, the courts apply per se– like rules keyed to an array of behaviors, regardless of the relation of prices to cost and thus in apparent disregard of the equally efficient competitor test. And despite the guidance, the Commission appears also to assess the lawfulness of a range of behaviors apart from the relation of prices to cost. Thus in both the United States and the European Union, there is an ongoing struggle over the role of the as efficient competitor test. Some apparent differences between the two jurisdictions may reflect unresolved internal jurisdictional differences, while others may mark an Atlantic divide.

a summing up

209

Fourth, at one time the EU was seen as institutionally split on the absorption of economic analysis into its competition law cases. The Commission was said to have accepted economic analysis, while the courts had not. The Commission’s greater receptivity to economics was referred to as effects-based analysis, where the effects were understood to be economic effects. The courts, by contrast, took an approach that tended to apply per se or presumptively illegal characterization to categories of behavior. This approach was usually described as form-based analysis. The story was that the Commission had been allowed to employ an effects-based analysis in proceedings involving Article 101 and the Merger Regulation, but that the courts were thwarting the application of effects-based analyses to Article 102 behaviors. Many observers believed that if and when the courts accepted effects-based analyses under Article 102, the stage would be set for a rapprochement with US antitrust law, which had taken a Chicago School economic approach since the 1970s. The trouble with this view is that there is more than one economic approach. Indeed, we can get a glimpse of different economic approaches within the Commission staff from comparing its discussion paper on Article 82 with the discussion paper on predation, the latter issued under the auspices of the Commission’s chief economist. There is no question that the Commission staff is well versed in the economics literature and that this literature is the same as that available to its American counterparts. Indeed, the bibliography at the end of the discussion paper on predation is composed almost exclusively of American journals. Nonetheless, these papers not only differ significantly in their policy recommendations from US approaches, but also the policy recommendations in the predation paper differ from those in the Article 82 paper. We have identified several of these intraCommission staff differences in chapter 5. Fifth, some differences in approach appear to be keyed to the respective institutional enforcement structures. US courts and enforcement agencies must constantly consider the impact of their actions on private litigation. Hence, even if thinking about the substance of policy were exactly the same on both sides of the ocean, there would be more US concern about the contagion of type one errors: condemning activity that is in fact innocuous or welfare enhancing.16 The FTC noted as an apparent strength that its successful attack on Intel under FTC Act Section 5 produced no impact on the prevailing standards of Sherman or Clayton Act cases and thus no effect on private actions.17 The Supreme Court has been especially concerned with the avoidance of type one errors, since the purpose of the antitrust laws is to foster efficiencyenhancing conduct. This concern dates back at least to 1986 when in its

chapter ten

210

Matsushita decision it remarked that mistaken inferences and resulting false condemnations “are especially costly, because they chill the very conduct the antitrust laws are designed to protect.”18 More recently the Court has indicated that the danger of such type one errors is a reason for declining to expand the scope of Section 2 liability.19 Indeed, the Court believes that some anticompetitive conduct may be “beyond the practical ability of a judicial tribunal to control.”20 By contrast, European authorities do not appear to be as concerned with the welfare-reducing potential of type one errors. This may partially be explainable by the lesser role of private-damage antitrust actions in Europe. The Commission in 2013 proposed a directive to expedite such suits, but that initiative suggests a huge Atlantic divide that will persist. In the explanatory memorandum that precedes the text, the Commission emphasizes that divide. “The Commission’s proposal is aimed at bringing compensation to those who suffered harm and not, as in some other jurisdictions, as a tool for punishment and deterrence of those who breach the antitrust rules.” Compensation is limited to actual damages and is expected to flow from breeches of conduct found by competition authorities. The directive aims principally to increase the clarity and consistency of procedures national authorities use to provide compensatory awards.21 Because the European Commission controls antitrust enforcement against major companies, the Commission may believe that by judiciously selecting the cases it brings, it can avoid type one errors. But the European guidelines and guidances that seem to be more theoretically complete than their US counterparts can potentially deter efficiency-enhancing behavior as well. Sixth, the administration of the US antitrust laws by the Department of Justice and their application by the courts tends to follow the Chicago School approach: the formulation and enforcement of rules derived from economic analysis. These rules are intended to be simple enough for business firms to self-apply them and for ordinary generalist judges to administer. As previously noted, the US Supreme Court is acutely aware of the potential for antitrust laws to be misapplied by officials and judges, thereby deterring socially beneficial behavior. The Justice Department has expressed similar concerns. The concerns just described are a large part of the reason the Justice Department and the courts have not embraced so-called post-Chicago antitrust analyses. Post-Chicago analysis tends to identify circumstances in which Chicago-based rules of analysis or actual judicially formulated rules of behavior are open to exploitation by a business firm bent on achieving an

a summing up

211

anticompetitive goal. Such analysis has shown how— despite the Chicagobased single-monopoly-profit theorem— a dominant firm could use tying to prevent entry of a rival into the market for a tied product.22 Similarly, a range of post-Chicago literature shows that above marginal cost pricing is capable of being used predatorily. The reason the US courts and enforcement agencies have not refashioned antitrust rules in the light of the theoretical insights achieved by post-Chicago theorists is not because these insights are wrong, but because of their belief that these insights cannot be administered without deterring more socially beneficial behavior than socially harmful behavior. Some creative attempts have been made to counter such skepticism,23 but their impact on the US competition policy system remains to be seen. As observed above, the Commission’s discussion paper on predation is based entirely on American economic journals, but the theoretical insights in those journal articles include substantial amounts of post-Chicago analysis. We believe the embrace of post-Chicago analysis by the Commission’s economists explains much of the differences between the antitrust stances of the two jurisdictions. Restated, many differences between the jurisdictions derive from their respective stances toward the potentialities for type one errors to infect their antitrust administration. Seventh, US officials charged with enforcing the antitrust laws act within a short time horizon, believing that it is impossible to predict the likelihood of economic events beyond two or three years. This reluctance to predict events far into the future blends into their reluctance to predict the concurrence of multiple events that would generate, in combination, an anticompetitive result. In an OECD roundtable reviewing the issues raised in the GE/ Honeywell merger case, the US Department of Justice faulted the European Commission for its willingness to make long-term predictions involving the concurrence of multiple events.24 Insofar as the European Commission has greater confidence in its predictive powers than US enforcement authorities, that would help to explain some of the differences in approach between the jurisdictions, including their different approaches in the GE/ Honeywell merger case. Eighth, it is possible that some of the differences in approach are the result of the relative insulation or exposure of the administering institutions to political concerns. When Barack Obama campaigned for the presidency in 2008, he attacked the Bush administration as having “one of the weakest records of antitrust enforcement in the last half century.”25 And the shift in DOJ policy by Obama appointees— both rhetorically and in the abrupt withdrawal of the recently completed study of Section 2— represented a

chapter ten

212

sharp break from one American administration to another,26 perhaps paralleling the Clinton administration’s revocation of the guidelines on vertical arrangements promulgated under the Reagan administration.27 Obama’s first deputy attorney general, Christine Varney, attracted great attention with her observation that “there is no such thing as a false positive,”28 bringing her into policy conflict with the Supreme Court’s strictures against type one errors. These acts— the revocation of the Section 2 report and the repudiation of the vertical guidelines and Varney’s rhetoric— reflect changing policies in antitrust administration injected as a result of political choices the electorate made. Yet these politically induced policy changes in antitrust tend not to play highly visible roles in the United States, where the rules of substantive law are mostly generated by the courts (which are insulated against political influence by sets of legal and cultural controls) and where private enforcement tends to dominate litigation. Even government enforcement appears relatively continuous from one administration to another.29 In fact, substantial departures from current practice by the enforcement agencies could attract the attention of Congress. For example, if mergers began to be approved that contemplated with equanimity likely price rises followed by a decline below the original level more than a couple of years later, a serious political backlash could ensue.30 The requirement that all EU commissioners approve DG Comp action— a development that resulted from political differences over the Aérospatiale merger in 1991— helps to enable political considerations being brought into play as the European Commission makes its enforcement decisions. The extent to which political factors underlie antitrust differences between the jurisdictions, however, is difficult to determine. Some political constraints in Europe on antitrust policy developments appear significant. As the first chapter explained, the strength of political views suspicious of markets and competition is stronger almost everywhere in the European Union than in the United States. Moreover, the uniting of Europe has already seen a vast and disruptive restructuring of national economies in myriad dimensions. Germany, in particular, has seen its Mittelstand of historically closely held, often family-owned, firms under siege from multinationals based both inside and outside the European Union. Part of the European left would like to see competition policy made more responsive to immediate government influence.31 This development is very unlikely, but rapid EU convergence to a perspective similar to that of the United States, in most of the areas documented above, lies far in future, if it occurs at all. Indeed, as some on the interventionist side of the US debate foresee, the United States might in some areas move in the EU direction. The US Intel case represents that possibility.

a summing up

213

Implications for Cooperation across the Atlantic and Beyond This book has concerned similarities and differences in transatlantic competition policy. These features have critical implications for transatlantic cooperation. Perhaps the most important cooperative issue in contemporary antitrust has not been treated in this book: the control of cartels. This lack of attention stems directly from cartel illegality in the United States and the European Union— and in virtually every other of the more than 120 jurisdictions that now have competition laws. A major barrier to cooperation in this area of policy turns on penalties; while they are rising in many jurisdictions, only the United States among the major states regularly puts violators in jail.32 This has led European business to be cautious about some elements of cooperation with Washington,33 but the overall level of cooperation across the Atlantic is higher and more effective than ever before.34 Many jurisdictions, including the European Union, have followed the US practice of offering leniency to the first (and only the first) member of a cartel to offer assistance to the competition authority in prosecuting a cartel.35 The current apparent consistency of declared US and EU policy toward mergers is critically important for cooperation because extraterritoriality propels merger policy immediately into the realm of international relations more forcefully than most other elements of competition policy. Shortly after the European Union introduced merger review in 1987, the United States and the European Union agreed to close cooperation on mergers. This has led to ever-closer cooperation. There is now typically a high level of coordination in simultaneous reviews.36 Moreover, while scholarship suggests that EU merger control activities sometimes favored EU firms until the new century, there is no indication of such favoritism since.37 Sometimes a merger that includes at least one foreign firm creates a purely local problem in which especially tailored remedies by the affected state are typically accepted by the home country without resistance, even though the expected profit stream accruing to the firm of its nationality may be reduced. Similarly, the merger of two multinational firms may create competitive problems in two or more jurisdictions, and little conflict is typically generated by national adjustments of the merger to avoid competitive problems. For example, the merger of Kimberly-Clark and Scott paper was approved in both the United States and the European Union in 1995 with separate provisions for limiting market power.38 Such actions involve ownership and control issues involving assets in the territory of the state making the stipulations. But a state may also attempt to block a completely foreign merger that portends important negative effects in its market, even though

chapter ten

214

no substantial activity by the merging firms takes place in the affected state. This was the case of the Boeing/ McDonnell-Douglas and the GE/ Honeywell mergers. The United States and the European Union make quite similar claims on the basis of the “effects doctrine” that justifies a voice in such completely foreign mergers having a substantial impact on the home market.39 In such a case, the more restrictive jurisdiction’s views will generally prevail.40 By extension, in a situation in which a merger will affect the entire global market, the objections of any one of a number of substantial jurisdictions may sink it, accommodations may be made to gain universal approval, or the merger may take place in the face of opposition from one or more governments whose threats or promises involve markets that can be ignored. Assuming that countries embrace some version of the effects doctrine, what practical recourse do they presently have if a merger takes place despite their opposition? In the celebrated Boeing/McDonnell-Douglas merger of 1997, the European Union was threatening to fine the companies for failing to comply. This could have been up to 10 percent of the annual revenue or daily fines of up to 100,000 European Currency Units (the predecessor to the euro).41 The companies never rejected the authority of the Commission, nor did they suggest that they would not pay the fines.42 Europe was a large market for Boeing, and recalcitrance about the fines could have resulted in more severe damage if the European Union had moved to block its carriers from using Boeing planes. Refusing to buy from a dominant firm typically hurts the purchaser more than it hurts the seller. But in this case, it would have provided a strong boost for Airbus, the outcome that the United States assumed to motivate the Europeans in the first place.43 It must be stressed, however, that EU objections were based on an opposition to increasing Boeing’s dominance in the large civilian airliner market— about two-thirds at the time, although it had been losing share to Airbus in most of the previous decade— and other motivations cannot be proved. An important inference from this episode, however, is that the objections of governments in which sales of the merging firms are modest and there are only minor assets to attach in retaliation for an unapproved merger might simply be ignored. Reactions to US-EU merger disputes have diverged markedly. Some believe an outcome such as that achieved in the Boeing case shows that the present system works tolerably well.44 But following that case and the subsequent GE/ Honeywell dispute, one authority noted that “the agencies redoubled their efforts to increase convergence, including through working groups on both substantive and procedural aspects of merger review.”45 As

a summing up

215

observed in chapter 3, however, it is not entirely clear that the portfolioeffects issues involved in that case have been resolved. The dynamic noted for merger control— the most interventionist policy of a major economic actor will control global outcomes— can also be seen in the currently vexed area of single firm conduct. It comes through clearly in EU interventions loosening the proprietary control over intellectual property. Once a firm has complied with a directive to share, the home government of the firm obliged to curb its property rights is presented with something of a fait accompli. Its only effective recourse would be to attempt to block imports involving the disputed intellectual property or to take a complaint to the World Trade Organization on grounds that the intervening country violated the Trade-Related Aspects of Intellectual Property Rights (TRIPS) agreement.46 A World Competition Organization? Such disputes as Boeing/McDonnell-Douglas and GE/ Honeywell led some to press for an international agency with powers over all states— or a subset of the “like-minded”— that would function within the WTO.47 Merger control would be part of its agenda. But this kind of development, if it ever occurs, will take place far in the future. Recent history suggests how great the obstacles are. OECD competition policy discussions over several decades have made little progress to bring congruence to various national competition policies. This stands in sharp contrast to the OECD’s experience in many other public policy areas and reflects the embeddedness of competition policy in national practices and legal structures of even those states with little competition policy experience.48 Elsewhere, in discussions preparatory to the Doha Round, the European Union proposed consideration of an exiguous agreement that would only have obliged states to adopt a competition policy that extended national treatment and banned hard-core cartels.49 But US experts feared any contamination of the US antitrust system, either by trade experts or by foreign competition authorities. As a response, the Clinton administration appointed an International Competition Policy Advisory Committee (ICPAC) of American experts that came out against a WTO role, advocating instead for a “virtual” organization to deal with cross-border competition issues. The result was a loosely structured and lightly funded International Competition Network (ICN)50 open to all interested states, which the European Union supported. Both powers valued the coordination and streamlining of national merger approval, a central substantive focus of the ICN.51

chapter ten

216

The ICN has been criticized as a creature of those who pay its modest costs, mainly public and private participants from the high income countries, but it probably represents the limit of current institutional feasibility in multilateral competition policy coordination. A Final Note Competition policy was born in the United States, and when the nascent European project elected to embrace a kindred policy more than a half century later, there was inevitably considerable borrowing and much indigenous content. By the time of the rapid global diffusion of antitrust policy in the 1990s, the EU approach proved more congenial than the American for most of the world. Nevertheless, lawyers and economists can find in virtually all national legislation ideas that are shared by the United States and the European Union. This book has argued that important differences remain between these two huge jurisdictions and that these are greater than often assumed. But the foundation for further convergence and increased cooperation is firmly in place.

Notes ChApter 1 1. Sherman Act of 1890, ch. 647, 26 Stat. 209 (codified as amended at 15 U.S.C. §§ 1– 7). 2. Treaty of Rome Establishing the European Economic Community, Mar. 25, 1957, 298 U.N.T.S. 11. Several treaties have amended the initial treaty. Unless time-context specific, all citations will accordingly be made to the current Consolidated Version of the Treaty on the Functioning of the European Union (TFEU). 3. Clayton Act ch. 323, § 7, 38 Stat. 730, 731– 32 (1914). 4. Cellar-Kefauver Act ch. 1184, § 7, 64 Stat. 1125 (1950) 15 U.S.C. § 18 (2006). 5. Council Regulation 4064/89, On the Control of Concentrations between Undertakings, 1989 O.J. (L. No. 395) (corrected version at 1989 O.J. (L 257) 13. 6. Council Regulation 139/2004, On the Control of Concentrations between Undertakings, 2004 O.J. (L. No. 24) 1. 7. Both jurisdictions treat horizontal price-fixing as per se illegal. On other per se categories, although the European categories are sometimes defined more broadly than their American analogue, this is sometimes a distinction without a difference, because the EU prohibitions extend only to business firms possessing dominant positions and thus to firms occupying market shares of, say, 40 percent or more. Thus in the case of tying arrangements, the US rule condemning tying arrangements conditions the prohibition on the seller possessing market power in the tying-product market, which means that the seller must hold more than a 30 percent market share. Thus the US rule on tying is similar to the more broadly phrased EU rule that applies only to firms holding

notes to chapter 1

218

dominant positions. In the case of exclusive-supply contracts (single branding, in the European terminology), a similar comparison could be made. Exclusive-supply contracts also fall into a category of behavior forbidden to firms holding dominant positions. So the European law appears similar to the US law that tends to bar such contracts when they apply to 40 or more percent of an industry’s output. The US law appears more flexible, however, and open to a showing that rival sellers, despite those contracts, do have effective access to buyers. Moreover, while the US Supreme Court has been narrowing or eliminating the categories of per se illegality, there has been less corresponding movement in Europe. 8. By competition policy we mean the practice of enforcing what in American parlance is called “the antitrust laws,” which, according to one prominent source, aim to “limit the market power exercised by firms and to limit how firms compete with each other.” D. W. Carlton and J. M. Perloff, Modern Industrial Organization, 4th ed. (Boston: Pearson/Addison Wesley, 2005), 631. Antitrust is a term also used in the European Union but typically with a narrower meaning than in the United States. Writing on the development of competition policy in Europe distinguishes antitrust policy from merger policy, probably because the latter came as a largely independent development in the European Union. See Tim Büthe, “The Politics of Competition and Institutional Change in European Union: The First Fifty Years,” in Making History: European Integration and Institutional Change at Fifty, ed. Sophie Meunier and Kathleen McNamara (Oxford: Oxford University Press, 2007). The distinction is also made in EU documents. When antitrust policy is used in this book, it is meant to refer to competition policy as defined above. And another distinction should be noted: competition policy as used here omits attention to nation-state aids and subsidies, a necessary concern in the European Union that lies beyond the scope of this work. 9. Some version of these and closely related categories appear in much writing about public policy differences and change. For examples, see Douglas Irwin and Randall Krozner, “Interests, Institutions, and Ideology in Securing Policy Change: The Republican Conversion to Trade Liberalization after Smoot Hawley,” Journal of Law and Economics 42 (1999); Richard Bird, Jorge Martinez-Vazquez, and Benno Torgler, “Tax Effort in Developing Countries and High Income Countries,” Economic Analysis and Policy 38, no. 1 (2008). In the strategic management literature, information available to the firm can replace ideology, which is considered with institutions, for example, David P. Barron, Business and Its Environment, 5th ed. (Upper Saddle River, NJ: Pearson Prentice Hall, 2006), 6– 10. 10. Louis Hartz, Liberalism in America (New York: Harcourt Brace, 1956). 11. Arthur M. Schlesinger Jr., The Politics of Hope (Princeton, NJ: Princeton University Press, 1957). 12. The Canadian anticompetition law of 1899, driven by many of the same forces that gave rise to the Sherman Act, receives much less attention partly because it proved so ineffective. Massimo Motta, Competition Policy: Theory and Practice (New York: Cambridge University Press, 2004), 14; Calvin Goldman, John D. Bodrug, and Mark A. A. Warner, “Canada,” in Global Competition Policy (Washington, DC: Institute for International Economics, 1997), 47– 85, at 50. 13. George J. Stigler, “The Origins of the Sherman Act,” Journal of Legal Studies (1985): 1– 12.

notes to chapter 1

219

14. Federal Trade Commission Act ch. 311, 38 Stat. 717 (1914). 15. See, for example, William E. Kovacic and Carl Shapiro, “Antitrust Policy: A Century of Economic and Legal Thinking,” Journal of Economic Perspectives 14, no. 1 (2001): 43– 60. 16. As late as 1990, there were fewer than thirty states with competition policies. That number had grown to over one hundred by 2009. Calculated from data in Mark R. A. Palim, “The Worldwide Growth of Competition Law: An Empirical Analysis,” Antitrust Bulletin 43, no. 1 (1998): 105– 45; Keith Hylton and Fei Deng, “Antitrust around the World,” Antitrust Law Journal 69 (2007): 469–526. 17. David S. Gerber, Global Competition: Law, Markets, and Globalization (Oxford: Oxford University Press, 2010). The Federal Trade Commission does firstround assessments with an administrative judge and the FTC itself, but all activity can be appealed to the court system. 18. David S. Evans, “Why Different Jurisdictions Do Not (and Should Not) Adopt the Same Antitrust Rules,” Chicago Journal of International Law 10 (2009): 4. 19. Standard Oil Co. v. United States, 221 U.S. 1, 66– 68 (1911); American Tobacco Co. v. United States, 221 U.S. 106, 179– 890 (1911). 20. Chicago Bd. of Trade v. United States, 246 U.S. 232, 238 (1918). 21. NIRA, ch. 90, 48 Stat. 195 (1933), declared unconstitutional in A.I.A. Schecter Poultry Corp. v. United States, 295 U.S. 495, 541– 51 (1935), and repealed by Act of Sept. 6, 1966, Pub. L. No. 89-554 § 8, 80 Stat. 648. See Donald R. Brand, Corporatism and the Rule of Law: A Study of the National Recovery Administration (Ithaca, NY: Cornell University Press, 1988). 22. Robinson-Patman Act ch. 592, 52 Stat. 1526 (1936). 23. United States v. Aluminum Co. of America, 148 F.2d 416 (2d Cir. 1945). 24. United States v. Griffith, 334 U.S. 100 (1948). 25. United States v. Grinnell Corp., 384 U.S. 563 (1966). 26. United States v. Columbia Steel Co., 334 U.S. 495 (1948). 27. Celler-Kefauver Act of 1950, 64 Stat. 1125 (1950), codified as amended at 15 U.S.C. § 18 (2012). 28. International Salt Co. v. United States, 332 U.S. 392 (1947). 29. United States v. E. I. du Pont de Nemours & Co., 353 U.S. 586 (1957). 30. United States v. Von’s Grocery Co., 384 U.S. 279 (1966). 31. United States v. Pabst Brewing Co., 384 U.S. 546 (1966). 32. Utah Pie Co. v. Continental Baking Co., 386 U.S. 685 (1967). 33. United States v. Arnold, Schwinn & Co., 388 U.S. 365 (1967). 34. See Carl Kaysen and Donald F. Turner, Antitrust Policy: An Economic and Legal Analysis (Cambridge, MA: Harvard University Press, 1959), 11– 15. 35. Richard Posner, Antitrust Law: An Economic Perspective (Chicago: University of Chicago Press, 1976); Robert H. Bork, The Antitrust Paradox: A Policy at War with Itself (New York: Free Press, 1978). 36. Joe Bain, the most influential writer in this tradition, did most of his work at the University of California– Berkeley, but his graduate training was at Harvard under Edward S. Mason. Joe S. Bain, “Workable Competition in Oligopoly,” American Economic Review (1950): 35– 47; Joe S. Bain, Barriers to New Competition (Cambridge, MA: Harvard University Press, 1956); Joe S. Bain, Industrial Organization (New York: Wiley, 1959).

notes to chapter 1

220

37. Continental T.V., Inc. v. GTE Sylvania, Inc., 429 U.S. 1059 (1977). 38. Compare Broadcast Music, Inc. v. Columbia Broadcasting System, 441 U.S. 1 (1979) with United States v. Socony-Vacuum Oil Co., 310 U.S. 150 1940); compare Northwest Wholesale Stationers, Inc. v. Pacific Stationary & Printing Co., 472 U.S. 284 (1985) with Klor’s v. Broadway-Hale Stores, Inc., 359 U.S. 207 (1959); compare Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007) with Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911); Compare Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209 (1993) with Utah Pie Co. v. Continental Baking Co., 386 U.S. 685 (1967). 39. Broadcast Music, Inc. v. Columbia Broadcasting System, Inc., 441 U.S. 1 (1979). 40. Jefferson Parish Hospital District No. 2 v. Hyde, 466 U.S. 2 (1984). 41. Northwest Wholesale Stationers, Inc. v. Pacific Stationery & Printing Co., 472 U.S. 284 (1985). 42. State Oil Co. v. Khan, 522 U.S. 3 (1997). 43. Leegin Creative Leather Products, Inc. v. PSKS, Inc., 551 U.S. 877 (2007). 44. U.S. Department of Justice and Federal Trade Commission, Horizontal Merger Guidelines, § 4 (2010), reprinted in 4 Trade Reg. Rep. (CCH) ¶ 13100. 45. Perhaps the most ingenious defense of cartels was offered in a British court decision: a savings on shopping costs generated by confidence that prices for a certain good would be the same everywhere! Motta, Competition Policy, 12. 46. World Trade Organization, World Trade Report (Geneva: WTO, 2007), 37– 44. 47. See, for example, H. Rep. No. 479, 66th Cong., 2d Sess. 2– 3 (1919) (describing policies pursued within the German chemical industry). 48. The characteristic socialist goal of nationalization of major industries, formally endorsed in 1925, was officially abandoned in 1959. 49. The Nazis more lightly controlled the southwestern city of Freiburg than most of Germany, but the ordoliberals still collaborated at great personal risk. Gerber, Global Competition, 167– 68. 50. The doctrine aimed to prevent periods such as Weimar as well as that of the Nazis; it therefore understandably had highly developed views about control of the money supply and other issues far removed from competition policy. Christian Ahlborn and Carsten Grave, “Walter Eucken and Ordoliberalism: An Introduction from a Consumer Welfare Perspective,” Competition Policy International 2, no. 2 (2006). 51. Ahlborn and Grave, “Walter Eucken and Ordoliberalism.” 52. The distinction between (perfectly) competitive “price-taking” firms and “price-making” firms with market power was coined by Tibor Scitovsky in his Welfare and Competition (Homewood, IL: Irwin, 1952). 53. David J. Gerber, Law and Competition in Twentieth Century Europe: Protecting Prometheus (Oxford: Clarendon Press, 1998), 241. There appears to have been tension among various adherents on the scope of social policy and hence on the size and scope of government. Christian Glossner, The Making of the German Post-War Economy (London: I. B. Tauris, 2010). 54. Ahlborn and Grave, “Walter Eucken and Ordoliberalism”; Gerber, Law and Competition, 7, 8, 239.

notes to chapter 1

221

55. Gerber, Global Competition, 169. 56. Gerber, Law and Competition, 173. 57. See Bain, “Workable Competition,” 35– 47; Bain, Barriers; Bain, Industrial Organization. 58. Treaty Establishing the European Economic Community, Mar. 25, 1957, 298 U.N.T.S. I-4300. 59. Section 1 of the Sherman Act condemns “every contract, combination, and conspiracy” in restraint of trade,” while Article 101 prohibits “all agreements by undertakings . . . which may affect trade between Member States and which have as their object or effect, the prevention restriction or distortion of competition within the internal market.” Section 2 of the Sherman Act attacks those who “monopolize or attempt to monopolize . . . any part of the trade or commerce of the several states or with foreign nations,” while Article 102 prohibits “any abuse by one or more undertakings of a dominant position within the internal market . . . insofar as it may affect trade among Member States.” 60. Ahlborn and Grave, “Walter Eucken and Ordoliberalism,” 207. 61. Id.; see also Alberto Pera, “Changing Views of Competition, Economic Analysis and EC Antitrust Law,” European Competition Journal 1 (2008): 127– 68. 62. Giorgio Monti, “EC and New Economy Markets, Article 82,” in Competition, Regulation, and the New Economy, ed. Cosmo Graham and Fiona Smith (Oxford: Hart Publishing, 2004), 40. 63. For example, Motta, Competition Policy, 22, 24. 64. United States v. Aluminum Co. of America, Inc., 148 F.2d 416, 429 (2d Cir. 1945). 65. 148 F.2d at 427, 429. 66. 148 F.2d at 429– 30. 67. Stephen Sosnick, “A Critique of Concepts of Workable Competition,” Quarterly Journal of Economics 72 (1958): 380– 423. 68. Together with Alcoa, Supreme Court decisions in Griffith in 1948 and Grinnell in 1966 are foundational for the modern US law of monopolization. 69. United States v. E. I. DuPont de Nemours & Co., 353 U.S. 586 (1957). 70. See Friedrich Kessler and Richard H. Stern, “Competition, Contract and Vertical Integration,” Yale Law Journal (1959): 1, 69. See also Northern Pac. Ry. v. United States, 356 U.S. 1, 4 (1958) (“The Sherman Act was designed to . . . [provide] an environment conductive to the preservation of our democratic political and social institutions”). 71. Brown Shoe Co. v. United States, 370 U.S. 294 (1962). 72. United States v. Von’s Grocery Co., 384 U.S. 279 (1966). 73. Kaysen and Turner, Antitrust Policy, 45. They framed their proposed approach to antitrust law as involving a limitation of market power, but would have made any market power limitation subject to the achievement of efficiency. 74. Council Regulation 4064/89, 1989 O.J. (L395/1) [1990]. 75. Consolidated Cases 56 and 58/64, Établissements Consten S.A. und Grundig-Verkaufs Gmbh v. EEC Commission (1966). 76. DG Competition is one of over thirty departments of the European Commission, the executive body of the European Union. The Commission’s competition department is headed by a director general and serves the Commission’s

notes to chapter 1

222

competition commissioner. DG Competition employs a large number of lawyers and economists to investigate and prepare competition cases. In addition to enforcing competition law in the private sector, DG Competition is also concerned with matters relating to state aid and to the application of competition rules to public enterprises. See Barry E. Hawk and Laraine L. Laudate, “Antitrust Federalism in the United States and Decentralization of Competition Law Enforcement in the European Union: A Comparison,” Fordham International Law Journal (1996): 18, 20, 31. 77. Pinar Akman, “Searching for the Long-Lost Soul of Article 82 EC,” Oxford Journal of Legal Studies 29, no. 2 (2009): 267– 303. 78. Id., 301. 79. Green Paper on Vertical Restraints in EC Competition Policy, Com (96) 721. Available at europa.eu/documents/comm/green_papers/pdf/com96_721_ en.pdf. 80. Commission Regulation (EC) No. 2790/1999 of 22 December 1999 on the application of Article 101(3) of the Treaty to categories of vertical agreements and concerted practices. 81. Commission Regulation (EU) No. 330/2010 of 20 April 2010 on the application of Article 101(3) of the Treaty on the Functioning of the European Union to categories of vertical agreements and concerted practices. 82. DG Competition, Discussion Paper on the Application of Article 82 of the Treaty to Exclusionary Abuses, 2005; DG Competion, Guidance on the Commision’s Enforcement Priorities on Applying Article 82 EC Treaty to Abusive Conduct by Dominant Undertaking, 2008. 83. Gerber, Global Competition, 187. 84. Brown Shoe Co. v. United States, 370 U.S. 294, 344 (1962). 85. Commission Decision of 03/07/2001 declaring a concentration to be incompatible with the common market and the EEA Agreement (Case No. COMP/ M.2220— General Electric/Honeywell); Honeywell International Inc. v. Commission of the European Communities (Case T-209/01). 86. Case T-201/04 Microsoft Corp. v. Commission of the European Communities (2007). 87. Ahlborn and Grave, “Walter Eucken and Ordoliberalism,” 203. 88. United States v. Microsoft Corp., 253 F.3d 34 (D.C. Cir. 2001). 89. Verizon Communications, Inc. v. Law Offices of Curtin V. Trinko, LLP, 540 U.S. 398, 410– 44 (2004). The law on these issues in both jurisdictions is complex and is examined extensively in chapter 9. 90. See, for example, Antitrust Modernization Commission, Report and Recommendations (Washington, DC: Antitrust Modernization Commission, 2007), 318– 20 (“AMC Report and Recommendations”); Jonathan Baker, “Competitive Price Discrimination: The Exercise of Market Power without Anticompetitive Effects,” Antitrust Law Journal 70 (2003): 643; William J. Baumol and Daniel G. Swanson, “The New Economy and Ubiquitous Competitive Price Discrimination: Identifying Defensible Criteria of Market Power,” Antitrust Law Journal 70 (2003): 661– 86. 91. Post Danmark A/S v. Konkurrencerådet, Case C-209/10 (27 March 2012). 92. LePage’s, Inc. v. 3M, 324 F.3d 141, 154– 57 (3D Cir. 2003).

notes to chapter 1

223

93. AMC Report and Recommendations, 94– 106; Cascade Health Solutions v. PeaceHealth, 515 F.3d 883 (9th Cir. 2008). 94. See discussion of the Intel case in chapter 7. 95. Major developmants include Robinson-Patman Act, Pub. L. No. 74-692, 49 Stat. 1526 (1936) (codified as amended at 15 U.S.C. §§ 13, 13b, 21a (2002)); Cellar-Kefauver Act of 1950, Pub. L. 81-899, 64 Stat. 1225 (codified as amended at 15 U.S.C. § 18 (2002)); Miller-Tydings Act, Pub. L. No. 314, ch. 690, Title VIII, 50 Stat. 693 (1937) (repealed 1975); McGuire Act, Pub. L. 82-542, 66 Stat. 631 (1952) (repealed 1975); Consumer Goods Pricing Act of 1975, Pub. L. No. 94-145, 89 Stat. 801 (1975); Hart-Scott-Rodino Antitrust Improvements Act of 1976, Pub. L. 94-145, 90 Stat. 1383 (codified at 15 U.S.C. § 18a (2002)); McCarran-Ferguson Act, Pub. L. 70-15, S9 Stat. 33 (codified at 15 U.S.C. §§ 1011– 15 (2002)). 96. No more than three commissioners can be from the same political party. 97. Albert Foer, “The Politics of Antitrust in the United States: Public Choice and Public Choices,” University of Pittsburgh Law Review 62 (2001): 475. 98. William J. Baer and David Balto, “The Politics of Federal Antitrust Enforcement,” Harvard Journal of Law and Public Policy 23, no. 1 (1999), cited in Foer, “Politics of Antitrust,” 478. Timothy Brennan has expressed a widely shared view from the perspective of the Justice Department: “the Division maintained a strong ethic of professionalism, with calls made on how the merits were dispassionately seen. Those of us lucky to work there could barely recognize the partisan, political worlds in which our colleagues in other parts of the government routinely operated— a contrast I saw firsthand when spending a year on the Council of Economic Advisers’ staff.” Timothy Brennan, “Predation, Exclusion, and Complement Market Monopolization,” Global Competition Policy Antitrust Chronicle 7 (2009): 4. 99. The FTC has always been influenced by the DOJ Guidelines, although they were not formally adopted by the FTC until 1992. For a formal model and quantitative evidence that covers many administrations and also finds only modest direct political impact in policy toward mergers, see Malcolm Coate, “A Test of Political Control of the Bureaucracy: The Case of Mergers,” Economics and Politics 1 (2002): 1– 15. 100. Despite its ambiguity; see chapter 2. Jacobs v. Tempur-Pedic Intern., Inc., 626 F.3d 1327 (11th Cir. 2010) (“consumer welfare, understood in the sense of allocative efficiency, is the animating concern of the Sherman Act”); Valuepest. com of Charlotte, Inc. v. Bayer Corp., 561 F.3d 282, 290– 91 (4th Cir. 2009) (“The purpose of the antitrust law, at least as articulated in the modern cases, is to protect the competitive process as a means of promoting economic efficiency”); Continental Airlines, Inc. v. United Airlines, 277 F.3d 499 (4th Cir. 2002); Pool Water Products v. Olin Corp., 258 F.3d 1024, 1034 (9th Cir. 2001) (“the antitrust laws are only concerned with acts that harm allocative efficiency and raise [. . .] the price of goods above their competitive level or diminish [. . .] their quality”); Chrome Lighting v. GTE Products Corp., 111 F.3d 653 (9th Cir. 1997); Chicago Professional Sports Ltd. v. National Basketball Ass’n, 95 F.3d 593, 602 (7th Cir. 1996) (Cudahy, concurring) (“the sole goal of antitrust is efficiency or, put another way, the maximization of total social wealth”); Morrison v. Murray Biscuit Co., 797 F.2d 1430 (7th Cir. 1986) (Posner) (“The purpose of antitrust law, at least as articulated in the modern cases, is to protect the competitive process as a means of

notes to chapter 1

224

promoting economic efficiency”); MCI Communications Corp. v. American Tel. & Tel. Corp., 708 F.2d 1081, 1113 (7th Cir. 1983) (“the antitrust laws are designed to encourage vigorous competition, as well as to promote economic efficiency and maximize consumer welfare”); Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 421 (9th Cir. 1995) (“allocative efficiency is synonymous with consumer welfare”); Arthur S. Landendefer, Inc. v. S. E. Johnson Co., 729 F.2d 1050 (6th Cir. 1984) (stating “the economic policies of the antitrust laws” are “to promote efficiency, encourage vigorous competition and maximize consumer welfare”). 101. Some believe that these incentives are too strong— perhaps especially in new economy industries. See Geoffrey E. Manne and Joshua D. Wright, “Innovation and the Limits of Antitrust,” Journal of Competition Law and Economics 6, no. 1 (2010): 153. Manne and Wright favor only single damages (rather than triple) for “overt” (i.e., nonhidden) offenses. An excellent discussion of many of the critical issues is presented in F. M Scherer and David Ross, Industrial Market Structure and Economic Performance, 3rd ed. (Boston: Houghton Mifflin, 1990), 326– 28. 102. W. Kip Viscusi, Joseph E. Harrington Jr., and John M. Vernon, Economics of Regulation and Antitrust, 4th ed. (Cambridge, MA: MIT Press, 2005), 72. 103. Richard A. Posner, “A Program for the Antitrust Division,” University of Chicago Law Review 38 (1971): 532, cited in Lawrence J. White, “The Growing Influence of Economics and Economists on Antitrust: An Extended Discussion,” New York University Law and Economics Working Papers No. 119 (2008), at 12. 104. Gerber, Law and Competition, 16– 114; Gerber, Global Competition, 163– 65. 105. Gerber, Global Competition, 163– 65. 106. Gerber, Law and Competition, 346– 47; Gerber, Global Competition, 182. 107. Gerber, Global Competition, 191– 92. 108. White Paper, COM (2008) 165 final; see also Commission Staff Working Paper annexed to the White Paper, SEC (2008) 404; Public consultation “Towards a coherent European approach to collective redress,” see http://ec.europa.eu /competition/consultations/2011_collective_redress/index_en.html. 109. European Commission, proposal for a directive of the European Parliament and of the Council on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union. Strasbourg, 11.6.2013 COM (2013) 404 final. 110. Damien Neven, “Competition Economics and Antitrust in Europe,” Economic Policy 48 (2006): 741–91. 111. Lars-Hendrick Roeller, “Economic Analysis and Competition Policy Enforcement in Europe,” in Modeling European Mergers: Theory, Competition Policy, and Case Studies, ed. Peter A. G. Van Bergeijk and Erik Kloosterhuis (Northampton: Edward Elgar, 2006). 112. For a study of various national competition agencies in a comparative perspective that includes quantitative measures, see Stephen Wilks, “Agencies, Networks, Discourses, and the Trajectory of European Competition Enforcement,” European Competition Journal 3, no. 2 (2007): 437– 64. 113. Airtours v. Commission (Case T-342/99), 2002 ECR II-2585; Schneider Electric v. Commission (Case T-310/01), 2002 ECR II 4071; Tetra Laval (Case T-5/02), 2002 ECR II-4381.

notes to chapter 1

225

114. This relates to what many political scientists call the “democratic deficit” and what the economist Dani Rodrik has referred to as the “augmented trilemma” of integrated economies, the traditional nation-state, and democratic governance. Dani Rodrik, “How Far Will International Integration Go?” Journal of Economic Perspectives 14, no. 1 (2000): 177; Dani Rodrik, “Feasible Globalizations,” in Globalization: What’s New?, ed. Michael Weinstein (New York: Columbia University Press, 2005), 196. 115. 50 Stat. 693 (1937). 116. 66 Stat. 631– 32 (1952). 117. Dr. Miles Medical Co. v. John D. Park & Sons Co., 220 U.S. 373 (1911). 118. Pub.L. No. 94-145 89 Stat 801 (1975). 119. Foer, “Politics of Antitrust,” 487. Organized labor in the United States has typically been little engaged on matters of product market competition, but it strongly opposed many of the industry deregulation measures of the 1970s. 120. One authoritative account characterizes Rhenish (in contrast to Anglo Saxon) capitalism as resting on four major characteristics: bank as opposed to market finance, cooperative labor relations and long tenure, firm-oriented labor training, and cooperation among firms. Andreas Busch, “European Integration, Varieties of Capitalism, and the Future of ‘Rhenish Capitalism,’” unpublished (2004). The last three characteristics are directly threatened by more intense competition and the first by EU attempts to create an integrated capital market that makes hostile takeover easier. 121. Angela Wigger and Andreas Nölke, “Enhanced Roles of Private Actors in EU Business Regulation and the Erosion of Rhenish Capitalism,” Journal of Common Market Studies 45, no, 2 (2007): 487. 122. The adoption of a more economic approach to antitrust and the growth of private litigation have generated a large demand for the services of economic consulting firms of the kind that have long been a fixture of the US competition policy scene. Their independent political and policy significance, however, is difficult to discern. Id.; Angela Wigger, “The Political Role of Transnational Experts in Shaping EU Competition Policy,” Legisprudence 3, no. 3 (2009): 251. 123. Hubert Buch-Hansen and Angela Wigger, The Politics of European Competition Regulation (New York: Routledge, 2011), 122. 124. Id. 125. Aérospatiale-Alenia/de Havilland (91/619) [1992], 1 CEC (CCH) 2034 (1991). 126. Political considerations were important in that case, with France and Italy, as well as the Commission’s Enterprise Directorate, lobbying the Commission on behalf of the French and Italian merger participants to approve the merger. Alissa A. Meade, “Modeling a European Competition Authority,” Duke Law Journal 46 (1996): 153, 167; D. G. Goyder, EC Competition Law, 2nd ed. (Oxford: Oxford University Press, 1993), 508; Gordon Borrie, “Personal View: Time for a Euro, MMC,” Financial Times, November 11, 1991, 16. 127. Ch. 50, 52 Stat. 516 (1918). 128. Motta, Competition Policy, 8. 129. Buch-Hansen and Wigger, Politics of European Competition Regulation. 130. For a discussion of the Boeing-Airbus and GE-Honeywell cases, see

notes to chapters 1–2

226

Daniel J. Gifford and Robert T. Kudrle, “European Competition Law and Policy: How Much Latitude for Convergence with the United States?” Antitrust Bulletin 48 (2003): 727. 131. Neven, “Competition,” 60.

ChApter 2 1. Jacobs v. Tempur-Pedic Intern., Inc., 626 F.3d 1327 (11th Cir. 2010) (“consumer welfare, understood in the sense of allocative efficiency, is the animating concern of the Sherman Act”); Valuepest.com of Charlotte, Inc. v. Bayer Corp., 561 F.3d 282, 290– 91 (4th Cir. 2009) (“The purpose of the antitrust law, at least as articulated in the modern cases, is to protect the competitive process as a means of promoting economic efficiency”); Continental Airlines, Inc. v. United Airlines, 277 F.3d 499 (4th Cir. 2002); Pool Water Products v. Olin Corp., 258 F.3d 1024, 1034 (9th Cir. 2001) (“the antitrust laws are only concerned with acts that harm allocative efficiency and raise [. . .] the price of goods above their competitive level or diminish [. . .] their quality”); Chrome Lighting v. GTE Products Corp., 111 F.3d 653 (9th Cir. 1997); Chicago Professional Sports Ltd. v. National Basketball Ass’n, 95 F.3d 593, 602 (7th Cir. 1996) (Cudahy, concurring) (“the sole goal of antitrust is efficiency or, put another way, the maximization of total social wealth”); Morrison v. Murray Biscuit Co., 797 F.2d 1430 (7th Cir. 1986) (Posner) (“The purpose of antitrust law, at least as articulated in the modern cases, is to protect the competitive process as a means of promoting economic efficiency”). 2. Rebel Oil Co. v. Atlantic Richfield Co., 51 F.3d 421 (9th Cir. 1995) (“allocative efficiency is synonymous with consumer welfare”); Arthur S. Landendefer, Inc. v. S. E. Johnson Co., 729 F.2d 1050 (6th Cir. 1984) (stating “the economic policies of the antitrust laws” are “to promote efficiency, encourage vigorous competition and maximize consumer welfare”); MCI Communications Corp. v. American Tel. & Tel. Co., 708 F.2d 1081, 1113 (7th Cir. 1983) (“the antitrust laws are designed to encourage vigorous competition, as well as to promote economic efficiency and maximize consumer welfare). 3. Bork, Antitrust Paradox, 107– 8. Bork includes benefit to producers resulting from cost reduction as adding to consumer welfare. 4. Reiter v. Sonotone Corp., 442 U.S. 330, 343 (1979); see Daniel J. Gifford and Robert T. Kudrle, “Rhetoric and Reality in the Merger Standards of the United States, Canada, and the European Union,” Antitrust Law Journal 72 (2005): 423, 428. 5. Aggregate welfare is maximized when the sum of allocative and productive efficiencies is maximized. When antitrust law fosters net efficiency, therefore, it fosters aggregate welfare. See Alan Devlin and Bruno Peixoto, “Reformulating Antitrust Rules to Safeguard Societal Wealth,” Stanford Journal of Law, Business, and Finance 13 (2008): 225, 259; Yedida Z. Stern, “A General Model for Corporate Acquisition Law,” Journal of Corporate Law 26 (2001): 675, 678. See also Joseph F. Brodley, “The Economic Goals of Antitrust: Efficiency, Consumer Welfare, and Technological Progress,” New York University Law Review 62 (1987): 1020 (describing these goals and making recommendations for achieving them). 6. Illustrative diagrams can be developed to show either industry or individual

notes to chapter 2

227

firm outcomes. Here industry outcomes are shown, and each seller is experiencing only a fraction of the cost and sales volume. Only if the diagram is meant to show the situation after a merger to monopoly is the distinction unimportant. 7. Joseph A. Schumpeter, Capitalism, Socialism, and Democracy (New York: Harper, 1942), 81– 86. 8. While the adoption of the total surplus standard is not presently feasible in the European Union, it is certainly advocated by many leading European economists. See, for example, Motta, Competition Policy, 18– 22. 9. See Oliver E. Williamson, “Economies as an Antitrust Defense Revisited,” University of Pennsylvania Law Review 125 (1977): 699, 728– 31, 734– 35; Oliver E. Williamson, “Economies as an Antitrust Defense: The Welfare Tradeoffs,” American Economic Review 58 (1968): 18, 34; Bork, Antitrust Paradox, 107– 15. Increased price accompanied by decreased cost has come to be known as the “Williamson trade-off.” 10. See, for example, Herbert Hovenkamp, “Exclusive Joint Ventures and Antitrust Policy,” Columbia Business Law Review 1 (1995): 51; Frank H. Easterbrook, “Vertical Arrangements and the Rule of Reason,” Antitrust Law Journal 53 (1984): 135, 159; Frank H. Easterbrook, “The Limits of Antitrust,” Texas Law Review 63 (1984). 11. Robert H. Lande, “Wealth Transfers as the Original and Primary Concern of Antitrust: The Efficiency Interpretation Challenged,” Hastings Law Journal 34 (1982): 65. See also Robert H. Lande, “Chicago’s False Foundation: Wealth Transfers (Not Just Efficiency) Should Guide Antitrust,” Antitrust Law Journal 58 (1989): 631; Alan A. Fisher and Robert H. Lande, “Efficiency Considerations in Merger Enforcement,” California Law Review 71 (1983): 1580. 12. Robert Pitofsky, former chairman of the Federal Trade Commission, has asserted that he would be willing to balance merger-generated efficiencies against anticompetitive aspects of a merger. Because he recognizes the difficulties involved in proving that the benefits of efficiencies would be passed on to consumers, Pitofsky would take efficiencies into accounting as a matter of enforcement discretion. Robert Pitofsky, “Antitrust Policy in a Clinton Administration,” Antitrust Law Journal 62 (1993): 217, 221. Nonetheless, his evaluations appear to be based entirely upon estimates of an increase or decrease in consumer surplus. Robert Pitofsky, “Proposals for Revised United States Merger Enforcement in a Global Economy,” Georgetown Law Journal 81 (1992): 195, 208– 9; Robert Pitofsky, “Efficiencies in Defense of Mergers: Two Years After,” George Mason Law Review 7 (1999): 485, 492. 13. Steven C. Salop, “Question: What Is the Real and Proper Antitrust Welfare Standard? Answer: The True Consumer Welfare Standard,” Loyola Consumer Law Review 22, no. 2 (2006): 3– 27. 14. Pinar Akman has concluded that some involved with the original drafting of Article 82 may have intended a total surplus standard. Nevertheless, she acknowledges that almost everyone involved in subsequent administration and adjudication has taken a different view. Akman, “Searching for the Long Lost Soul,” 298. 15. Joseph Farrell and Michael L. Katz, “The Economics of Welfare Standards in Antitrust,” Competition Policy International 2, no. 2 (2006).

notes to chapter 2

228

16. See also Bruce R. Lyons, “Could Politicians Be More Right Than Economists?” Revised CCR Working Paper CCR 02-1 (2002). 17. See Gary L. Roberts and Steven C. Salop, “Efficiency Benefits in Dynamic Merger Analysis,” 1994 (unpublished); Gary L. Roberts and Steven C. Salop, “Efficiencies in Dynamic Merger Analysis: A Summary,” World Competition: Law and Economics Review 19, no. 5 (1996). 18. Roberts and Salop, “Efficiency Benefits,” 35. 19. Scherer and Ross, Industrial Market Structure, 4. 20. Dennis W. Carlton, “Does Antitrust Need to Be Modernized?” Antitrust Division, U.S. Department of Justice Economic Analysis Group Discussion Paper No. 07-6, January 2007. 21. United States v. Aluminum Co. of America, 148 F.2d 416, 424 (2d Cir. 1945). 22. Market definition is a challenging policy issue that will be considered in the following chapter. 23. Copperweld Corp. v. Independence Tube Corp., 467 U.S. 752 (1984). 24. United States v. U.S. Steel Corp., 251 U.S. 417, 451 (1920). 25. American Tobacco v. United States, 328 U.S. 781 (1946). 26. United States v. E. I. du Pont de Nemours & Co., 351 U.S. 377 (1956). 27. The term was used by Judge Knox in the Alcoa case, 92 F.Supp. 333, 341 (S.D.N.Y. 1950). See Kenneth Elzinga and David E. Mills, “The Lerner Index of Monopoly Power: Origins and Uses,” American Economic Review 101, no. 3 (2011): 558– 64 at 560. 28. United States v. Grinnell Corp., 384 U.S. 563, 571 (1966). 29. A. P. Lerner, “The Concept of Monopoly and the Measurement of Monopoly Power,” Review of Economic Studies 1 (1934): 157, 169. 30. The formula is (P – MC)/P = –(1/e) where e is the price elasticity of demand. 31. Elzinga and Mills, “Lerner Index.” Even though Lerner employed the term “monopoly power” to refer to all situations in which price exceeds marginal cost, others have used that term differently. A leading industrial organization textbook, for example, uses “monopoly power” to refer to the profitable setting of optimal prices by a firm, while employing “market power” to describe break-even situations (such as the long run in monopolistic competition), while suggesting that most economists use the terms interchangeably. Carlton and Perloff, Modern Industrial Organization, 93. The authors observe that “people do not always make this distinction, and generally use the two terms interchangeably, sometimes creating confusion.” As if to verify the latter point, the same textbook later uses the term “market power” in the context of policy interventions that clearly aim at supernormal profits. Carlton and Perloff, Modern Industrial Organization, 643. Gregory Werden advocates using “monopoly power” synonymously with the Lerner index— where marginal cost is explicitly long run. Gregory Werden, “Demand Elasticities in Antitrust Analysis,” Antitrust Law Journal 66 (1997– 98): 373. But as a policy relevant measure, this seems to ignore problems of sunk costs and economies of scale in which marginal cost is always lower than average total cost. 32. Werden, “Demand Elasticities.” See also Lawrence J. White, “Market Power and Market Definition in Monopolization Cases,” in ABA Section of Antitrust Law, Issues in Competition Law and Policy (2008): 913, 917.

notes to chapter 2

229

33. See, for example, Eastman Kodak Co. v. Image Technical Services. Inc., 504 U.S. 451, 481 (1992) (“Monopoly power under § 2 [of the Sherman Act] requires . . . something greater than market power under § 1”). Werden states that “circuit courts have commonly distinguished ‘market power’ from ‘monopoly power’ as a matter of degree, and the Supreme Court has used the two terms in essentially this manner.” Werden, “Demand Elasticities,” 363, 378. 34. Einer Elhauge, “Defining Better Monopolization Standards,” Stanford Law Review 56 (2003): 253, 258– 59. 35. This can be seen by rotating a marginal revenue curve clockwise around a fixed intersection with marginal cost. 36. For a typical assessment, see Richard Schmalensee, “Sunk Costs and Antitrust Barriers to Entry,” American Economic Review 94, no. 2 (2004): 471– 75 at 471– 72. 37. The real goal was to be the extirpation of the trapezoid of lost consumer surplus, so minimizing profits would be formally correct only if all relevant demand elasticities were the same. A total surplus standard would aim to minimize the summed deadweight loss triangles across markets. Maximizing consumer surplus is not the same as aiming to lower the highest rates of return on equity capital corrected for market size. The goal would be the same only if all investment were properly measured and depreciated, the ratio of equity to total investment were the same in all industries, and all industries were equally capital intensive. The “performance” part of the original “Harvard” “structure, conduct, performance” approach to competition policy implicitly employed a consumer surplus standard and was very concerned with price-cost margins. It sometimes attempted to use firm profitability as a rough index of them, but limitations were recognized. See Joe S. Bain, “The Profit Rate as a Measure of Monopoly Power,” Quarterly Journal of Economics 55 (1941): 271, 288. 38. 196 U.S. 375 (1905). 39. Don T. Hibner, “Attempts to Monopolize: A Concept in Search of Analysis,” Antitrust Law Journal 34 (1967): 165– 77; Daniel J. Gifford, “The Role of the Ninth Circuit in the Development of the Law of Attempt to Monopolize,” Notre Dame Law Review 61 (1986): 1021–51. 40. The Ninth Circuit’s expansionary construction of the attempt offense began in Lessig v. Tidewater Oil Co., 327 F.2d 459 (9th Cir.), cert. denied, 377 U.S. 993 (1964) and continued until the Supreme Court’s decision in Spectrum Sports, Inc. v. McQuillan, 506 U.S. 447 (1993). 41. 506 U.S. 447 (1993). 42. Gerber, Global Competition, 131. 43. Case 85/76 Hoffmann-La Roche v. Commission [1979] ECR 461 at paragraph 38. 44. Id., at paragraph 71. 45. See John Vickers, “Abuse of Market Power,” Economic Journal 115 (2005): F244– F261. 46. The conceptual basis for determining what goods and services in what area constitute a “market” will be considered in the following chapter. 47. France v. Commission, paragraph 221 notes the need for “correlative factors.” 48. Guidelines sections 39 and 40.

notes to chapter 2

230

49. Treaty on the Functioning of the European Union, Art. 263. Tetra Laval ECJ, Case C-12/03 P [2004] C.C.R. I-987, ¶ 39. 50. Joined Cases C204/00, Aalborg Portland and Others v. Commission, 2004 E.C.R. I-403 at ¶ 279; Case 42/84, Remia and Others v. Commission, 1985 E.C.R. 2545 at ¶ 34; Joined Cases C142/84, BAT and Reynolds v. Commission, 1987 E.C.R. 4487 at ¶ 62; Joined Cases C-68/94, French Republic v. EC Commission, 1998 E.C.R. I-1375 at 224. See Bo Vesterdorf, “Judicial Review in EC Competition Law: Reflections of the Role of the Community Courts in the EC System of Competition Law Enforcement,” Competition Policy International 1 (2005): 3, 11. 51. Deutsche Telecom AG v. European Commission (C-280/08 P) [2010] 5 C.M.L.R. 27 at ¶ 143. 52. American courts have long attempted to delineate the respective roles allocated to courts and administrative bodies. As early as 1951, that issue was a focus of attention in Universal Camera cases. Universal Camera Corp. v. NLRB, 340 U.S. 474 (1951). Judge Frank’s concurring opinion on remand set out a conceptual framework for issue allocation that has retained its vitality over the intervening years. NLRB v. Universal Camera Corp., 190 F.2d 429, 432 (2d Cir. 1951). Under Judge Frank’s approach, courts defer to agencies on matters of policy. Since 1984 judicial deference to agency statutory interpretations has been mandated under the so-called Chevron doctrine. Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984). Justice Breyer has offered a complementary analysis of when such deference is required that, inter alia, allocates final resolution of issues to courts or agencies, depending on their overall importance. Barnhart v. Wilson, 535 U.S. 212, 222 (2002). See also Mayburg v. Secretary of Health and Human Services, 740 F.2d 100, 106– 7 (1st Cir. 1984) (Breyer). By contrast, Christian Ahlborn and David Evans fault the EU courts for failing adequately to explain their different approaches to judicial review involving complex economic assessments. They point out that the courts accord greater deference to the Commission in Article 102 cases than in merger cases. Yet under the rationale that the Commission possesses a greater competence on complex economic assessments, the judicial practice should be the opposite. Article 102 cases involve rules that are more form-based, and economic theory and evidence are less relevant whereas merger cases involve rules that are effects-based, and it is in these cases where economic theory and evidence matter more. See Christian Ahlborn and David S. Evans, “The Microsoft Judgment and Its Implications for Competition Policy towards Dominant Firms in Europe,” Antitrust Law Journal 75 (2009): 887– 932. 53. Joined Cases C-68/94, French Republic and Others v. EC Commission, 1998 E.C.R. I-1375. 54. Id., at ¶¶ 223, 224. 55. Id., at ¶ 226, ¶¶ 227– 32. 56. Airtours v. Commission, Case T-342/99, 2002 E.C.R. II-2585 (Court of First Instance [CFI]); Schneider Electric SA v. Commission, Case T-310/01, 2002 E.C.R. II-4071 (CFI); Tetra Laval BV v. Commission, Case T-5/02, 2002 E.C.R. II4381 (CFI). In these cases, the CFI reversed the Commission’s disapproval of mergers. In Sony/BMG, the CFI reversed the Commission’s unconditional approval of a merger. Sony/BMG, Case COMP/M, 3333 Sony/BMG, 2005 O.J. (L 62) 30. The CFI had previously reversed a Commission’s conditioned approval. Case T-114/02,

notes to chapters 2–3

231

BaByliss v. Commission, 2002 O.J. (C 144) 57; 2003 O.J. (C135) 27. Commentators have suggested that the Commission’s standard of proof should be higher under Articles 101 and 102 than under the Merger Regulation on the ground that the effects of a merger lie in the future and are accordingly matters of prediction whereas violations of Articles 101 and 102 are examined after the fact when evidence of their effects is more readily available. Under this theory, the Commission should have been held to a higher standard of proof in Microsoft (which was an Article 102 proceeding) than in the merger cases. Yet the opposite appears to be true. Despite the existence of evidence about past events that characterize Article 102 cases, the Court deferred to the Commission’s assessments. 57. Sony/BMG, note 56. 58. Airtours, note 56 at ¶ 64; Tetra Laval, note 56 at ¶ 119. 59. Airtours, note 56, at ¶¶ 120, 181, 182, 294; Schneider, note 56 at ¶¶ 262, 296, 297, 314, 315, 349, 350, 369, 403, 404, 411. 60. Tetra Laval ECJ, Case C-12/03 P [2004] C.C.R. I-987, ¶ 39. 61. Microsoft Corp. v. Commission, Case T-101/04, 2004 E.C.R. II-3619. 62. ¶ 87. See also ¶¶ 379, 482. 63. IMS Health GmbH & Co. OHG v. NDC Health GmbH & Co. KG, 2004 E.C.R. I-5039; Oscar Bronner GmbH& Co. KG v. Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG, 1998 E.C.R. I-07791; Tierce Ladbroke SA v. Commission, Case T-504/93 (1997) [1997] 5 C.M.L.R. 309; Radio Telefis Eireann (RTE) v. Commission, 1995 E.C.R. I-743; AB Volvo v. Erik Veng (UK) Ltd., 1988 E.C.R. 6211. 64. Chevron, U.S.A., Inc. v. Natural Res. Def. Council, 467 U.S. 837 (1984). 65. Baltimore Gas & Electric Co. v. Natural Resources Def. Council, Inc., 462 U.S. 87, 103 (1983). “A reviewing court must remember that the Commission is making predictions, within its area of special expertise, at the frontiers of science. When examining this kind of scientific determination, as opposed to simple findings of fact, a reviewing court must generally be at its most deferential.” 66. Compare Ethyl Corp. v. EPA, 541 F.2d 1, 66 (D.C. Cir. 1976) (Bazelon, concurring) with id., 68 (Leventhal, concurring).The two protagonists were Judges Bazelon and Leventhal, both of whom sat on the DC Circuit. Judge Bazelon believed that judges were incapable of evaluating scientific issues, and he took the position that in judicial review the courts should limit themselves to ensuring that the agency had assembled the right evidence and provided an adequate hearing. Judge Leventhal, by contrast, was willing to look at the evidence before the agency and assess its substantive response. See Ronald J. Krotoszynski Jr., “‘History Belongs to the Winners’: The Bazelon-Leventhal Debate and the Continuing Relevance of the Process/Substance Dichotomy in Judicial Review of Agency Action,” Administrative Law Review 58, no. 4 (2006): 995. 67. See Lands Council v. McNair, 537 F.3d 981, 993– 94 (9th Cir. 2008). 68. Business Elec. Corp. v. Sharp Elec. Corp., 485 U.S. 717, 731– 32 (1988).

ChApter 3 1. Horizontal mergers increase the concentration level in specific markets, although not necessarily product markets as usually defined. The important Staples–

notes to chapter 3

232

Office Depot case involved the sales of ranges of goods through a particular type of retail outlet. Vertical mergers involve combinations between suppliers and their customers. Conglomerate mergers involve firms that are neither in a competitive relationship nor in a supplier-customer relationship. The European Commission investigates conglomerate mergers with a focus on whether the relationships among their products provide incentives for the merged firm to act anticompetitively. 2. In the United States, the best-known example may be the forced divestiture of GM stock by DuPont in 1956. United States v. E. I. DuPont de Nemours & Co., 353 U.S. 586 (1957). In Europe, considerations of employment impacts and other elements of “industrial policy” have driven competition policy outcomes. For example, the proposed De Havilland merger in the European Union in 1991 seems to have been determined by myriad political forces more than economic analysis. Mark A. Pollack, The Engines of European Integration: Delegation, Agency, and Agenda Setting (Oxford: Oxford University Press, 2003), 292– 94. 3. Or buyers. This issue will not be pursued here, but in the chapters on vertical arrangements, the possible consumer welfare or total welfare-enhancing impact of “power buyers” is considered. Such buyers may sometimes result from mergers. 4. Bain, Industrial Organization. The Chicago School also acknowledges this mechanism as a possibility; see George Stigler, “A Theory of Oligopoly,” Journal of Political Economy 72 (1964): 44– 61. 5. A vast literature on the limitations of the early research is succinctly reviewed in Scherer and Ross, Industrial Market Structure, 411– 47. 6. One possible case would arise from a merger by a dominant firm selling homogeneous products that, by assumption, already faced a downward sloping demand curve, with smaller price-taking firms. Another possibility would be a case of Cournot competition with homogeneous products, no entry, and constant and equal marginal cost: price-cost margins are inversely related to the number of firms because the elasticity of the demand curve facing each firm increases with firm numbers. In the simplest model, Cournot firms make output decisions simultaneously on the assumption that other firms keep their outputs fixed at the observed level. 7. To the degree that the two firms were tacitly colluding before the merger, of course, there would be less (or no) change. 8. Bain, Industrial Organization, 270– 73, 460. 9. 15 U.S.C. § 18 (1994). 10. 15 U.S.C. §§ 1, 2 (1994). 11. The 1950 amendment also rephrased the language that had apparently limited Section 7’s prohibition to horizontal acquisitions; Celler-Kefauver Act of 1950, 64 Stat. 1125 (1950) (codified as amended at 15 U.S.C. § 18 (1994)). 12. United States v. Columbia Co., 334 U.S. 495 (1948). 13. These included: Brown Shoe Co. v. United States, 370 U.S. 294 (1962); United States v. Philadelphia National Bank, 374 U.S. 321 (1963); United States v. Aluminum Co. of America, 377 U.S. 271 (1964); United States v. Continental Can Co., 378 U.S. 441 (1964); United States v. Von’s Grocery Co., 384 U.S. 270 (1966); United States v. Pabst Brewing Co., 384 U.S. 546 (1966). 14. Brown Shoe Co. v. United States, 370 U.S. at 344. 15. United States v. Philadelphia National Bank, 374 U.S. 321 (1963) at 362.

notes to chapter 3

233

16. United States v. Von’s Grocery Co., 384 U.S. 270 (1966). 17. United States v. Pabst Brewing Co., 384 U.S. 546 (1966). 18. United States v. El Paso Natural Gas Co., 376 U.S. 651 (1964); United States v. Penn-Olin Chem. Co., 378 U.S. 158 (1964); FTC v. Consolidated Foods Corp., 380 U.S. 592 (1965); FTC v. Proctor and Gamble Co. 386 U.S. 568 (1967); United States v. Falstaff Brewing Co., 410 U.S. 526 (1973); Ford Motor Co. v. United States, 405 U.S. 562 (1972). 19. The acquiring firm might be a potential competitor because it was likely that it would expand its product line into the lines of the acquired firm (a productmarket extension merger) or because it was likely that it would expand geographically into the territory presently served by the acquired firm (a geographic-market extension merger). The potential competition doctrine raised two competition issues: perceived potential competition and actual potential competition. The former raised the question whether the potential of the acquiring firm to enter the target market (i.e., the market in which the acquired firm operated) exerted a constraining effect upon the pricing and product policies of the firms active in that market. The latter raised the question whether, if the merger were barred, the acquiring firm would enter the target market de novo and thereby help to de-concentrate that market. It will be observed that neither aspect of the potential competition doctrine can apply unless the target market is itself concentrated. 20. United States v. General Dynamics Corp., 415 U.S. 486 (1974). 21. In General Dynamics the government carried its initial burden by introducing statistics showing that the sales of the merged firm would constitute a large share of sales. The defendant rebutted that evidence however, by showing that the acquired firm held few coal reserves and that the market was for new long-term supply contracts. Because of its low reserves, the acquired firm was not a significant participant in the actual market, the market for long-term contracts. Hence the acquisition did not result in an increase in concentration among market participants. 22. United States v. Marine Bancorporation, Inc., 418 U.S. 602 (1974); United States v. Connecticut Nat’l Bank, 418 U.S. 656 (1974). 23. 418 U.S. at 61; 418 U.S. at 669– 70. 24. See, for example, Proctor & Gamble Co., 63 F.T.C. 1465, 1580– 81 (1963). 25. United States v. Aluminum Co. of America, 148 F.2d 416, 429 (2d Cir. 1945). 26. Proctor & Gamble Co., 63 F.T.C. 1465 (1963). 27. FTC v. Proctor & Gamble Co., 386 U.S. 568, 579 (1967). The Federal Trade Commission in that case had ruled that advertising efficiencies generated entry barriers. It also rejected “as specious in law and unfounded in fact, the argument that the Commission ought not, for the sake of protecting the “inefficient” small firms in the industry, proscribe a merger so productive of “efficiencies.” Proctor & Gamble Co., 63 F.T.C. 1465, 1580 (1963). In Brown Shoe the Court thought that because distributional efficiencies were unavailable to competing small retailers, the merger was condemned. 28. Bork, Antitrust Paradox, 128. 29. U.S. Dep’t of Justice, 1982 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13,102. 30. U.S. Dep’t of Justice, 1982 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13,102 at 20, 549– 13.

notes to chapter 3

234

31. Id. 32. U.S. Dep’t of Justice, 1982 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13,102 at 20,549-13 n. 53. 33. U.S. Dep’t of Justice, 1984 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13,103. 34. U.S. Dep’t of Justice, 1984 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13,103 at 20, 554. 35. Id. 36. 1984 Merger Guidelines § 3.5. The 1984 guidelines also indicate that the department will give more weight to expected efficiencies in its assessment of vertical mergers. Id., § 4.24. 37. U.S. Dep’t of Justice, 1992 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13,104. 38. U.S. Dep’t of Justice, 1992 Merger Guidelines § 4, Trade Regulation Reporter 4 (CCH) ¶ 13,104 at 20,573-13-20,574 (effective April 2, 1992– April 7, 1997). 39. U.S. Dep’t of Justice, 1992 Merger Guidelines § 4, Trade Regulation Reporter 4 (CCH) ¶ 13,104 at 20,573-11-20,573-13 (as revised April 8, 1997). 40. See Gregory J. Werden, “An Economic Perspective on the Analysis of Merger Efficiencies,” Antitrust (Summer 1997): 11, 12, 14; Robert M. Vernail, “One Step Forward, One Step Back: How the Pass-On Requirement for Efficiencies Benefits in FTC v. Staples Undermines the Revisions to the Horizontal Merger Guidelines Efficiencies Section,” George Mason Law Review 7, no. 1 (1998): 133– 62. 41. Vernail, “One Step Forward,” § 4 n. 37 (contrasting a short-term effect with “delayed benefits”). 42. Thomas J. Horton, “The New United States Horizontal Merger Guidelines: Devolution, Evolution, or Counterrevolution?” Journal of European Competition Law and Practice 2, no. 2 (2011): 158–64. 43. U.S. Dep’t of Justice, 1992 Merger Guidelines, § 5. 44. Id., § 7. 45. Dennis W. Carlton, “Revising the Horizontal Merger Guidelines,” Journal of Competition Law and Economics 6, no. 3 (2010): 619– 52. 46. Barry C. Harris and Joseph J. Simons, Focusing Market Definition: How Much Substitution Is Necessary? Research in Law and Economics 12 (Greenwich, CT: JAI Press, 1989), 207– 26. 47. Joseph Farrell and Carl Shapiro, “Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition,” B. E. Journal of Theoretical Economics 10, no. 1 (2010). 48. Richard Schmalensee, “Should New Merger Guidelines Give UPP Market Definition?” Antitrust Chronicle (2009): 1. 49. Jerry Hausman, “2010 Merger Guidelines: Empirical Analysis.” MIT, mimeographed. 50. The Merger Guidelines have moved from hostility toward accepting an efficiencies defense in the 1982 version toward greater acceptance of efficiencies in successive iterations in 1984, 1992, and 1997. See Gifford and Kudrle, “Rhetoric and Reality.” In the 1984 US Guidelines, the DOJ attempted to distance itself from some of the skeptical language that it used about the proof of efficiencies

notes to chapter 3

235

in the 1982 version. In its introduction to the 1984 US Guidelines, it described the language in the earlier version as having “a restrictive, somewhat misleading tone.” U.S. Dep’t of Justice, Merger Guidelines (1984) (Statement— Efficiencies), reprinted in Trade Regulation Reporter 4 (CCH) ¶ 13, 103 (hereinafter 1984 US Guidelines). The 1984 US Guidelines state that the DOJ will “consider” a claim that a merger was required to achieve “significant net efficiencies” so long as the parties establish such efficiencies “by clear and convincing evidence.” Trade Regulation Reporter 4 (CCH) ¶ 13,103 § 3.5. The 1992 revision kept the language of the 1984 version but dropped the requirement that the efficiencies must be established by clear and convincing evidence. The efficiency provisions of the Merger Guidelines were further revised in 1997 so as superficially to widen the acceptance of efficiencies as a merger defense. Yet the 1997 revision, which took place under a Democratic administration, appears to favor operating efficiencies (which would lower marginal cost and thus directly affect price) over efficiencies achieved by reductions in fixed cost (which do not). Thus the 1997 revision appears to favor a consumer-surplus approach to the evaluation of cost savings. See Werden, “An Economic Perspective,” 12, 14. Deborah Garza observes that “while Republicans generally have worried about the efficiency (and liberty) losses of false positives [i.e., false indicators of price increases or other anticompetitive effects], the Democrats worried about false negatives [i.e., false indicators of competitive results such as those falsely indicating efficiency].” Deborah A. Garza, “A Comparative Analysis of the Clinton Antitrust Program and Suggestion of Changes to Come,” Antitrust 15 (2001): 64. She states that “because Clinton enforcers had greater faith in the predictive strength of concentration ratios, they imposed substantial hurdles in regard to efficiency claims.” Garza, “A Comparative Analysis,” 64. Robert Pitofsky, the FTC chairman during the Clinton administration, had denounced a requirement that efficiencies be passed on to consumers before he was appointed to the Commission. Pitofsky, “Proposals,” 208– 9. Yet other Clinton administration antitrust officials appear to have endorsed that requirement. Richard Steuer points out that the efficiencies revision did not require a pass-on of cost savings, but he nonetheless observes that there was considerable staff support for such a provision. Richard M. Steuer, “Capitalist Welfare,” Antitrust (2001): 4, 5. Timothy Muris (a Bush appointee) has criticized the 1997 efficiency revisions for preferring operational over fixed-cost efficiencies. See Stephen Stockum, “An Economist’s Margin Notes: The Antitrust Writings of Timothy Muris,” Antitrust (2002): 60, 62. 51. See William J. Kolasky and Andrew R. Dick, “The Merger Guidelines and the Integration of Efficiencies into Antitrust Review of Horizontal Mergers,” US Department of Justice, http://www.justice.gov/atr/hmerger/11254.htm. 52. Horton, “The New United States Horizontal Merger Guidelines,” 158. 53. Carlton, “Revising the Horizontal Merger Guidelines,” 619– 52. 54. Goetz Drauz, Stephen Mavroghenis, and Sara Ashall, “Recent Developments in EU Merger Control, 1 September 2009– 31 August 2010,” Journal of European Competition Law and Practice 2 (2011): 46. 55. FTC v. H. J. Heinz Co., 246 F.3d 708, 720– 21 (D.C. Cir. 2001); FTC v. Tenet Health Care Corp., 186 F.3d 1045, 1054 (8th Cir. 1999); FTC v. Univ. Health, Inc., 938 F.2d 1206, 1222 (11th Cir. 1991); FTC v. Cardinal Health, Inc., 12 F.Supp.2d 34, 61– 62 (D.D.C. 1998); FTC v. Staples, Inc., 970 F.Supp. 1066, 1088– 89 (D.D.C. 1997);

notes to chapter 3

236

FTC v. Butterworth Health Corp., 946 F.Supp. 1285, 1300– 01 (W.D. Mich. 1996), aff’d mem., 121 F.3d 708 (6th Cir. 1997). Many courts have indicated that some of the merger-generated cost savings must be passed on to consumers, thus suggesting a consumer-surplus approach to the evaluation of efficiency. See Cardinal Health, 12 F.Supp.2d at 62; Univ. Health, 938 F. 2d at 1223; Butterworth Health, 946 F.Supp. at 1301; FTC v. Swedish Match, 131 F.Supp.2d 151, 172 (D.D.C. 2000); United States v. United Tote, Inc., 768 F.Supp. 1064, 1084– 85 (D. Del. 1991); California v. American Stores Co., 697 F.Supp. 1125, 1132– 33 (C.D. Cal. 1988), aff’d in part and rev’d in part on other grounds, 872 F.2d 837 (9th Cir. 1989), rev’d, 495 U.S. 271 (1990). 56. Gifford and Kudrle, “Rhetoric and Reality,” 423. 57. In an unpublished opinion, however, the Sixth Circuit has affirmed a district court’s refusal to issue a preliminary injunction against a merger on the ground that cost savings would be passed on to consumers, thus providing additional support for the existence of an efficiency defense. Butterworth Health, 946 F.Supp. 1285, 1301 (W.D. Mich. 1996), aff’d mem., 121 F.3d 708 (6th Cir. 1997). 58. See, for example, Cardinal Health, 12 F.Supp.2d at 62– 63 (recognizing merger-generated savings of between $38 and $52 million per year, but rejecting an efficiencies defense on the ground that continued competition might also generate those savings, albeit not as quickly as the merger). 59. See United States v. Long Island Jewish Med. Ctr., 983 F.Supp. 121, 148– 49 (E.D.N.Y. 1997); United States v. Country Lake Foods, Inc., 754 F.Supp. 669, 674, 680 (D. Minn. 1990). 60. FTC v. Staples, Inc., 970 Supp. 1066, 1088– 90 (D.D.C. 1997). Staples involved an FTC challenge to a proposed merger between Staples, Inc. and Office Depot, Inc., two of the three major office-supply superstores. The merger proponents presented evidence of substantial projected savings, but the court discovered flaws in almost all of this evidence. See Staples, 970 F.Supp. at 1089– 90. The defendants argued that two-thirds of the projected cost savings would be passed through as a 3 percent cost saving that would outweigh its estimated price increase of 0.8 percent and would result in a net price drop of 2.2 percent. The FTC’s parallel estimates were 0.2 percent and 7.3 percent for a net price increase of 7.1 percent. The presiding judge found the government’s case more persuasive and decided against the merger. Serdar Dalkir and Frederick R. Warren-Boulton, “Prices, Market Definition, and the Effects of Merger: Staples-Office Depot (1997),” in The Antitrust Revolution: Economics, Competition, and Policy, 5th ed., ed. John E. Kwoka Jr. and Lawrence J. White (Oxford: Oxford University Press, 2009), 178– 201; see also Staples 970 F.Supp. at 1090. 61. FTC v. Tenet Health Care Corp., 186 F.3d 1045, 1054 (8th Cir. 1999). 62. See cases cited in note 55. 63. FTC v. H. J. Heinz Co., 246 F.3d 708 (D.C. Cir. 2001). 64. Heinz, 116 F.Supp.2d 190, 193 (D.D.C. 2000), rev’d, 246 F.3d 708 (D.C. Cir. 2001). 65. Heinz at 195– 96. 66. 246 F.3d at 721. 67. Id. 68. 246 F.3d at 711. On the assumption that Heinz would add Beech-Nut’s

notes to chapter 3

237

premerger output to its own premerger output, the former would constitute approximately 47 percent of the total output of the merged firm. 69. See Jonathan B. Baker, “Efficiencies and High Concentration: Heinz Proposes to Acquire Beech-Nut (2001),” in Kwoka and White, The Antitrust Revolution, 150, 164. 70. 246 F.3d at 720. 71. US Guidelines at § 4 (footnote omitted). 72. Id. 73. Council Regulation (EC) No 139/2004 of 20 January 2004 on the control of concentrations between undertakings; Council Regulation (EEC) No 4064/89 of 21 December 1989 on the control of concentrations between undertakings. In evaluating a merger, article 2 of the 2004 Merger Regulation requires the Commission to take into account: “(b) the market position of the undertakings concerned and their economic and financial power, the alternatives available to suppliers and users, their access to supplies or markets, any legal or other barriers to entry, supply and demand trends for the relevant goods and services, the interests of the intermediate and ultimate consumers, and the development of technical and economic progress provided that it is to consumers’ advantage and does not form an obstacle to competition.” Similar language was contained in Article 1 of the 1989 Merger Regulation. 74. See, for example, Paul L. Yde and Michael G. Vita, “Merger Efficiencies: Reconsidering the ‘Passing-On’ Requirement,” Antitrust Law Journal 64 (1996): 735, 742. 75. Int’l Salt Co., Inc. v. United States, 332 U.S. 392 (1947); see IBM Corp. v. United States, 298 U.S. 131, 138 (1936). The concept in US law is broader than antitrust. See Alan O. Sykes, “The Least Restrictive Means,” University of Chicago Law Review 70 (2003): 403– 19. 76. Council Regulation 4064/89, 1989 O.J. (L 395/1) [1990]. 77. The U.S. Hart-Scott-Rodino Act of 1976 in the United States also requires mergers of a certain size to be examined by the DOJ or FTC. Prior to that time, the US enforcement agencies typically intervened only after the merger was consummated. 78. Commission Notice, O.J. C372 9.12 1997 (definition of relevant market for community competition law). 79. Dennis W. Carlton and William D. Bishop, “Merger Policy and Market Definition under the EC Merger Regulation,” Fordham Corporate Law Institute 422 (1993), edited by Barry E. Hawk, 1994. 80. Article 102 prohibits “abuse” of a “dominant position.” 81. The merger of the third set of two firms would result in a Herfindahl index of 1,600 and an increase of 200 points. According to the 2010 guidelines, this level and increase would “potentially raise significant competitive concerns and often warrant scrutiny,” § 5.3. Earlier guidelines were cast somewhat differently, but all would have directed attention to the third merger. 82. See Società Italiana Vetro SpA and Others v. Commission, [1992] 2 CEC (CCH) 33, 113. Although the court agreed that an Article 82 [102] case could be based upon a theory of collective dominance, the court was not convinced that the Commission had adequately established such a case.

notes to chapter 3

238

83. 1992 O.J. (L 356) 1, [1993] CEC (CCH) 2,018 (1992). 84. Guidelines on the assessment of horizontal mergers under Council Regulation on the control of concentrations between undertakings ¶ 4 (2004/C 31/03) (hereafter “EU Horizontal Guidelines”). 85. EU Horizontal Guidelines ¶¶ 39– 60. 86. U.S. Dep’t of Justice, 1992 Merger Guidelines §§ 0.1, 2.1, Trade Regulation Reporter 4 (CCH) ¶ 13, 104. 87. Article 2 of the Merger Regulation requires the Commission to take into account “the actual or potential competition from undertakings located either within or without the Community” as well as “legal or other barriers to entry.” Merger Regulation, Article 2(1)(a), (b). 88. Id., Article 2(1)(b). 89. Thus in Accor/Wagons-Lits (Decision 92/385), [1992] 1 CEC (CCH) 2,170 (1992), the Commission ruled that because of the demand structure of the motorway restaurant market, the benefits of any efficiencies would not be passed on to consumers. Id. ¶ 26(f). 90. Aérospatiale-Alenia/de Havilland (Decision 91/619), [1992] 1 CEC (CCH) 2,034 (1991). 91. The Commission noted that two competitors of the merged company “expect that the proposed concentration would lead to ATR/de Havilland pursuing a strategy of initially lowering prices so as to eliminate the competitors at least in the key markets of 40 seats and above. . . . Neither Fokker nor British Aerospace consider it possible for them to withstand such a price war. Consequently, both could leave the markets,” Aérospatiale/de Havilland Decision, ¶ 69. This part of the de decision is reminiscent of the now discredited approach of the US Supreme Court in Brown Shoe Co. v. United States, 370 U.S. 294, 344 (1962), where the Court disapproved of the merger in part because the distributional efficiencies resulting from vertical integration would have disadvantaged small unintegrated rival retailers. 92. Thus, for example, the German government has long recognized the welfare-generating potential of vertical restraints while the Commission was much more grudging in its recognition. Robert van den Bergh, “The Subsidiary Principle and the EC Competition Rules: The Costs and Benefits of Decentralization,” in Constitutional Law and Economics of the European Union, ed. Dieter Schmidtchen and Robert Cooter (London: Edward Elgar, 1997), 142, 155. Indeed, in the first cases brought before the European Court of Justice under the Treaty of Rome, the German government had contended that so long as interbrand competition was maintained, consumer welfare would be furthered. Établissements Consten and Grundig-Verkaufs-Gmbh v. Commission, Common Market Law Rep. [1961– 66] (CCH) ¶ 8046, at 7645 (ECJ 1966). That position was only adopted by the US Supreme Court eleven years later. Continental T.V., Inc. v. GTE Sylvania, Inc., 433 U.S. 36 (1977). 93. Meade, “Modeling a European Competition Authority,” 153, 167 n. 78. 94. The EU process parallels that of the United States in that the agencies examine the proposed merger and either give it a green light or make a “second request” for information in the United States or launch a “second stage” investigation in the European Union. Henriette K. B. Andersen, “EC Merger Control Regu-

notes to chapter 3

239

lation as Applied in the De Havilland Case,” New York International Law Review 37, no. 87 (1994): 7, 25, 43. 95. Commission Decision of 2 October 1991 Declaring the Incompatibility with the Common Market of a Concentration (Case No. IV/M.053— AérospatialeAlenia/de Havilland), 1991 O.J. (L 334) 42, ¶¶ 1– 4, 53– 56, 72. 96. Commission Decision of 30 July 1997 Declaring a Concentration Compatible with the Common Market and the Functioning of the EEA Agreement (Case No. IV/M.877— Boeing/McDonnell Douglas), 1997 O.J. (L 336) 16, ¶¶ 46, 68– 71, 116. 97. See FTC, Statement of Commissioner Mary L. Azcuenaga (July 1997), available at 1997 WL 359762. On another reckoning, the contracts may have accounted for 15.29 percent. See Daniel J. Gifford and E. Thomas Sullivan, “Can International Antitrust Be Saved for the Post-Boeing Merger World? A Proposal to Minimize International Conflict and to Rescue Antitrust from Misuse,” Antitrust Bulletin (2000): 45, 55, 76 n. 64. 98. Commission Decision of 3 July 2001 Declaring a Concentration to be Incompatible with the Common Market and the EEA Agreement (Case No. COMP/M.2220— General Electric/Honeywell), 2004 O.J. (L 048) 1. 99. Commission Decision (2004) ¶¶ 353, 378. No satisfactory scenario related to consumer harm was ever offered. See Barry Nalebuff, “Bundling: GE-Honeywell (2001),” in Kwoka and White, The Antitrust Revolution, 388; see also Donna E. Patterson and Carl Shapiro, “Transatlantic Divergence in GE/Honeywell: Causes and Lessons,” Antitrust (2001): 18, 21 (“Based on GE/Honeywell, we must conclude that mergers in the EU may be subject to an efficiency offense whereby they are blocked precisely because they provide incentives for the merged entity to set lower prices”). Just as a merger of firms producing substitutes produces a unilateral price increasing effect, the joining of firms producing complements tends to lower prices— even at the same costs. 100. Patterson and Shapiro, “Transatlantic Divergence.” They cite the Commission at ¶ 349 (“The complementary nature of GE and Honeywell product offerings coupled with their respective existing market positions will give the merged entity the ability and the economically rational incentive to engage in bundled offers or cross-subsidization across product sales to both categories of customers”). See also Patterson and Shapiro, “Transatlantic Divergence,” citing the Commission at ¶¶ 374– 76 (“The Cournot Effect of Bundling”), where it discusses and rejects the parties’ contentions that they lacked incentives to bundle. 101. Patterson and Shapiro, “Transatlantic Divergence.” 102. Gifford and Kudrle, “Rhetoric and Reality”; Nalebuff, “Bundling: GEHoneywell.” More generally, Massimo Motta has pointed out that such portfolio effects can arise either as nonstrategic cost advantages for merging firms or possibly as a weapon for consciously driving rivals from the market. But the former variant could well produce higher consumer surplus at a permanently lower price level for the goods in question. This would be particularly likely if rivals imitated the original “portfolio.” And an attempt to prevent possible future advantage in predation by thwarting immediate cost saving is unlikely to enhance welfare. Motta, Competition Policy, 275– 76.

notes to chapter 3

240

103. General Electric Co. v. Commission of the European Communities Case No. T-210/01, 14 Dec. 2005, at ¶¶ 470– 73. 104. Commission Decision of 24 January 2001 Declaring a Concentration to be Compatible with the Common Market (Case No. COMP/M.2033— Metso/ Svedala) 2001 O.J. (L 088) 1. 105. Commission Decision ¶¶ 137– 39, 201. 106. Commission Decision ¶ 228. 107. Commission Decision ¶ 138. 108. Commission Decision of 9 March 1999 Relating to a Proceeding Under Council Regulation (EEC) 4064/89 (Case IV/M.1313— Danish Crown/VestjyskeSlagterier) 2000 O.J. (L 020) 1. 109. Commission Decision ¶ 198. 110. 1989 Merger Regulation, Article 2(1)(b). 111. 2004 Merger Regulation, art. 2; 1989 Merger Regulation, art. 2. 112. 1989 Merger Regulation, art. 2(3). Similar language is contained in the 2004 Merger Regulation, art. 2(3): “A concentration which would significantly impede effective competition, in the common market or in a substantial part of it, in particular as a result of the creation or strengthening of a dominant position, shall be declared incompatible with the common market.” See also 2004 Merger Regulation, pmbl., cl. 24 (crediting the 1989 Merger Regulation with establishing the principle that “a concentration with a Community dimension which creates or strengthens a dominant position as a result of which effective competition in the common market or in a substantial part of it would be significantly impeded should be declared incompatible with the common market”). 113. Guidelines on the assessment of horizontal mergers under the Council Regulation on the control of concentrations between undertakings (2004/C 31/03). 114. EU Guidelines, at ¶ 17. Generally, the EU Guidelines employ a HerfindahlHirschman Index (HHI) of concentration as a guide to determining competitive effects. EU Guidelines at III. 115. Green Paper on the Review of Council Regulation (EEC) 4064/89, COM(2001)745 final, at 37, ¶ 160. 116. The 2004 Merger Regulation, however, raises the prominance of the “significantly impede competition” clause, thereby signaling the focus of the regulation on competition. 117. Thomas Janssens, “European Merger Reform— More EU-US Convergence Than Meets the Eye?” Mergers and Acquisitions Newsletter 4, no. 2 (2004). 118. Merger Regulation, Articles 2(2) and 2(3). 119. EU Horizontal Guidelines ¶ 4. 120. EU Horizontal Guidelines ¶¶ 39– 60. 121. EU Horizontal Guidelines ¶¶ 24– 38. 122. Merger Regulation, Recital 29. 123. Horizontal Merger Guidelines, paragraph 78. 124. OECD, Portfolio Effects in Conglomerate Mergers (2001). 125. U.S. Dep’t of Justice, Antitrust Division Submission, OECD Roundtable on Portfolio Effects in Conglomerate Mergers, Range Effects: The United States Perspective (2001) (hereafter “DOJ Submission”).

notes to chapter 3

241

126. “DOJ Submission” 4. 127. Commission Decision of 30 October 2001 (Case No. Comp/M. 2416 Tetra Laval/Sidel) 2004 O.J. L 043. 128. Commission Decision at ¶ 345. 129. Tetra Laval BV v. Commission of the European Communities, Case T-5/02, 2002 ECR II-04381. 130. Commission of the European Communities v. Tetra Laval BV, Case C-13/03 P, 2005 ECR I-01113. 131. Tetra Laval, at ¶ 155. 132. Tetra Laval, at ¶ 155. 133. Guidelines on the Assessment of Nonhorizontal Mergers under Council Regulation on the Control of Concentrations between Undertakings (2008/ C 265/07) (hereafter “EU Non-Horizontal Guidelines”). 134. EU Non-Horizontal Guidelines ¶ 55. 135. Commission Decision of 30 March 2009 (Case No. COMP/M.5449— ABF/AZUCARERA) (¶ 72); Commission Decision of 2 July 2008 (Case COMP/ M.4942— Nokia/Navteq) (¶¶ 365– 69). 136. EU Non-Horizontal Guidelines ¶ 101. The provisions of the NonHorizontal Guidelines dealing with bundling appear to describe the foreclosure theory described in Michael Whinston, “Tying, Foreclosure, and Exclusion,” American Economic Review 80, no. 4 (1990): 837, where the anticompetitive effects result from the seller’s commitment to tie, resulting in its expansion in the tied product market, preventing a rival in that market from attaining scale efficiencies. Gregory Werden has criticized guidelines that recite theoretical possibilities of foreclosure without identifying generalizable circumstances where this is likely. Gregory J. Werden, “Should the Agencies Issue New Merger Guidelines? Learning from Experience,” George Mason Law Review 16 (2009): 839, 848. 137. EU Non-Horizontal Guidelines ¶¶ 116, 117. 138. James Langenfeld, “Non-Horizontal Merger Guidelines in the United States and the European Commission: Time for the United States to Catch Up?” George Mason Law Review 16 (2009): 851– 83. 139. Werden, “Should the Agencies Issue New Merger Guidelines?,” 848. 140. U.S. Dep’t of Justice, 1984 Merger Guidelines, Trade Regulation Reporter 4 (CCH) ¶ 13, 103 at 23. 141. In a 2004 action under the 2004 Merger Regulation, the EU Commission ruled that the proposed acquisition of PeopleSoft, Inc. by the Oracle Corporation would be lawful. Press Release, EU Commission, Commission Clears Oracle’s Takeover Bid for PeopleSoft (Oct. 26, 2004), available at http://europa.eu.int /rapid/pressReleasesAction.do?reference=IP/04/1312. Although the acquisition would reduce the number of major firms providing high-function human resources processes software and financial planning and reporting software to large and complex enterprises from three to two, there were in fact more actual or potential suppliers of that software. In so ruling, the Commission reached the same conclusion as did a US court. United States v. Oracle Corp., 331 F.Supp.2d 1098 (N.D. Cal. 2004). Both the EU Commission and the US court decisions were based on an assessment of the relevant market. Although the efficiencies defense Oracle

notes to chapter 3

242

raised was not critical to the US decision, the US court rejected it. 331 F.Supp.2d at 1175. 142. Case COMP/M.4000 (Inco/Falconbridge). 143. Case COMP/M.4439 (Ryanair/Aer Lingus). 144. In February 2011 NYSE Euronext and Deutsche Börse AG, two of the largest financial exchanges, agreed to merge. NYSE Euronext operates the New York Stock Exchange, NYSE Arca (an all-electronic exchange), and NYSE Amex. Deutsche Börse, through subsidiaries, is the largest stockholder in Direct Edge, a company that through a subsidiary operates two US stock exchanges that compete with the NYSE Euronext exchanges. The proposed merger was reviewed by the US Justice Department, which concluded that the merger raised antitrust concerns because of competition between Direct Edge’s exchanges and those of NYSE Euronext. The Justice Department’s position was that the relevant markets within the United States were not affected by competition elsewhere. The Department withdrew its objections to the merger when the parties agreed to divest Direct Edge. The European Commission took the view that trading in derivatives was dominated by NYSE Euronext and Deutsche Börse, and, contrary to the contentions of the parties, the market was not global but European. On this reasoning, the Commission disapproved the merger. But as a DOJ official explained, US approval contingent on a divestiture and EU blockage of the merger as a whole came after nearly a year of close transatlantic discussion and simply reflects “different competitive conditions in the United States than in Europe” and not a conflict between jurisdictions. Rachel Brandenburger, “Recent Developments in Merger Control: Views from the U.S. Department of Justice’s Antitrust Division,” Remarks as Prepared for the International Bar Association’s 16th Annual Competition Conference, Florence, Italy, September 14, 2012. 145. Richard Gilbert and Daniel L. Rubinfeld, “Revising the Horizontal Merger Guidelines: Lessons from the US and the EU,” in Competition Policy and Regulation: Recent Developments In China, the US, and Europe, ed. Michael Faure and Xinzhu Zhang (Cheltenham, UK: Edward Elgar, 2011). 146. Potential competition issues are nominally horizontal but often seemingly speculative. 147. Rachel Brandenburger, “Transatlantic Antitrust Past and Present,” St. Gallen International Competition Forum, May 2010. An ambitious approach to the problem might involve a joint US-EU merger board to supersede national authorities on mergers that have a strong impact on both sides of the Atlantic, but this appears to lack all political feasibility for the foreseeable future. See David Snyder, “Mergers and Acquisitions in the European Community and the United States: A Movement toward a Uniform Enforcement Body,” Law and Policy in International Business 29 (1997): 115, 143– 44. 148. U.S.-EU Merger Working Group, Best Practices on Cooperation in Merger Investigation, http://ec.europa.eu/competition/international/bilateral/eu_us.pdf. 149. Rachel Brandenburger and Randy Tritell, “Global Antitrust Policies: How Wide is the Gap? Interview,” Concurrences 1 (2012): 3– 11. 150. Nihat Aktas, Eric de Bodt, Marieke Delanghe, and Richard Roll, “Market Reactions to European Merger Regulation: A Reexamination of the Protectionism Hypothesis” (2011); electronic copy available at http://ssrn.com/abstract=1961188.

notes to chapter 4

243

ChApter 4 1. Guidance on the Commission’s Enforcement Priorities in Applying Article 82 of the EC Treaty to Abusive Exclusionary Conduct by Dominant Undertakings, 2009/C 45/02. 2. Benjamin Klein and Kevin Murphy, “Exclusive Dealing Intensifies Competition for Distribution,” Antitrust Law Journal 75 (2008): 433– 66. 3. Clayton Act ch. 323, § 2, 38 Stat. 730 (1914) (current version at 15 U.S.C. § 13 (2006)). 4. Robinson-Patman Act of June 19, 1936, ch. 592, §§ 1– 4, 49 Stat. 1526 (codified at 15 U.S.C. §§ 13(a)– (b), 21a (2006)). 5. See Civil Aeronautics Act of 1938, ch. 601, 52 Stat. 973 (1938); Motor Carrier Act of 1935, ch. 498, 49 Stat. 543 (1935); Act to Regulate Commerce, ch. 104, 24 Stat. 379 (1887). 6. Consolidated Version of the Treaty on the Functioning of the European Union, art. 101(1)(d), 102(c)(2012). 7. Aradhna Aggarwal, “Patterns and Determinants of Anti-Dumping: A World-Wide Perspective,” Indian Council for Research on International Economic Relations, Working Paper No. 113, 2003, 25– 30, available at http://icrier.org/pdf /wp113.pdf. Before the 1990s, the United States, the European Union, Canada, and Australia had been the primary users of antidumping laws. Since the early 1990s, developing countries have begun to employ antidumping laws to protect their markets. In 1993, for example, antidumping laws were employed by Argentina, Brazil, India, Korea, Mexico, and South Africa. Christopher F. Corr, “Trade Protection in the New Millennium: The Ascendancy of Antidumping Measures,” Northwestern Journal of International Law and Business 18, no. 1 (1997): 49, 55– 56; Wei Huo, “Introduction and Critical Analysis of Anti-Dumping Regime and Practice in China Pending Entry of WTO: Transition toward a WTO-Modeled Trade Legal Mechanism,” International Law 36 (2002): 197, 198; Meredith Schutzman, Note, “Antidumping and the Continued Dumping and Subsidy Offset Act of 2000: A Renewed Debate,” Cardozo Journal of International and Comparative Law 11 (2004): 1069, 1077– 78. 8. For example, the original Clayton Act § 2 was enacted in 1914. As early as 1887, the Congress enacted the Interstate Commerce Act condemning discrimination in railroad rates. The leading economic work on discrimination, Arthur C. Pigou, The Economics of Welfare, was published in 1920. Jules Dupuit was analyzing discrimination as a tool for paying for public works in the mid-nineteenth century. See Jules Dupuit, On Tolls and Transport Charges 7, trans. Elizabeth Henderson, International Economic Papers No. 11, 1962. 9. Pigou identified three categories of price discrimination and assessed their effects: first-degree, second-degree, and third-degree price discrimination. First-degree price discrimination involves charging every customer the maximum amount he is willing to pay for each unit of the product sold. This removes all “consumer surplus,” the usual excess of value that people obtain from buying multiple units of a good at a fixed price. In theory and if perfectly carried out, however, first-degree price discrimination would generate no deadweight loss. First-degree discrimination can only be roughly approximated and only occurs in specialized circumstances. Second-degree price discrimination is the practice of setting two or

notes to chapter 4

244

more prices for a good, depending on the amount purchased but allowing each purchaser to face the same schedule. The most familiar variant is the two-part tariff where there is an entrance fee into the market followed by a single price for all units purchased. Two- or multipart or pricing is often used to increase output in a regulated monopoly while allowing total costs to be covered by total revenues. Buyers may not all face the same price for marginal purchases under second-degree price discrimination, however, and this generates allocative inefficiencies. In third-degree price discrimination, a seller identifies separable market segments, each of which possesses its own demand for its product. Assuming that some of the good is sold in every market at the simple monopoly price, Joan Robinson demonstrated that a monopolist’s output remains constant with linear demand curves when the monopolist discriminates. This is done by setting a price in each submarket in accordance with that segment’s demand elasticity. Further, she notes that under third-degree discrimination, output is misallocated and welfare is reduced by comparison with sales at a single monopoly price because purchasers do not face the same price for their purchasers. Where a product is sold in both markets at a single price, overall welfare can improve under discrimination only with nonlinear demand and only if output expands enough to outweigh the inefficiency of different marginal prices. When one (more elastic) market is not served at simple monopoly price, however, discrimination increases welfare even with linear demand curves. This is because discrimination in this case leaves the “stronger” market unchanged and output expands in the “weaker” market (with constant marginal cost). 10. Stigler, “A Theory of Oligopoly,” Journal of Political Economy 72 (1964): 44, 47. 11. See Kenneth Corts, “Third-Degree Price Discrimination in Oligopoly: AllOut Competition and Strategic Commitment,” RAND Journal of Economics 29, no. 2 (1998): 306–23. 12. Stigler, “A Theory of Oligopoly,” 47. 13. Id. 14. Baumol and Swanson, “The New Economy,” 661– 62. 15. Id., 665. 16. Clayton Act ch. 323, § 2, 38 Stat. 730 (1914) (current version at 15 U.S.C. § 13 (2006)). 17. See HR Rep. No. 63-627, at 8– 9 (1914): “The necessity for legislation to prevent unfair discriminations in prices with a view of destroying competition needs little argument to sustain the wisdom of it. In the past it has been a most common practice of great and powerful combinations engaged in commerce— notably the Standard Oil Co., and the American Tobacco Co., and others of less notoriety, but of great influence— to lower prices of their commodities, oftentimes below the cost of production in certain communities and sections where they had competition, with the intent to destroy and make unprofitable the business of their competitors, and with the ultimate purpose in view of thereby acquiring a monopoly in the particular locality or section in which the discriminating price is made. Every concern that engages in this evil practice must of necessity recoup its losses in the particular communities or sections where their commodities are sold below cost or without a fair profit by raising the price of this same class of commodities

notes to chapter 4

245

above their fair market value in other sections or communities.” Identical language is found in the Senate Report. S. Rep. No. 63-698, at 2– 4 (1914). 18. HR Rep. No. 63-627 at 9; S. Rep. No. 63-698 at 3. 19. Id. 20. See John S. McGee, “Predatory Price Cutting: The Standard Oil (N.J.) Case,” Journal of Law and Economics 1 (1958): 137, 143. 21. Antidumping Act of Sept. 8, 1916, ch. 463, §§ 801– 2, 39 Stat. 798 (1916), repealed by Pub. L. No. 108-429, tit. 2, § 2006(a), 118 Stat. 2597 (2004). 22. Antidumping Act of 1921, ch. 14, § 201, 42 Stat. 11 (1921), repealed and substantially reenacted by Trade Agreements Act of 1979, 93 Stat. 162 (1979) (codified at 19 U.S.C. § 1673 (2006)). 23. Thus the 1916 act required proof that the seller sold the articles in question “with the intent of destroying or injuring an industry in the United States, or of preventing the establishment of an industry in the United States, or of restraining or monopolizing any part of trade and commerce in such articles in the United States.” Antidumping Act of Sept. 8, 1916, ch. 463, §§ 80102, 39 Stat. 798 (1916), repealed by Pub. L. No. 108-429, tit. 2, § 2006(a), 118 Stat. 2597 (2004)). 24. See Antidumping Act of 1921 § 201. 25. See Daniel J. Gifford, “Rethinking the Relationship between Antidumping and Antitrust Laws,” American University Journal of International Law and Policy 6 (1991) 227, 299– 300. 26. See Antidumping Act of 1921 § 201. 27. Robinson-Patman Act of June 19, 1936, ch. 592, §§ 14, 49 Stat. 1526 (codified at 15 U.S.C. §§ 13(a)– (b), 21a (2006)). 28. See, for example, Frederick M. Rowe, Price Discrimination under the Robinson-Patman Act (Boston: Little, Brown, 1962), 1113; see also Hugh C. Hansen, “Robinson-Patman Law: A Review and Analysis,” Fordham Law Review 51 (1983): 1113, 1122. 29. See, for example, “Federal Trade Commission, Final Report on the Chain Store Investigation,” S. Doc. No. 74-4, at 6671 (1st Sess. 1935) (Using gross margins as measure of efficiency and reporting lower gross margins for both grocery and drug chains) (hereinafter “Chain Store Report”). 30. For an example of the discussion of A&P’s purchasing strategy involving differentiated products, see M. A. Adelman, A&P: A Study in Price-Cost Behavior and Public Policy (Cambridge, MA: Harvard University Press, 1959), 140– 46; see also “Chain Store Report,” 24. 31. “Chain Store Report,” 9. 32. See FTC v. Morton Salt Co., 334 U.S. 37, 43 (1948) (“The legislative history of the Robinson-Patman Act makes it abundantly clear that Congress considered it to be an evil that a large buyer could secure a competitive advantage over a small buyer solely because of the large buyer’s quantity purchasing ability”). 33. Dagher v. Saudi Refining, Inc., 369 F.3d 1108, 1123 (9th Cir. 2004) (quoting Alan’s of Atlanta, Inc. v. Minolta Corp., 903 F.2d 1414, 1422 (11th Cir. 1990)), rev’d, 547 U.S. 1 (2006). 34. See Adelman, A&P, 10– 12. 35. Jack Triplett and Barry Bosworth, Productivity in the U.S. Services Sectors:

notes to chapter 4

246

New Sources of Economic Growth (Washington DC, Brookings Institution Press, 2004). 36. Legal and common usage equates price discrimination with a price difference. FTC v. Anheuser-Busch, Inc., 363 U.S. 536, 549 (1960) (“a price discrimination . . . is merely a price difference”). See also Texaco Inc. v. Hasbrouck, 496 U.S. 543, 558– 59 (1990). When economists use the term, however, they mean that two or more similar goods are being sold at prices that bear different ratios to their marginal costs. George J. Stigler, The Theory of Price (London: Macmillan, 1966), 299. 37. John K. Galbraith, American Capitalism: The Concept of Countervailing Power (Boston: Houghton Mifflin, 1952). 38. 334 U.S. 37, 54– 55 (1948). 39. See, for example, Rowe, Price Discrimination, 4. 40. Utah Pie Co. v. Cont’l Baking Co., 386 U.S. 685, 702– 3 (1967); Samuel H. Moss, Inc., v. FTC, 148 F.2d 378, 379– 80 (2d Cir. 1945), cert. denied, 326 U.S. 734 (1945); In re Maryland Baking Co., 52 F.T.C. 1679, 1683– 84 (1956), modified and aff’d, 243 F.2d 716 (4th Cir.1957), order modified, 53 F.T.C. 1106 (1957). 41. 15 U.S.C. § 13(a) (2006). 42. In re Morton Salt Co., 39 F.T.C. 35, 42– 43 (1944), modified, 40 F.T.C. 388 (1945), order vacated, 162 F.2d 949 (7th Cir. 1947), rev’d, 334 U.S. 37 (1948); HR Rep. No. 74-2287, at 7 (1936); S. Rep. No. 74-1502, at 4– 6 (1936); Daniel J. Gifford, “Assessing Secondary-Line Injury under the Robinson-Patman Act: The Concept of ‘Competitive Advantage,’” George Washington Law Review 44 (1975): 48– 51. 43. See, for example, Utah Pie Co. v. Cont’l Baking Co, note 40 above. 44. See, for example, Ward S. Bowman, “Restraint of Trade by the Supreme Court: The Utah Pie Case,” Yale Law Journal 77 (1967): 70 (protesting the Supreme Court’s ruling in Utah Pie as antagonistic to competition understood as striving by firms to capture business from their rivals by underpricing them). 45. Samuel H. Moss, Inc., 36 F.T.C. 640, 648 (1943), aff’d, Samuel H. Moss, Inc. v. FTC, 148 F.2d 378, 379 (2d Cir. 1945), cert. denied, 326 U.S. 734 (1945). 46. Samuel H. Moss, Inc. v. FTC, 148 F.2d 378, 379 (2d Cir. 1945), cert. denied, 326 U.S. 734 (1945). 47. Id. The Commission took a similar approach in In re Anheuser-Busch, Inc., 54 F.T.C. 277, 300 (1957), order set aside, 265 F.2d 677 (7th Cir. 1959), rev’d, 363 U.S. 536 (1960), remanded to 289 F.2d 835 (7th Cir. 1961). 48. See Samuel H. Moss, Inc., 148 F.2d at 379. The presumption the FTC and the Second Circuit used in Moss made possible the proof of a primary-line case (a case in which competitors of the discriminating seller were adversely affected) through proof only of discrimination. In FTC v. Morton Salt Co., 334 U.S. 37, 46– 47 (1948), the Supreme Court authored a presumption that was directed toward proof of a secondary-line case (a case in which the purchasers or their customers were adversely affected) through proof only of discrimination in a “substantial” amount and proof that the favored and disfavored purchasers were in competition for the resale of the goods involved. 49. Samuel H. Moss, Inc., 148 F.2d at 379. 50. See Daniel J. Gifford, “Primary-Line Injury under the Robinson-Patman

notes to chapter 4

247

Act: The Development of Standards and Erosion of Enforcement,” Minnesota Law Review 64, no. 1 (1979): 69. 51. See Anheuser-Busch, 54 F.T.C., 300. 52. See id. 53. See Standard Oil Co. v. FTC, 340 U.S. 231, 249– 50 (1951); Cont’l Baking Co. v. Old Homestead Bread Co., 476 F.2d 97, 104 (10th Cir. 1973), cert. denied, 414 U.S. 975 (1973); United States v. N.Y. Great Atl. & Pac. Tea Co., 67 F.Supp. 626, 671 (E.D. Ill. 1946), aff’d, 173 F.2d 79 (7th Cir. 1949); United Fruit Co., 82 F.T.C. 53, 151– 54 (1973), aff’d in part, rev’d in part sub nom., Harbor Banana Distrib., Inc. v. FTC, 499 F.2d 395 (5th Cir. 1974); Forster Mfg. Co., 62 F.T.C. 852, 902 (1963), vacated and remanded, 335 F.2d 47 (1st Cir. 1964); C. E. Nieoff & Co., 51 F.T.C. 1114, 1126 (1955), modified and aff’d, 241 F.2d 37 (7th Cir. 1957), order reinstated, aff’d sub nom., Moog Indus., Inc. v. FTC, 355 U.S. 411 (1958); see also Daniel J. Gifford, “Promotional Price-Cutting and Section 2(a) of the Robinson-Patman Act,” Wisconsin Law Review (1976): 1045, at 1076– 77 and n. 126. 54. Utah Pie Co. v. Cont’l Baking Co., 386 U.S. 685 (1967). 55. Utah Pie 703 & n. 14. 56. Utah Pie 691– 92 n. 7 (tables). 57. For a discussion of the phases of US competition policy that considers this transition, see Daniel J. Gifford and Robert T. Kudrle, “Alternative National Merger Standards and the Prospects for International Cooperation,” in The Political Economy of International Trade Law: Essays in Honor of Robert E. Hudec, ed. Daniel L. M. Kennedy and James D. Southwick (Cambridge: Cambridge University Press, 2002). 58. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 U.S. 209, 219– 27 (1993) (effectively overruling Utah Pie Co. v. Cont’l Baking Co., 386 U.S. 685 (1967)). 59. Brown & Williamson, 509 U.S. at 222. 60. See text above at notes 16– 20. 61. Brown & Williamson, 509 U.S. at 222. 62. See United States v. United States Gypsum Co., 438 U.S. 422, 458– 59 (1978); see also Automatic Canteen Co. of America v. FTC, 346 U.S. 61, 74 (1953) (upholding Sherman Act policies over potentially conflicting Robinson-Patman Act policies). In the 2006 Simco decision finding Volvo not in violation of the RobinsonPatman Act in its differential treatment of dealers, Justice Stevens’s dissenting opinion hinted that the myriad technical arguments made by the majority could mask antipathy to the substance of the law: “[Although] I do not suggest that disagreement with the policy of the Act has played a conscious role in my colleagues’ unprecedented decision today.” Volvo Trucks North America, Inc. v. Reeder-Simco GMC, Inc., 546 U.S. 164, 188 (2006). Stevens’s dissent (for himself and Thomas) pointedly noted that “the exceptional quality of this case provides strong reason to enforce the Act’s prohibition against discrimination even if Judge Bork’s evaluation [that the law was based on ‘wholly mistaken economic theory’] (with which I happen to agree) is completely accurate.” Id. One inference from all of this is that the entire Supreme Court rejects the policy contained in the Robinson-Patman Act, but members differ in how that rejection should be expressed.

notes to chapter 4

248

63. See Scott Martin and Irving Scher, “The Robinson-Patman Act: Sellers’ and Buyers’ Violations and Defenses,” 1649 PLI/Corp. 553, 561 (2008) (reporting that FTC had brought an average of forty Robinson-Patman cases per year from 1937 to 1971, that after 1980, FTC instituted only one to two cases per year, and that currently there are no Robinson-Patman cases on the Commission’s docket). 64. AMC Report and Recommendations, 312, 317. 65. See Boise Cascade Corp. v. FTC, 837 F.2d 1127, 1139– 40 (D.C. Cir. 1988); Richard Short Oil Co. v. Texaco, Inc., 799 F.2d 415, 420 (8th Cir. 1986). 66. AMC Report and Recommendations, 316. 67. See Feeser’s, Inc. v. Michael Foods, Inc., 498 F.3d 206, 216 (3d Cir. 2007); Freightliner of Knoxville, Inc. v. DaimlerChrysler Vans, LLC, 484 F.3d 865, 871– 74 (6th Cir. 2007); Lewis v. Philip Morris, Inc., 355 F.3d 515, 534 (6th Cir. 2004); Caribe BMW, Inc. v. Bayerische Moteren Werke Aktiengesellschaft, 19 F.3d 745, 748– 52 (1st Cir. 1994); DeLong Equip. Co. v. Washington Mills Abrasive Co., 887 F.2d 1499, 1515– 17 (11th Cir. 1989); Krist Oil Co. v. Bernick’s Pepsi-Cola of Duluth, Inc., 354 F.Supp.2d 852, 858 (W.D. Wis. 2005); Allied Sales & Serv. Co. v. Global Indus. Techs., Inc., 2000-1 Trade Cas. (CCH) ¶ 72, 953, No. 97-0017-CB-M, 2000 U.S. Dist. LEXIS 7774, at *31– 34 (S.D. Ala. May 1, 2000); Calumet Breweries, Inc. v. G. Heileman Brewing Co., 951 F.Supp. 749, 753– 756 (N.D. Ind. 1994). 68. EC Treaty art. 101(1)(d). 69. EC Treaty art. 102(c). 70. Thus, for example, in a contemporary discussion of antitrust policy, Kaysen and Turner, Antitrust Policy, although criticizing the rigidities of the RobinsonPatman Act, nonetheless contemplated that a prohibition of price discrimination should be a part of the antitrust laws. Indeed, these authors provided a model for legislation prohibiting price discrimination that they believed was superior to the Robinson-Patman Act. See Kaysen and Turner, Antitrust Policy, 184– 85. 71. Gifford, “Assessing Secondary-Line Injury under the Robinson-Patman Act,” 48. 72. See Case C-73/95, Viho Europe BV v. Comm’n, E.C.R. I-5457, ¶ 61 (1996). In that decision Viho complained that a supplier (Parker) offered it discriminatorily unfavorable supply prices. The Court rejected that contention on the ground that the discrimination at which Article 101(1)(d) is directed cannot be “the result of unilateral conduct by a single undertaking.” Id. See S. O. Spinks, “Exclusive Dealing, Discrimination, and Discounts under EC Competition Law,” Antitrust Law Journal 67, no. 3 (2000): 641– 70, at 668– 69; Michel Waelbroeck, “Price Discrimination and Rebate Policies under EU Competition Law,” Fordham Corporate Law Institute (1995): 147, 149. 73. Case 27/76, United Brands Co. v. Comm’n, E.C.R. 207, ¶ 300 (1978). 74. See Damien Geradin and Nicolas Petit, “Price Discrimination under EC Competition Law: Another Antitrust Doctrine in Search of Limiting Principles?” Journal of Competition Law and Economics 2 (2006): 479– 531, at 522– 23. 75. See joined Cases 56/64 and 58/64, Consten and Grundig v. Comm’n, E.C.R. 299, ¶ 341 (1966). 76. Case 85/76, Hoffmann-La Roche & Co. v. Comm’n, E.C.R. 461, ¶ 90 (1979). 77. Id. 78. See Geradin and Petit, “Price Discrimination,” 525.

notes to chapter 4

249

79. Case T-219/99, British Airways plc v. Comm’n, E.C.R. II-5917, ¶¶ 246– 47 (2003). 80. Id. 81. See Case 322/81, NV Michelin v. Comm’n, E.R.C. 3461, ¶¶ 81– 82 (1983); see also Commission Decision 92/163/EEC, Tetra Pak II, 1992 O.J. (L 72/1) 174 (1991) (barring aggregation on quantity discounts). 82. See NV Michelin, E.R.C. 3461 at ¶¶ 86, 91. 83. British Airways, E.C.R. II-5917 at ¶¶ 244– 45, 248. 84. Id. 85. Case C-95/04 P, British Airways plc v. Comm’n, 2007 E.C.R. I-02331. 86. See Damien Geradin and Nicolas Petit, “Price Discrimination under EC Competition Law: The Need for a Case-by-Case Approach,” 9, Global Competition Law Centre, Working Paper No. 07/05, 2007, available at http://www .coleurope.eu/content/gclc/documents/GCLC%20WP%2007-05.pdf. 87. Case 85/76, Hoffmann-La Roche & Co. v. Comm’n, E.C.R. 461, ¶ 90 (1979); Case 27/76, United Brands Co. v. Comm’n, E.C.R. 207, ¶ 300 (1978); see also Case T-83/91, Tetra Pak International SA v. Comm’n, E.C.R. II-755, 4 C.M.L.R. 726, at ¶ 173 (1994). 88. See previous note. 89. See Brooke Group Ltd. v. Brown and Williamson Tobacco Corp., 509 U.S. 209, 209 (1993). 90. Case C-62/86, Akzo Chemie BV v. Comm’n, 1991 E.C.R. I-03359, 5 C.M.L.R. 215 (1993), ¶ 115; Case T-30/89, Hilti AG v. Comm’n, 1991 E.C.R. II-01439, 4 C.M.L.R. 16, ¶ 100 (1992); Case T-228/97, Irish Sugar plc v. Comm’n, 1999 E.C.R. II-02969, 5 C.M.L.R. 1300 (1999), ¶¶ 124, 259, 261 (recognizing dominant firm’s selective price-cutting as abusive but finding failure of proof). In Case C-395/96 P, Compagnie Mar. Belge Transps. SA v. Comm’n, 2000 E.C.R. I-1365, 4 C.M.L.R. 1076 (2000), the losing defendant argued lawfulness of selective pricecutting. See the advocate general’s presentation to the Court at ¶ 114. The Court decided otherwise. See 4 C.M.L.R. at ¶¶ 117– 21. 91. See, for example, Compagnie, 2000 E.C.R. I-1365 at ¶ 132. 92. Akzo, 1991 E.C.R. I-03359. 93. Hilti, 1991 E.C.R. II-01439. 94. 88/138/EEC: Commission Decision of 22 Dec. 1987 relating to a proceeding under Article 86 of the EEC Treaty (IV/30.787 and 31.488— Eurofix-Bauco v. Hilti) at ¶¶ 80– 81. 95. Case C-395/96 P, Compagnie Mar. BelgeTransps.SA v. Comm’n, 2000 E.C.R. I-1365, 4 C.M.L.R. 1076 (2000) at ¶¶ 97, 118– 20. 96. Post Danmark A/S v. Konkurrencerådet Case C-209/10 (27 March 2012). 97. Post Danmark at ¶ 29. 98. Post Danmark at ¶ 30. 99. Post Danmark at ¶ 37. 100. Discussion Paper at ¶¶ 121, 128, 140, 141, 179. 101. Guidance at ¶ 20 and n. 3. In Irish Sugar, the Commission determined again that a policy of selective low prices to customers of its competitor constituted an abuse albeit without a finding whether these low prices were above or below cost. Commission Decision of 14 May 1997 relating to a proceeding pursuant to

notes to chapters 4–5

250

Article 86 of the EC Treaty (IV/34.621, 35.059/F-3— Irish Sugar plc) at ¶ 123. On review, the General Court vacated this aspect of the Commission’s decision, not on the law (i.e., whether selectively low prices to customers of a competitor constitute abusive conduct) but on the facts. According to the General Court, there was evidence that the company contemplated this conduct but with no evidence that it actually carried out the plan. Case T-228/97 Irish Sugar plc v. Comm’n, 199 E.C.R. II-02969 at ¶¶ 121, 124. Another aspect of Irish Sugar concerned discriminatory rebates condemned under Article 102(c) which, according to the General Court, the company financed from higher prices elsewhere. This aspect is of interest to the discussion of selectively low prices because the guidance references that part of the General Court’s decision in its discussion of selectivity. 102. Case C-62/86, Akzo Chemie BV v. Comm’n, 1991 E.C.R. I-03359, 5 C.M.L.R. 215 (1993), ¶ 71. 103. Akzo at ¶ 72. 104. See, for example, Akzo 1991 E.C.R. I-03359 at ¶¶ 114– 15; Hilti, 1991 E.C.R. at ¶¶ 100– 101; Irish Sugar 1999 E.C.R. at ¶ 203; Compagnie, 2000 E.C.R. I-1365 at ¶¶ 117– 21. 105. Akzo, 1991 E.C.R. I-03359 at ¶ 115; see also Compagnie, 2000 E.C.R. I-1365 at ¶ 128. 106. Case C-333/94 P, Tetra Pak Int’l SA v. Comm’n, 1996 E.C.R. I-5951, 4 C.M.L.R. 662 (1997), ¶ 74; Akzo, 1991 E.C.R. I-03359 at ¶¶ 70– 71.

ChApter 5 1. HR Rep. No. 63-627, at 8– 9 (1914). Identical language is contained in the Senate Report. S. Rep. No. 63-698, at 2– 4 (1914). 2. 386 U.S. 685, 697, 702 and n. 14 (1967). 3. Phillip Areeda and Donald F. Turner, “Predatory Pricing and Related Practices under Section 2 of the Sherman Act,” Harvard Law Review 88 (1975): 697– 733. 4. Areeda and Turner, “Predatory Pricing,” 699. 5. Id., 698– 99. 6. See, for example, Oliver E. Williamson, “Predatory Pricing: A Strategic and Welfare Analysis,” Yale Law Journal 87 (1977): 284; Paul L. Joskow and Alvin K. Klevorick, “A Framework for Analyzing Predatory Pricing Policy,” Yale Law Journal 89 (1979): 213; Janusz A. Ordover and Robert D. Willig, “An Economic Definition of Predation: Pricing and Product Innovation,” Yale Law Journal 91 (1981): 8; Frank H. Easterbrook, “Predatory Strategies and Counterstrategies,” University of Chicago Law Review 48 (1981): 263. 7. 698 F.2d 1377, 1387 (9th Cir. 1983). 8. Arthur S. Langenderfer, Inc. v. S. E. Johnson Co., 729 F.2d 1050, 1056 (6th Cir. 1984). 9. McGahee v. Northern Propane Gas Co., 858 F.2d 1487, 1503 (11th Cir. 1988). The Eleventh Circuit cast its ruling in language different from the Ninth and Sixth Circuits but acknowledged that its approach to pricing between average total cost and average variable cost did not differ substantively from those circuits. McGahee v. Northern Propane Gas Co, at n. 37.

notes to chapter 5

251

10. 858 F.2d at 1503. 11. Matsushita Elec. Indus. Co. v. Zenith Radio Corp., 475 U.S. 574, 588– 89 (1986). 12. Brooke Group Ltd. v. Brown and Williamson Tobacco Corp., 509 U.S. 209, 226 (1993); Cargill, Inc. v. Montfort of Colo., Inc., 479 U.S. 104, 121 n. 17 (1986). 13. 475 U.S. at 597 (1986). See also Brooke Group Ltd. v. Brown and Williamson Tobacco Corp., 509 U.S. 209, 227 (1993). 14. Patrick Bolton, Joseph F. Brodley, and Michael H. Riordan, “Predatory Pricing: Strategic Theory and Legal Policy,” Georgetown Law Journal 88 (2000): 2258. 15. See Aaron Edlin, “Predatory Pricing,” in Research Handbook on the Economics of Antitrust Law, ed. Einer Elhauge (London: Edward Elgar, 2011), 144– 73. 16. 549 U.S. 312 (2007). 17. William J. Baumol, “Predation and the Logic of the Average Variable Cost Test,” Journal of Law and Economics 39 (1996): 49– 72. 18. Baumol, “Predation,” 58– 59. 19. Baumol, “Predation,” 60– 61. 20. In United States v. AMR Corp. (American Airlines), 335 F.3d 1109 (10th Cir. 2003), the government charged American Airlines with predatory pricing on four routes out of its Dallas/Fort Worth hub. The critical issue was whether American’s capacity additions on those routes brought its costs to a level higher than its revenues from those routes. 21. US Department of Justice Report, “Competition and Monopoly: SingleFirm Conduct under Section 2 of the Sherman Act,” 65– 67 (2008). 22. Justice Department Withdraws Report on Antitrust Monopoly Law, Antitrust Division to Apply More Rigorous Standard with Focus on the Impact of Exclusionary Conduct on Consumers, May 11, 2009, www.usdoj.gov. 23. There remain nonpredatory dynamic reasons for below-cost pricing, notably learning by doing and the development of network effects. 24. Bolton, Brodley, and Riordan, “Predatory Pricing,” 2239. The Bolton, Brodley, and Riordan article was praised and extensively cited in a 2005 discussion paper issued by the Office of the Chief Economist, DG Competition, European Commission: Miguel de la Mano and Benolt Durand, “A Three-Step Structured Rule of Reason to Assess Predation under Article 82 (12 Dec 2005).” 25. Laker Airways v. Sabena, Belgian World Airlines, 731 F.2d 909, 917, 955 (D.C. Cir. 1984) involved allegations of financial predation. The DC Circuit viewed the allegations as stating a valid cause and ordered the case to go forward despite objections of the UK government based on considerations of international comity. 26. Laker Airways Ltd. v. Sabena, Belgian World Airlines, 731 F.2d 909, 917 (D.C. Cir. 1984). 27. Guidance on Article 82, ¶ 26. 28. Baumol, “Predation,” 58. 29. Bolton, Brodley, and Riordan, “Predatory Pricing,” 2272 and n. 180. 30. The problem comes when a commentator simply uses “incremental” cost without modification; this is innately ambiguous. 31. Rebel Oil Co., 146 F.3d 1088, 1095 (9th Cir. 1998); In re IBM Peripheral

notes to chapter 5

252

EDP Devices Antitrust Litigation, 459 F.Supp. 626, 631 (N.D. Cal. 1978); Continental Airlines, Inc., v. American Airlines, Inc., 824 F.Supp. 689, 701 (S.D. Texas 1993). 32. Spirit Airlines, Inc. v. Northwest Airlines, Inc., 431 F.3d 917, 949– 53 (6th Cir. 2005). 33. Baumol, “Predation,” 71. 34. United States v. AMR Corp., 335 F.3d 1109 (10th Cir. 2003). In that case, American responded to challenges by a number of low-cost airlines on several routes from Dallas/Fort Worth, an American hub, by shifting capacity to that airport. The Justice Department brought suit against American, alleging predatory pricing. 35. 335 F.3d at 1120. 36. Commission Decision of 14 December 1985 (IV/30.698— ECS/Akzo). 37. Commission Decision at ¶ 75. 38. Commission Decision at ¶ 77. 39. Commission Decision at ¶ 76. 40. Commission Decision at ¶ 81. 41. Commission Decision at ¶ 81. 42. Commission Decision at ¶ 77. In support of this economic analysis, the Commission cites B. S. Yamey, “Predatory Price Cutting: Notes and Comments,” Journal of Law and Economics 15 (1972): 129, 133. 43. Akzo Chemie BV v. Commission of the European Communities, Case C-62/86, [1991] ECR I-03359. 44. Akzo Chemie BV v. Commission at ¶ 64. 45. Akzo Chemie BV v. Commission at ¶ 71. 46. Akzo Chemie BV v. Commission at ¶ 72. 47. Tetra Pak International SA v. Commission of the European Communities, Case C-333/94, [1996] ECR I-05951. 48. Tetra Pak International SA v. Commission at ¶ 42. 49. Tetra Pak International SA v. Commission at ¶ 41. 50. Tetra Pak International SA v. Commission at ¶ 42 (ECJ reciting finding of General Court). 51. Tetra Pak International SA v. Commission at ¶ 44. Tetra Pak was found to be dominant in the aseptic carton market, although not in the nonaseptic carton market. It contended that because of its lack of dominance in the nonaseptic market, it could not recoup its losses in that market. The Court of Justice answered this contention by ruling that likelihood of recoupment was not a constitutive element of predatory pricing. Tetra Pak International SA v. Commission at ¶ 44. After ruling that it was unnecessary to show that Tetra Pak had a realistic chance of recouping its losses, the Court switched its focus from recoupment to the elimination of competitors. On that issue, the Court stated that it was unnecessary to wait to see whether competitors were actually eliminated. In further response to Tetra Pak’s contention that it lacked dominance in the nonaseptic carton market, the Court also ruled that Tetra Pak held a dominant position on the aggregate of the markets in which it was operating. Tetra Pak International SA v. Commission at ¶ 2. 52. Tetra Pak International SA v. Commission of the European Communities, Case T-83/91, [1994] ECR II-00755 at ¶ 150. The ECJ followed this ruling of the General Court. See note 51, supra.

notes to chapter 5

253

53. Guidance on Article 82. 54. DG Competition Discussion Paper on the Application of Article 82 of the Treaty to Exclusionary Abuses (2005) (hereafter Article 82 Discussion Paper). 55. DG Competition, Office of the Chief Economist, Discussion Paper: Miguel de la Mano & Benoit Durand, A Three-Step Structured Rule of Reason to Assess Predation under Article 82 (2005) (hereafter Predation Discussion Paper). 56. Predation Discussion Paper § 2.1. 57. Predation Discussion Paper, at § 4.3.3.2. 58. Predation Discussion Paper, at § 4.2 59. Predation Discussion Paper, § 4.3.2. In taking this position, the authors necessarily reject Baumol’s admonition to the contrary, even though they embrace other Baumol recommendations that are contained in the same article. 60. Predation Discussion Paper § 4.3.1. 61. Predation Discussion Paper § 2.3.2. 62. This position is also a feature of the “consumer betterment” standard advocated in Edlin, “Predatory Pricing.” 63. Article 82 Discussion Paper ¶ 6.1 (97) (expanded understanding of predation). 64. Article 82 Discussion Paper ¶ 6.2.2.2 (122) (recognizing the importance of recoupment but accepting the legitimacy of presuming recoupment from dominance). 65. Article 82 Discussion Paper ¶ 6.2.1 (average avoidable cost). See also ¶ 6.2.3 (long-run average incremental cost). Curiously, the Article 82 Discussion Paper prefers average avoidable cost in analyzing predation but falls back on average total cost in evaluating loyalty discounts. See Article 82 Discussion Paper ¶ 7.22 (154). Its rationale is that a seller can provide rebates over a long period without incurring actual losses. The subsequent guidance employs average avoidable cost and longrun average incremental cost. Guidance ¶ 44. 66. Article 82 Discussion Paper ¶ 66. 67. Guidance ¶ 26. 68. See text above at notes 26– 29. 69. Guidance ¶ 26. 70. Guidance, at ¶¶ 26, 67. 71. Guidance, at ¶ 63. 72. Guidance, at ¶ 65. 73. In the Guidance, at ¶ 5, the Commission indicates that its enforcement actions will be keyed to consumer harm. 74. Guidance, at ¶ 70. 75. In a footnote, the Commission also recites the Court of Justice ruling that “actual” recoupment need not be shown. The issue, however, is not actual recoupment but the likelihood of recoupment at the time of the allegedly predatory pricing. Then the Commission goes on to describe a case in which recoupment turns out to be difficult, but the predator continues with its below-cost pricing on the rationale that its initial losses are sunk and that continuance of below-cost pricing will enable it to gain monopoly profits, although not in sufficient quantity as to recoup all of its losses over the entire predatory period. The Commission’s identification of the predator’s increased market power with consumer harm suggests

notes to chapters 5–6

254

that the Commission wants to require recoupment but feels that it is constrained in doing so by the ruling of the Court of Justice in Tetra Pak. 76. United States v. Aluminum Co. of America, 148 F.2d 416, 436– 38 (2d Cir. 1945). 77. 555 U.S. 438 (2008). 78. Had the FCC required the incumbent telephone company to provide access to subscriber lines, there would have been no antitrust issue either on the price charged. Because the duty to provide access would fall within the scope of a regulatory statute, the enforcement of that duty would be the task of the FCC and not a duty enforced by the antitrust laws. Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 U.S. 398, 410 (2004). 79. Case C-280/08 P [2010] ECR. 80. Deutsche Telekom AG v. Commission of the European Communities, Case T-271/03 [2008] ECR II-427. 81. On a “sacrifice” approach, see discussion in Opportunity Cost section preceding note 30. 82. The embrace of average avoidable cost by the Commission is an innovation not yet reflected in case law.

ChApter 6 1. In this chapter, which concentrates on purchase for untransformed resale, the next level of commerce is distribution; purchasing for production inputs will receive more attention in the next two chapters. 2. This outline ignores inventories and assumes purchases take place only for immediate use or resale. In most cases, this simplification will not mislead. But in some cases, such as the EU Tomra case discussed in the following chapter, the durable nature of a product is a critical element in the evaluation of competitive impact. 3. Timothy Brennan has written frequently on this subject. See, for example, Timothy J. Brennan, “The Complement Market/Final Consumer Distinction: Exclusion and Predation in the U.S. Department of Justice Section 2 Report,” GCP: The Online Magazine for Global Competition Policy 10 (October 8, 2008, Release One): 1– 12. 4. Viscusi, Harrington, and Vernon, Economics of Regulation, 258. 5. Exclusive-supply contracts are governed by section one of the Sherman Act. 15 U.S.C. § 1. Exclusive-supply arrangements involving commodities are also governed by section three of the Clayton Act, a provision whose language is broad enough to apply both to exclusive-supply and tying arrangements. 15 U.S.C. § 14. Exclusive-supply arrangements undertaken by monopolists can also fall under the sweep of section two of the Sherman Act. 15 U.S.C. § 2. 6. For a discussion that stresses exclusive-supply contracts as a means of overcoming contracting problems and as an alternative to vertical integration, see Jonathan Jacobson, “Exclusive Dealing, ‘Foreclosure,’ and Consumer Harm,” Antitrust Law Journal 70, no. 9 (2002): 311. 7. 337 US 293 (1949). 8. 332 US 392 (1947).

notes to chapter 6

255

9. 332 US at 396. 10. $500,000 in sale was involved. 332 US at 395– 96. 11. See 337 US at 306– 7. “Requirements contracts may well be of economic advantage to buyers as well as to sellers, and thus indirectly of advantage to the consuming public. In the case of the buyer, they may assure supply, afford protection against rises in price, enable long-term planning on the basis of known costs, and obviate the expense and risk of storage in the quantity necessary for a commodity having a fluctuating demand. From the seller’s point of view, requirements contracts may make possible the substantial reduction of selling expenses, give protection against price fluctuations, and— of particular advantage to a newcomer to the field to whom it is important to know what capital expenditures are justified— offer the possibility of a predictable market.” 12. 337 US at 314. 13. The law governing tying arrangements has changed substantially since the late 1940s. A few years subsequent to the cases described in text, the Court reaffirmed its International Salt ruling in its Times-Picayune decision. Times-Picayune Pub. Co. v. United States, 345 US 594 (1953). In Times-Picayune, the Court ruled that section 3 of the Clayton Act condemned tying arrangements involving a “not insubstantial” amount of commerce while section 1 of the Sherman Act required, in addition, substantial market power in the tying product. Later cases appeared to collapse the Sherman Act’s market power requirement into the “not insubstantial” amount of commerce requirement. See, for example, Fortner Enterprises, Inc. v. U.S. Steel Corp., 394 US 495, 501– 2 (1969); Northern Pac. Ry. Co. v. United States, 356 US 1 (1958). Jefferson Parish, however, rejuvenated the market power requirement. Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 U.S. 2 (1984). Since that time, it appears that the tests of legality under the Sherman and Clayton Acts have coalesced, and both now require a showing of market power to establish illegality. 14. 365 US 320, 327– 29 (1961). 15. 365 US at 334. 16. 365 US at 335. Because the Clayton Act was designed to bar potential Sherman Act violations in their incipiency, it was thought that the substantive scope of the Clayton Act was broader than the Sherman Act. This view is incorporated in a number of judicial decisions applying section 3 to exclusive-supply contracts. See Twin Cities Sportservice, Inc. v. Charles O. Finley & Co., 676 F.2d 1291, 1304 n. 9 (9th Cir. 1982). In recent years the courts have concluded that the standards of section 3 and section 1 have converged. Sheridan v. Marathon Petroleum Co., 530 F.3d 590, 592 (7th Cir. 2008); Southern Card & Novelty, Inc. v. Lawson Mardon Label, Inc., 138 F.3d 869, 874 (11th Cir. 1998); Town Sound & Custom Tops, Inc. v. Chrysler Motors Corp., 959 F.2d 468, 495– 96 n. 42 (3d Cir. 1992); Bob Maxfield, Inc. v. American Motors Corp., 637 F.2d 1033, 1037 (5th Cir.), cert. denied, 454 US 860 (1981). See also Bork, Antitrust Paradox. 17. 337 US at 304 n. 6. 18. 365 US at 333. 19. 466 US 2 (1984). 20. 466 US at 26. 21. See, for example, Northern Pac. Ry. v. United States, 356 US 1, 6 (1958). 22. Illinois Tool Works Inc. v. Independent Ink, Inc., 547 US 28, 35 (2006).

notes to chapter 6

256

23. 466 US at 45. 24. 373 F.3d 57 (1st Cir. 2004). 25. 373 F.3d at 68. Accord, B&H Medical, LLC v. ABP Administration, Inc., 526 F.3d 257, 266 (6th Cir. 2008) (dictum). See also Eastern Food Services, Inc. v. Pontifical Catholic University Services Ass’n, 357 F.3d 1, 9 (1st Cir. 2004). 26. United States v. Microsoft Corp., 87 F.Supp.2d 30, 52 (D.D.C. 2000). 27. 87 F.Supp.2d at 53– 54. 28. United States v. Microsoft Corp., 253 F.3d 34, 70 (D.C. Cir. 2001). 29. 127 F.3d 1157 (9th Cir. 1997). 30. United States v. Dentsply, International, Inc., 399 F.3d 181 (3d Cir. 2005). 31. Compare the court’s ruling that Dentsply’s 67 percent market share (measured on a unit basis), which supported an inference of monopoly with Judge Learned Hand’s statement in his Alcoa opinion that “it was doubtful whether sixty or sixty-four percent would be enough” to support an inference of monopoly. United States v. Aluminum Co. of America, 148 F.2d 416, 424 (2d Cir. 1945). 32. 399 F.3d at 190, 196. See also Benjamin Klein and Andres V. Lerner, “The Expanded Economics of Free-Riding: How Exclusive Dealing Prevents Free-Riding and Creates Undivided Loyalty,” Antitrust Law Journal 74 (2007): 473– 76. 33. 399 F.3d at 190, 193– 94. LePage’s dealt with bundled discounts. We discuss LePage’s in chapter 8 in connection with our treatment of bundled discounts. 34. 253 F.3d at 70. 35. 696 F.3d 254 (3d Cir. 2012). 36. The Block Exemption Regulation was revised in 2010 to supplement the maximum seller share with a maximum buyer share of 30 percent of the relevant product or geographic market. 37. Hoffmann at ¶¶ 90, 120. See also id., at ¶ 116, where the Court holds that compliance with Article 101(3) does not preclude application of Article 102. 38. Manufacture française des pneumatiques Michelin v. Commission of the European Communities, Case T-203/01, 30 Sept. 2003 (Michelin II), at ¶ 58. 39. Nicolas Petit, “From Formalism to Effects? The Commission’s Communication on Enforcement Practices in Applying Article 82 EC,” World Competition: Law and Economics Review 32 (2009): 494. 40. Guidance ¶ 20. 41. Ginevra Bruzzone and Marco Boccaccio, “Impact-Based Assessment and Uses of Legal Presumptions in EC Competition Law: The Search for the Proper Mix,” World Competition: Law and Economics Review 32 (2009): 479– 80. 42. See Richard A. Posner, Antitrust Law, 2nd ed. (Chicago: University of Chicago Press, 2001), 231; Bork, Antitrust Paradox, 140. For a review of the theoretical development summarized here, see Michael D. Whinston, “Exclusivity and Tying in U.S. v. Microsoft: What We Know, and Don’t Know,” Journal of Economic Perspectives 15 (2001): 63. 43. Eric B. Rasmusen, J. Mark Ramseyer, and John Shepard Wiley Jr., “Naked Exclusion,” American Economic Review 81 (December 1991): 1137– 45. For a judicial discussion of this problem, see Nicsand, Inc., v. 3M Co., 457 F.3d 534, 544– 45 (6th Cir. 2006), rev’d en banc, 507 F.3d 442 (6th Cir. 2007). 44. For an excellent summary of several possibilities from the economics literature, see Viscusi, Harrington, and Vernon, Economics of Regulation, 258– 62.

notes to chapters 6–7

257

45. Christodoulos Stefanadis, “Selective Contracts, Foreclosure, and the Chicago School View,” Journal of Law and Economics 41 (1998): 430. 46. Chiara Fumagalli and Massimo Motta, “Exclusive Dealing and Entry When Buyers Compete,” American Economic Review 96 (2006): 785– 95. 47. Fumagalli and Motta, “Exclusive Dealing.” 48. Thomas G. Krattenmaker and Steven C. Salop, “Anticompetitive Exclusion: Raising Rivals’ Costs to Achieve Power Over Price,” Yale Law Journal 96 (1986): 209. 49. Thus in a raising-rivals’-costs context, the focus is horizontal: Has the supplier monopolized a complement market to which its rivals in the product market must have access? But the vertical emphasis found in raising-rivals’-costs models can aid an analyst in identifying the relevant complement market. See Timothy J. Brennan, “‘Vertical Market Power’ as Oxymoron: Horizontal Approaches to Vertical Antitrust,” George Mason Law Review 12 (2004): 903. 50. Nicsand, Inc. v. 3M Co., 457 F.3d 534 (6th Cir. 2006), rev’d en banc, 507 F.3d 442 (6th Cir. 2007). 51. 457 F.3d at 544– 45. 52. 507 F.3d at 452. 53. 507 F.3d at 453– 54. The court also observed that exclusivity was a condition imposed by (most of) the retailer customers themselves. 54. Timothy J. Brennan, “Bundled Rebates as Exclusion Rather Than Predation,” Journal of Competition Law and Economics 4 (2008): 338– 39.

ChApter 7 1. In the period immediately following the enactment of the Robinson-Patman Act, the Federal Trade Commission challenged some such discount schedules as violations of that act, when, in its view, significant discounts attached to very large volumes which would be out of the reach of smaller buyers. See FTC v. Morton Salt Co., 334 US 37 (1948). 2. The market outcome of such bilateral monopoly is indeterminate over a range of price and quantity, but the term power buyer is more appropriate than monopsonist exactly because of the seller’s market power. For a good discussion, see Zhiqi Chen, “Buyer Power: Economic Theory and Antitrust Policy,” in Research in Law and Economics, vol. 22, ed. Richard O. Zerbe and John B. Kirkwood (Bingley, West Yorkshire, England: Emerald Group Publishing, 2000), 17– 40. 3. In this connection, bundled discounts may also be a response by a seller to the organizational structure of important buyers. It can more easily obtain the attention of the buyer’s higher-level management when it offers a discount keyed into the buyer’s aggregate purchases. When buying decisions in, say, a large retailer is allocated among buyers in each of the retailer’s departments, there is more of a chance that one or more of the buyers will be unresponsive to a price reduction, balancing the price reduction off against perceived quality or other nonprice factors. When the price reduction is spread over all of the buyer’s purchases, that price reduction (because the aggregate benefit to the buyer is likely to be substantial) is more likely to persuade higher-level management that it merits a positive response. Thus, for example, in the LePage’s case, Judge Greenberg’s dissent noted that “Le-

258

notes to chapter 7

Page’s tape had been selling well when its buyers were directed by senior management to ‘maximize’ all purchases from 3M to maximize the EGF/PGF rebate.” 324 F.3d, at 173. 4. Concord Boat Corp. v. Brunswick Corp., 207 F.3d 1039 (8th Cir. 2000). 5. Allied Orthopedic Appliances Inc. v. Tyco Health Care Group LP, 592 F.3d 991 (9th Cir. 2010); Masimo Corp. v. Tyco Health Care Group, L.P., 350 Fed. Appx 95, 2009 WL 3451725 (9th Cir. 2009). 6. 696 F.3d 254 (3d Cir. 2012). 7. 207 F.3d at 1059. 8. 207 F.3d at 1062. 9. Id. 10. 207 F.3d at 1063. 11. Tyco’s original product line (R-Cal) was protected by a patent that expired in 2003. Before the expiration of that patent, Tyco’s share of stand-alone pulse oximetry monitor sales in the United States was between 62 and 64 percent. When Tyco introduced its new (OxiMax) line of equipment, it ceased producing R-Cal equipment, both sensors and monitors. In Tyco’s new OxiMax line, the sensors were compatible with its new (OxiMax) monitors as well as with its old (R-Cal) monitors, but the new OxiMax monitors were incompatible with the older (R-Cal) sensors produced under the expired patent. The plaintiffs, who produced generic sensors using the R-Cal technology, claimed that, partially because many hospitals had a significant installed base of R-Cal monitors, Tyco was using design to maintain its monopoly. 12. 592 F.3d at 997 n. 2. 13. FTC, In the Matter of Intel Corp., Decision and Order, 2010 WL 3180281 (FTC Aug. 4, 2010). 14. The duopoly aspect of the Intel case, while unusual, is not unique: Nicsand, an exclusive-supply case, discussed above, raises comparable issues. 15. FTC Complaint, ¶ 53. 16. We use the abbreviation LAIC. This abbreviation was used by Bolton, Brodley, and Riordan and is identical to Baumol’s AIC and the European Commission’s LRAIC. 17. FTC, Decision and order, part iv, ¶ 6. 18. FTC, Decision and order., at part iv, ¶¶ 5, 7. 19. In addition, the FTC’s Analysis of Proposed Consent Order to Aid Public Comment (p. 10) notes that the agency will not necessarily take a particular approach to bundling practices in other cases; thus, there are no “safe harbors.” 20. This can be represented symbolically as follows: Let p = list price, d = discount, t = target set for the individual purchaser by the supplier, and q = amount already purchased. The average per-unit effective price for the block of additional purchases required to meet the target is then:

(p – d)(t – q) – qd (qd) =p–d– . (t – q) (t – q) 21. See European Commission, DG Discussion Paper on the Application of Article 82 of the Treaty to Exclusionary Abuses ¶¶ 153– 54 (2005) (describing this

notes to chapter 7

259

“suction effect”); Frank P. Maier-Rigaud, “Article 82 Rebates: Four Common Fallacies,” European Competition Journal 2 (2006): 87– 88. 22. A number of scholars have analyzed the effect of loyalty rebates in tying the purchaser increasingly to the supplier offering them. In addition to the citations in the previous footnote, see Patrick Greenlee, David S. Reitman, and David S. Sibley, “An Antitrust Analysis of Bundled Loyalty Discounts,” Economic Analysis Group Discussion Paper No. 04-13, 2005, 5, available at http://ssrn.com/abstract=600799; Patrick Greenlee and David S. Reitman, “Competing with Loyalty Discounts,” Economic Analysis Group Discussion Paper No. 04-2, 2005, 7– 8, available at http:// ssrn.com/abstract=502303. Frank P. Maier-Rigaud of the European Commission’s General Competition Directorate, for example, demonstrates the high switching costs that would be incurred by a purchaser as it approached the target amount. Maier-Rigaud offers a diagram similar to the one above to demonstrate the “suction effect” encountered by any purchaser approaching the target. Maier-Rigaud, “Article 82 Rebates,” 87. Similar graphs appear numerous times in the ProkentTomra decision. Commission Decision of 29 March 2006, Comp/E-1/38.113 (Tomra) (figures 15, 18, 21, 22, 23, 24, 27). Indeed, Maier-Rigaud challenges other writers’ contention that the “suction effect” is overstated. Maier-Rigaud focuses on the argument that when the demand of a particular customer is greater than the target amount, there are no suction effects on its purchases that exceed the latter. He is particularly concerned with the contention that the price a competitor would have to offer to persuade the customer to switch prior to the point at which the customer’s purchases reached the target amount would rise as demand increases, because the competitor could offer a price that averages the posttarget price with the low pretarget prices. Meier-Rigaud, “Article 82 Rebates,” 6; see also Giulio Federico, “When Are Rebates Exclusionary?” European Competition Law Review 26 (2005): 477– 80. Maier-Rigaud dismisses these contentions, claiming that it would be irrational for a seller offering a target rebate to set the target in excess of the expected demand of the customer or substantially below it. 23. 926 F.Supp. 371 (S.D.N.Y. 1996). 24. See Discussion Paper at ¶ 155 and ¶ 156. 25. Discussion Paper at ¶ 155. 26. Guidance on the Commission’s Enforcement Priorities in Applying Article 82 of the EC Treaty to Abusive Exclusionary Conduct by Dominant Undertakings. 2009 O.J. (C 45) 7 (Feb. 24, 2009). 27. Guidance on Article 82, ¶ 26. 28. Discussion Paper, 155. 29. Guidance on Article 82, 42 n. 1 30. American antitrust observers will recall a parallel in the analysis proposed by the Antitrust Modernization Commission (AMC) for bundled discounts. The first step of that analysis involves a discount attribution rule whose application similarly involves the facts of each particular customer’s purchases. But the ultimate focus is on aggregate market effects. The AMC’s treatment of bundled discounts is considered later in chapter 8. 31. Discussion Paper, 145. 32. Discussion Paper, 145.

notes to chapter 7

260

33. Guidance, 20. 34. Tomra Systems ASA v. European Commision, Case C-549/10 P (2012), affirming Tomra Systems ASA v. European Commision, Case T-155/06 (2010). 35. Tomra does not use the term “commercially viable amount,” but it does use “contestable amount,” a term the discussion paper uses interchangeably with the former. 36. In the Netherlands, the noncontestable market share was below 20 percent in 1998, above 60 percent in 1999– 2000, 58 percent in 2001, and 37 percent in 2002 (Tomra, 163 and figure 12). In Sweden, the noncontestable market share was 20 percent in 2000, rising to over 50 percent in 2002 (Tomra, 185 and figure 17). In Germany, the noncontestable share was 22 percent in 1998, rose to above 33 percent in 1999, fell to under 30 percent in 2000, and to 26 percent in 2001 (Tomra, 219 and figure 20). In Austria, the noncontestable share of demand was 36 percent in 1999, 50 percent in 2000, and around 11 percent in 2001 (Tomra, 238 and figure 25). In Norway, Tomra’s noncontestable market share was 5 percent in 1998, exceeded 90 percent in 1999, was 85 percent in 2000, and 0 in 2002 (Tomra, 270.) 37. Its share was over 99 percent in the Netherlands in 1999– 2000, 87– 97 percent in 2001, and 85– 95 percent in 2002 (Tomra, 163 and figure 12); approximately 95 percent in Austria in 1999 and 2000, and falling to over 70 percent in 2001 (Tomra, 238 and figure 25). Its share of the Swedish market was 70 in 2000, rising to 90 percent or more by 2002 (Tomra, 185 and figure 17). Its share of the German market ranged from 60– 70 percent during the period (Tomra, figure 20); and its share of the Norwegian market was 94 percent in 1999, rising to 94– 99 percent in 2000 and 2001, and falling to 88– 93 percent in 2002 (Tomra, 270). 38. Tomra, 121 and 115 at nn. 222, 223. 39. Tomra, 122, 220, 309. 40. Frank P. Maier-Rigaud and Dovile Vaigauskaite, “Prokent/Tomra: A Textbook Case? Abuse of Dominance under Perfect Information,” EC Competition Policy Newsletter 2 (2006): 19– 24. 41. Judgment of the General Court (Fifth Chamber) 9 Sept. 2010, Case T– 155/06 (Tomra) (hereafter GC Tomra), 243. 42. GC Tomra, 241. 43. Judgment of the Court of Justice (Third Chamber), 19 April 2012, Case C-549/10 P (Tomra Systems ASA v. European Commission) (hereafter CJ Tomra) at ¶¶ 42, 44. 44. See discussion of Hoffmann-La Roche in chapter 6. See also British Airways plc v. Commission of the European Communities, Case C-95/04 P, 15 March 2007 at ¶¶ 71– 76; Manufacture française des pneumatiques Michelin v. Commission of the European Communities, Case T-203/01, 30 Sept. 2003 (Michelin II) at ¶ 57. 45. Guidance at ¶ 20. 46. CJ Tomra at ¶ 46. See also id., at ¶¶ 44, 48. 47. See, for example, Daniel A. Crane and Graciela Miralles, “Toward a Unified Theory of Exclusionary Vertical Restraints,” Southern California Law Review 84 (2011): 605– 60, 84 S. Cal. L.Rev. 605, 608, 639– 46 (2011) (applying foreclosure test based on minimum efficient scale). See Joshua D. Wright, “Moving beyond Naïve Foreclosure Analysis,” George Mason Law Review 19, no. 5 (2012): 1163, 1166,

notes to chapter 7

261

(recognizing consensus supporting minimum efficient scale as critical to foreclosure analysis). 48. CJ Tomra at ¶ 81. 49. Commision Decision of 13 May 2009 relating to a proceeding under Article 82 of the EC Treaty and Article 54 of the EEA Agreement (COMP/C-3/37.990Intel). The Commission decision was affirmed by the General Court on June 12, 2014. Intel Corp. v. European Commision, Case T-286/09 (2014). 50. One source, Passmark Software, put the AMD share at 26 percent in April of 2009, www.passmark.com. 51. ¶ 1598. 52. ¶ 1598. 53. Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc., 920 F.Supp. 455 (S.D.N.Y. 1996). 54. AMC Report and Recommendations, 94– 106. 55. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 US 209, 223– 24 (1993). The lower courts have largely determined that the proper measure of cost for this analysis is marginal cost, or its practical surrogate, average variable cost. See discussion in chapter 5. 56. Phillip Areeda and Donald F. Turner, “Predatory Pricing and Related Practices under Section 2 of the Sherman Act,” Harvard Law Review 88 (1975): 716– 18. 57. ¶ 1577, ¶¶ 1584– 92. 58. ¶ 1603. 59. One of the recipients of payments Intel made was Media-Saturn Holding GmbH (MSH), a large German consumer-electronics retailer. MSH did not purchase directly from Intel, but the Intel payments compensated MSH for buying products containing Intel chips. The Commission applied its required and contestable share analysis to MSH in the same way that it did for Intel’s direct customers. But MSH, as a large retailer, is a player in the consumer market where AMD held a 33 percent share during the relevant period. Therefore, it is unclear what the Commissions required and contestable share analysis, which it applied to MSH, was designed to reveal. Perhaps that analysis would show that AMD was unable to sell to MSH, but since AMD held one-third of the consumer market anyway, the fact that it was unable to sell to MSH does not carry competitive significance. 60. ¶ 1604. 61. ¶ 1605. 62. ¶ 1607. 63. ¶ 1605. 64. “The aim of the Commission’s enforcement activity in relation to exclusionary conduct is to ensure that dominant undertakings do not impair effective competition by foreclosing their competitors in an anti-competitive way, thus having an adverse impact on consumer welfare, whether in the form of higher price levels than would have otherwise prevailed or in some other form such as limiting quality or reducing consumer choice.” Guidance, 19. 65. Greater FTC intervention based on Section 5 was anticipated by Franklin Foer of the American Antitrust Institute as a “bridge” to the EU before the FTC announced the action against Intel. Albert A. Foer, “Section 5 as a Bridge toward

notes to chapters 7–8

262

Convergence,” FTC Workshop on Section 5, October 17, 2008. After the Intel decision Keith Hylton saw the FTC’s shift toward the European Union as a very retrograde development. Keith N. Hylton, “Intel and the Death of U.S. Antitrust Law,” CPI Antitrust Journal 2 (February, 28, 2010). 66. See chapter 6, text following note 54.

ChApter 8 1. SmithKline Corp. v. Eli Lilly & Co., 427 F.Supp. 1089 (E.D. Pa. 1976), aff’d, 575 F.2d 1056 (3d Cir. 1978); Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc., 920 F.Supp. 455 (S.D.N.Y. 1996). 2. 324 F.3d 141 (3d Cir. 2003). 3. AMC Report and Recommendations, 94– 106. 4. Cascade Health Solutions v. PeaceHealth, 515 F.3d 883 (9th Cir. 2008). 5. In a tying arrangement, the buyer must take the package if he wants the tying product. The buyer is thus “forced” or “coerced” into taking the tied product. In a bundle, US courts see this element of coercion as absent because the buyer has the option of buying (or not buying) either product. Were the seller to formally offer a bundle but price the “tying” product so high that it was practically unavailable separately, the “bundle” would be transformed into a tie. 6. Jefferson Parish Hosp. Dist. No. 2 v. Hyde, 466 US 2, 26– 29 (1984). 7. Varying approaches to the antitrust treatment of bundled discounts are reflected in diametrically opposed analyses of the problem contained in successive editions of the leading antitrust treatise by Areeda and Hovenkamp. See Areeda and Hovenkamp Antitrust Law, ¶ 749 (arguing that [c]ourts should not entertain claims that while a defendant’s overall price is remunerative, the separate ‘price’ for one particular component is predatory”) with id. ¶ 749 (Supp. 2004) (approving the LePage’s result while conceding that “the defendant’s aggregated discounts did appear to produce immediate gains in sales volume and profits.”) All of this attests to the troubling nature of this marketing tool and the uncertainties attending attempts to evaluate it. 8. In this connection, bundled discounts may also be a response by a seller to the organizational structure of important buyers. It can more easily obtain the attention of the buyer’s higher-level management when it offers a discount keyed into the buyer’s aggregate purchases. When buying decisions in, say, a large retailer are allocated among buyers in each of the retailer’s departments, there is more of a chance that one or more of the buyers will be unresponsive to a price reduction, balancing the price reduction off against perceived quality or other nonprice factors. When the price reduction is spread over all of the buyer’s purchases, that price reduction (because the aggregate benefit to the buyer is likely to be substantial) is more likely to persuade higher-level management that it merits a positive response. Thus, for example, in the LePage’s case, Judge Greenberg’s dissent noted that “LePage’s tape had been selling well when its buyers were directed by senior management to ‘maximize’ all purchases from 3M to maximize the EGF/PGF rebate.” 324 F.3d at 173. 9. This dynamic is well captured in the Automatic Canteen case of 1953. Automatic Canteen Co. v. FTC, 346 U.S. 61 (1953).

notes to chapter 8

263

10. See, for example, Chen, “Buyer Power.” 11. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 US 209, 223– 24 (1993). 12. 509 US at 222, 223. 13. See chapter 5, text at notes 13– 16. 14. Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc., 920 F.Supp. 455 (S.D.N.Y. 1996). 15. 920 F.Supp. at 467. The passage was disapproved in Phillip E. Areeda and Herbert Hovenkamp, Antitrust Law, rev. ed. 749, n. 6. The authors concluded that in Ortho, the plaintiff was “not complaining of pricing, but rather of the package discount, which is a form of tying arrangement” (italics in original). Judge Greenberg, dissenting in LePage’s, cited that portion of the Areeda and Hovenkamp treatise. 324 F.3d at 176 n. 2. The treatise appears to approve of the LePage’s majority decision in later editions. See Phillip E. Areeda and Herbert Hovenkamp, Antitrust Law, 2nd ed. 749 at 183 (Supp. 2004). 16. SmithKline Corp. v. Eli Lilly & Co., 427 F.Supp. 1089 (E.D. Pa. 1976), aff’d, 575 F.2d 1056 (3d Cir. 1978). 17. Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc., 920 F.Supp. 455 (S.D.N.Y. 1996). 18. 920 F.Supp. at 469. 19. 324 F.3d at 161– 62. 20. 324 F.3d at 177. 21. Barry Nalebuff, “Bundling as an Entry Barrier,” Quarterly Journal of Economics 119 (2004): 159– 87. 22. Benjamin Klein and Andres Lerner, “The Law and Economics of Bundled Pricing: LePage’s, PeaceHealth, and the Evolving Antitrust Standard,” Antitrust Bulletin 53, no. 3 (2008): 555– 85. 23. Yet there is little doubt they would pass any test as “separate products” based on the branded goods version of transparent tape to profitably raise its price by a very significant margin above its marginal cost without being checked by a generic version. Indeed this is much of what the case is about. 24. It is also true, however, that any lowering of the price of the high mark-up product might, by itself (if passed on to the final purchaser) create more consumer surplus than a similar cut to the price of the generic product. Klein and Lerner stress this effect. 25. See discussion later in this chapter. 26. AMC Report and Recommendations, 99– 100. 27. 920 F.Supp. at 467. 28. There are differing views in the US literature about how much burden the plaintiff should assume to demonstrate attempts at the necessary coordination among bundled goods sold under competition. Compare Thomas A. Lambert, “Evaluating Bundled Discounts,” Minnesota Law Review 89 (2005): 1722– 23, and Thomas A. Lambert, “Appropriate Liability Rules for Tying and Bundled Discounting,” Ohio State Law Journal 72 (2011): 909, with Daniel A. Crane, “Mixed Bundling, Profit Sacrifice, and Consumer Welfare,” Emory Law Journal 55 (2006): 423– 86. 29. This is the approach taken by the DOJ Section 2 report of 2008 (101– 2),

notes to chapter 8

264

where sales above average variable cost using this technique would constitute a “safe harbor” from legal attack. The Obama administration rejected that report. 30. Crane, “Mixed Bundling”; Herbert Hovenkamp and Erik Hovenkamp, “Exclusionary Bundled Discounts and the Antitrust Modernization Commission,” Antitrust Bulletin 53 (2008): 517– 53. 31. See discussion in Crane, “Mixed Bundling,” and in Hovenkamp and Hovenkamp, “Exclusionary Bundled Discounts.” 32. Oliver E. Williamson, The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting (New York: Free Press, 1985), 1– 63. 33. And hence straightforward predatory pricing analysis can be employed. 34. Dennis Carlton, Patrick Greenlee, and Michael Waldman, “Assessing the Anticompetitive Effects of Multiproduct Pricing,” National Bureau of Economic Research Working Paper No. 14199, 2008, 34. 35. Cascade Health Solutions v. PeaceHealth, 515 F.3d 883 (9th Cir. 2007, 2008). Subsequently, the Eighth Circuit rejected a bundled discount claim on the grounds that (1) at least three of the defendant’s rivals sell the full line of products sold by the defendant and (2) the plaintiff did not show that the defendant sold items below cost, even applying the discount attribution rule. Southeast Missouri Hosp. v. C. R. Bard, Inc., 616 F.3d 888 (8th Cir. 2010). That opinion was vacated, however, and the case was decided on other grounds. Southeast Missouri Hosp. v. C. R. Bard., 642 F.3d 608 (8th Cir. 2011). 36. 515 F.3d at 899. See AMC Report and Recommendations, 97. In the Southeast Missouri case, the Eighth Circuit raised the question whether the Ninth Circuit continues to recognize the discount attribution rule. 616 F.3d at 893. The basis for the Eighth Circuit’s doubt was the Ninth Circuit’s rejection of a price-squeeze claim in Doe v. Abbott Laboratories, 571 F.3d 930, 935 (9th Cir. 2009), where the court followed the Supreme Court’s ruling in Pacific Bell Tel. Co. v. Linkline Communications, Inc., 129 S.Ct. 1109 (2009). Because the court decided the case based on the Linkline precedent, it stated that it “need not discuss Cascade’s impact on this case or others pending in the district court.” The district court for the Northern District of California has ruled that the discount attribution rule continues to apply to bundled discount pricing claims. Safeway Inc. v. Abbott Laboratories, 2010 WL 147988 (N.D. Cal. 2010). 37. 515 F.3d at 899– 900. See AMC Report and Recommendations, 97. 38. 515 F.3d at 899– 900 at 911 n. 21. 39. 515 F.3d at 910 n. 21. 40. 515 F.3d at 910 n. 21. 41. Einer Elhauge, “Tying, Bundled Discounts, and the Death of the Single Monopoly Theory,” Harvard Law Review 123 (December 2009). 42. Guidance ¶ 60. 43. Because exclusive-supply contracts are presumed to violate Article 102’s prohibition of abusive conduct by dominant firms, and because dominance is determined separately from abuse, there is a possibility that Article 102 would bar a firm with a 40 percent market share from entering into exclusive-supply contracts affecting only a small percentage (say 10 percent) of the market. If so, then the superficial similarity of the law under Article 102 to the law under section 1 of the Sherman Act and section 3 of the Clayton Act would be undermined.

notes to chapter 8

265

44. The AMC did not define the term incremental cost. In its Brooke Group opinion, the Supreme Court used the phrase “some measure of incremental cost” in describing the price/cost relationship necessary for a determination of predation. Brooke Group v. Brown & Williamson Tobacco Corp., 509 US 209, 223, 257 (1993). Later, in a widely read article, Professor William Baumol discussed various measures of costs and their possible uses in predatory pricing analysis. Baumol, “Predation,” 49– 72. In that article, Baumol distinguished average incremental cost from average avoidable cost. Sunk cost necessarily incurred in producing the incremental output that is allegedly being sold at predatory prices does not affect average avoidable cost. “Predation,” 56. It is unclear whether the AMC used that term in the more imprecise way that the Court used it in Brooke Group or whether it was using the term in a sense closer to that of Baumol’s usage. 45. If the firm prices a bundle at the list prices of the components less a discount as 3M apparently did, the pricing is transparent. If the firm just offers a bundle at a package price, however, then the separate components have no individual prices. Nonetheless, all other things being equal, those buyers choosing to accept the bundle are stating by their actions that the monopoly product included in the bundle is more of a bargain than if purchased separately at the monopoly price. 46. This is not to deny that in some industries an accurate forecast of demand is critically important for low cost production or to avoid capacity constraints. See Herbert J. Hovenkamp, “The FTC’s Anticompetitive Pricing Case against Intel,” CPI Antitrust Chronicle 2 (2010). 47. As discussed in previous chapters, most economists in both the United States and the European Union probably accept the consumer welfare standard as a political accommodation rather than really embracing it because they expect the Kaldor-Hicks criterion (that employs total welfare) to guide public policy. 48. The concept of “impediment competition” was developed by the ordoliberals. Ahlborn and Grave, “Walter Eucken and Ordoliberalism,” 205. 49. Nicholas Economides and Ioannis Lianos, “The Elusive Antitrust Standard on Bundling in Europe and in the United States in the Aftermath of the Microsoft Cases,” Antitrust Law Journal 76 (2009). Skepticism that competition policy can protect the optimum price-variety-quality margins is strongly supposed by arguments and examples given in Klein and Murphy, “Exclusive Dealing Intensifies Competition for Distribution,” 433. See also Crane and Miralles, “Toward a Unified Theory of Exclusionary Vertical Restraints,” 605– 60. 50. Economides and Lianos, “The Elusive Antitrust Standard,” imply that behavioral economics vindicates the European approach, but there is no explicit argument or example given. 51. Gerber, Law and Competition, 314– 15. 52. Phillip Areeda and Herbert Hovenkamp, Antitrust Law: An Analysis of Antitrust Principles and Their Application, 3rd ed. (New York: Aspen Publishers, 2008), ¶ 749 at 323. 53. Posner, Antitrust Law, 194– 95. For exclusionary practices generally, Posner would require that the plaintiff prove the defendant possess monopoly power and that, in the circumstances, the practice is likely to exclude an equally or more efficient competitor. The defendant then can rebut by proving that the practice is,

notes to chapters 8–9

266

on balance, efficient. If the courts were capable of applying the analyses presented in this chapter, Posner’s approach, which describes the standard antitrust burden allocations, would work. The problem, however, is that the courts have generally employed flawed analyses. 54. Aspen Skiing Co. v. Aspen Highlands Skiing Corp., 472 US 585 (1985). 55. Easterbrook, “The Limits of Antitrust,” 15.

ChApter 9 1. Patent Act, 35 U.S.C. § 103(a). 2. European Patent Convention, Article 56. Article 56 contains a “nonobvious” requirement similar to that contained in §103 of the US patent act. 3. In the language of patent law, such drastic innovations are commonly referred to as pioneer inventions. The US Supreme Court described “pioneering” inventions in Boyden Power Brake Co. v. Westinghouse, 170 US 537, 561– 62 (1898) in terms that are fully consistent with the type of drastic innovation that Evans and Schmalensee find characteristic of the new economy. (The views of Evans and Schmalensee are discussed below.) In its Boyden opinion the Court described “pioneer” inventions as follows: “This word [pioneer], although used somewhat loosely, is commonly understood to denote a patent covering a function never before performed, a wholly novel device, or one of such novelty and importance as to mark a distinct step in the progress of the art, as distinguished from a mere improvement or perfection of what had gone before. Most conspicuous examples of such patents are the one to Howe, of the sewing machine; to Morse, of the electrical telegraph; and to Bell, of the telephone.” 170 U.S. at 563. 4. In one of the foundational cases of copyright, the Court ruled that copyright could not protect the inventor (Selden) of a new form of accounting from the competition of a rival (Baker) who was selling a book explaining Selden’s system. Baker could not copy Selden’s book, but he was free to write his own book explaining Selden’s system. Copyright thus contemplates that copyrighted works will be in competition with similar works. Baker v. Selden, 101 U.S. 99 (1879). 5. See Uniform Trade Secrets Act § 1; American Law Institute, Restatement (Third) of Unfair Competition §§ 39– 43 (1995); Richard A. Posner, “Intellectual Property: The Law and Economics Approach,” Journal of Economic Perspectives 19 (2005): 62. 6. 250 US 300, 307 (1919). 7. 224 US 383 (1912). 8. Hecht v. Pro-Football, Inc., 570 F.2d 982 (D.C. Cir. 1976). 9. Intergraph Corp. v. Intel Corp., 195 F.3d 1346, 1356– 57 (Fed. Cir. 1999). 10. 472 US 612 (1985). 11. Aspen Skiing was competing with many other destination ski resorts in the United States, Canada, and Europe for the patronage of holiday skiers, and whether Aspen and Highlands together could raise prices significantly above its costs (the merger Guidelines test for a market) cannot be known. Once a skier had chosen Aspen as a destination, however, Aspen Skiing’s only rival was Highlands. 12. 472 US at 611 n. 44. 13. 342 US 143 (1951).

notes to chapter 9

267

14. 410 US 366 (1973). 15. 540 US 398 (2004). 16. 1988 ECR 6211. 17. 1974 ECR 223. 18. AB Volvo v. Erik Veng (UK) Ltd., 1988 ECR 6211. 19. AB Volvo at ¶ 10. 20. AB Volvo at ¶ 8. 21. Id. 22. 224 US 1 (1912). 23. 243 US 502 (1917). 24. Carbice Corp. of America v. American Patents Development Corp., 283 US 27 (1931); Morton Salt Co. v. G. S. Suppinger Co., 314 US 488 (1942). 25. Mercoid Corp. v. Mid-Continent Inv. Co., 320 US 661 (1944); Mercoid Corp. v. Minneapolis-Honeywell Regulator Co., 320 US 680 (1944). 26. This falls within the logic of the single monopoly profit theorem. 27. A staple of article commerce is any product with substantial uses beyond those with the patented product. 28. See, e.g., United States v. Loew’s Inc., 371 US 38, 45 (1962). 29. Illinois Tool Works Inc. v. Independent Ink, Inc., 547 US 28 (2006). 30. 504 US 451 (1992). 31. 504 US at 477 n. 24. This point was confirmed in lower court decisions. See for example, Psi Repair Services, Inc. v. Honeywell, Inc., 104 F.3d 811, 820 (6th Cir. 1997). 32. The two cases discussed in this paragraph are Image Technical Services, Inc. v. Eastman Kodak Co., 125 F.3d 1195, 1219 (9th Cir. 1997), and In re Independant Service Organizations Antitrust Litigation (Xerox), 203 F.3d 1322 (Fed. Cir. 2000), cert. denied 531 U.S. 1143 (2001). 33. Illinois Tool Works Inc. v. Independent Ink, Inc., 547 U.S. 28 (2006). 34. In this and the following paragraph, we draw from an earlier article by one of the authors. See Daniel J. Gifford, “The Antitrust/Intellectual Property Interface: An Emerging Solution to an Intractable Problem,” Hofstra Law Review 31 (2002): 363, at 410– 11. 35. Public policy in both the United States and the European Union favors copyright treatment of computer software to provide only limited protection. See discussion in text preceding note 4. 36. Case T-201/04 Microsoft Corp. v. Commission of the European Communities (2007). 37. The new Windows 2000 server operating systems linked servers and client PCs together in a more integrated system than was possible under the earlier Windows NT system. The new system was based in large part upon Active Directory (controlling security and allocating functions) and multimaster replication. In the Windows 2000 system, directory services are performed by many small, relatively inexpensive servers often separated from each other by substantial distances, linked in a so-called ensemble. Now changes to domain accounts could be made on any server with those changes automatically propagated to the other domain controllers. Again, Windows 2000 server systems could borrow the identity of a PC and request a service from another server on behalf of that client PC.

notes to chapter 9

268

38. Compare Steven Anderman, “Does the Microsoft Case Offer a New Paradigm for the ‘Exceptional Circumstances’ Test and Compulsory Copyright Licensees under EC Competition Law?” Competition Law Review 1 (2004): 15 (stating that “The Sun Solaris work group server was not a clone of the Microsoft server; it actually preceded it in the market”) with Paula Rooney and Barbara Darrow, “Sun’s Strategy— Vendor Takes on Microsoft on Desktop, Server Fronts with Upgrades,” Computer Reseller News 102 (May 20, 2002); Paula Rooney, “Microsoft Calls Sun Copycat,” CMPnetasia.com, May 24, 2002. 39. See ¶¶ 1194, 1225. 40. Radio Telefis Eireann (RTE) v. Commission of the European Communities, 1995 ECR I-743. 41. Radio Telefis Eireann 46. 42. AB Volvo v. Erik Veng (UK) Ltd., 1988 ECR 6211. 43. 1995 ECR I-743 at ¶¶ 49– 50. 44. 1995 ECR I-743 at ¶¶ 54– 56. 45. Tierce Ladbroke SA v. E.C. Commission, Case T-504/93 (1997), [1997] 5 C.M.L.R. 309. 46. Tierce Ladbroke at ¶ 131. 47. Oscar Bronner GmbH & Co. KG v. Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG, 1998 ECR I-07791. 48. Oscar Bronner at ¶ 41. 49. Oscar Bronner at ¶ 43. 50. Oscar Bronner at ¶ 46. 51. IMS Health GmbH & Co. OHG v. NDC Health GmbH & Co. KG, 2004 ECR I-5039. 52. IMS Health at ¶ 52. 53. IMS Health at ¶¶ 37– 38. 54. Estelle Derclaye, “The IMS Health Decision: A Triple Victory,” World Competition: Law and Economics Review 27 (2004): 404. Compare Michele Messina, “Article 82 and the New Economy: Need for Modernisation?” Competition Law Review 2 (2006): 92, who suggested that IMS would be likely to prevail on remand to the national court because the three conditions (set forth above) are cumulative. Derclaye also believed that the cumulative nature of the Magill conditions ensured that IMS was likely to prevail. Estelle Derclaye, “Abuse of a Dominant Position and Intellectual Property Rights: A Suggestion to Reconcile the Community Courts’ Case Law,” in Competition, Regulation, and the New Economy, ed. Cosmo Graham and Fiona Smith (Portland, OR: Hart Publishing, 2004), 55– 75. Messina, however, cautioned that the Court had left unspecified how different the new product would have to be from the product currently sold by the copyright holder. Id., at 93. On this issue, Derclaye saw the issue turning on whether the national court attributed a narrow or broad construction of the new product requirement. Estelle Derclaye, “The IMS Health Decision and the Reconciliation of Copyright and Competition Law,” European Law Review 29 (2004): 695– 96. 55. 2004 ECR I-5039 at ¶ 52. 56. Michele Messina has pointed out that because the Court in IMS chose to refer “to a soft concept, such as ‘new’ rather than to a well-established and clearly defined competition law concept, such as the notion of substitution,” the Court

notes to chapter 9

269

was indicating that it considered it “sufficient that the new product presents some novel features while remaining substitutable with the existing product, thereby implying a much lower threshold for the application of Article 82 to refusal to license cases.” Messina, “Article 82 and the New Economy,” 93. Derclaye, “The IMS Health Decision and the Reconciliation of Copyright and Competition Law,” 695– 96. 57. Messina, “Article 82 and the New Economy,” 93. 58. EC Treaty, Article 102(e). 59. Case T-201/04 at ¶¶ 857– 63; Commission Decision of 24.03.2004, Case COMP/C-3/37.792 at ¶ 841. 60. See note 59. 61. Case T-201/04 at ¶ 917; Commission Decision of 24.03.2004, Case COMP/C-3/37.792 at ¶ 803. 62. 466 US 2 (1984). In the Microsoft antitrust case, the DC Circuit determined that the consumer demand test could not be applied to new functionalities incorporated into platform software without discouraging innovation. 253 F.3d at 85– 96. 63. Case T-201/04 at ¶ 932; Commission Decision of 24.03.2004, Case COMP/C-3/37.792 at ¶ 804. 64. 253 F.3d at 80– 84. 65. 253 F.3d at 95. Because of the way plaintiffs had litigated the tying claim at trial, the court of appeals barred them from arguing any theory of harm that depended upon a precise definition of browsers or barriers to entry. 66. It is unclear how the government’s case would be constructed. Would the government be permitted to show that the actual effects in the browser market produced a net negative welfare effect or would it be required to show that in general adding functionalities of similar kinds would produce a negative welfare effect? This issue is discussed in the text. Since the Netscape Navigator had been the nation’s most popular browser until improvements in the Explorer reduced the Navigator’s advantage, the browser market was not lacking in quality substitutes. 67. United States v. Grinnell Corp., 384 US 563, 570– 71 (1966). 68. Eastman Kodak Co. v. Image Technical Services, Inc., 504 US 451 (1992); United Shoe Machinery Corp. v. United States, 258 US 451, 457– 58 (1922). 69. In its findings, the district court quoted a memorandum from AOL indicating that in 1998 Netscape occupied approximately 50 percent of the browser market. United States v. Microsoft Corp., 65 F.Supp. 1, 79 (D.D.C. 1999) (finding 301). 70. See above, text at note 64. 71. Commission Decision of 24.03.2005 relating to a proceeding under Article 82 of the EC Treaty (Case COMP/C-3/37, 792 Microsoft), at ¶ 783. In addition, the Commission expressed skepticism as to whether Microsoft’s incentives to innovate would be reduced as a result of the Commission’s order to make knowledge of its protocols available to rivals. ¶¶ 727, 729. The General Court affirmed these rulings. Judgment of the Court of First Instance of 17 September 2007, Microsoft Corp. v. Commission of the European Communities, at ¶¶ 701– 12. 72. The telephone is often used as an example. As more telephones are added to the network, the value of each telephone on the network increases. 73. See W. Brian Arthur, Increasing Returns and Path Dependence in the

notes to chapter 9

270

Economy (Ann Arbor: University of Michigan Press, 1994), 1; W. Brian Arthur, “Competing Technologies, Increasing Returns, and Lock-In by Historical Events,” Economic Journal 99 (1989): 127. 74. See Gregory J. Werden, “Network Effects and the Conditions of Entry: Lessons from the Microsoft Case,” Antitrust Law Journal 69 (2001): 87 (discussing Joe Bain’s and George Stigler’s differing definitions of “entry barriers”). 75. See Schumpeter, Capitalism, 81– 86 (describing a kind of competition as involving “creative destruction”). 76. Microsoft ¶¶ 568– 72, 714. 77. New York v. Microsoft Corp., 224 F.Supp.2d 76, 228 (D.D.C. 2002), aff’d, 373 F.3d 199, 1216– 25 (D.C. Cir. 2004). See United States v. Microsoft Corp., Second Modified Final Judgment ¶ I(E) (D.D.C. 2002, 2006, 2009). Carl Shapiro, a former deputy assistant attorney general, has criticized the final judgment as inadequate. Carl Shapiro, “Microsoft: A Remedial Failure,” Antitrust Law Journal 75 (2009): 739. 78. Microsoft ¶¶ 979– 80. 79. 253 F.3d at 49– 51. 80. Copyright Act, 17 U.S.C. § 102(a)(1) (protection of literary works); id., § 101 (definitions of computer programs and literary works). 81. Computer Associates Int’l v. Altai, Inc., 982 F.2d 693, 707– 9 (2d Cir. 1992). 82. Sega Enterprises Ltd. v. Accolade, Inc., 977 F.2d 1510 (9th Cir. 1992); Sony Computer Entertainment, Inc. v. Connectix Corp., 203 F.3d 596 (9th Cir.), cert. denied, 531 U.S. 871 (2000). 83. 17 U.S.C. § 1201(f) (2000). Congress has long looked favorably on reverse engineering. In the Semi-Conductor Chip Protection Act of 1984, Congress included a provision authorizing reverse engineering. 17 U.S.C. § 906 (2000). 84. United States v. Microsoft Corp., Modified Final Judgment ¶ III(D) (APIs) III(E), ¶ VI(B) (protocols) (Civ. Action No. 98-1232 [CKK], Filed Sept. 7, 2006). 85. See Massachusetts v. Microsoft Corp., 373 F.3d 1199, 1204, 1250 (D.C. Cir. 2004). 86. 547 U.S. 388 (2006). 87. See, for example, Carbice Corp. v. American Patents Development Corp., 283 US 27 (1931). See also Dawson Chemical Co. v. Rohm & Hass Co., 448 U.S. 176, 191– 97 (1980) (discussing the evolution of the patent misuse doctrine). Lasercomb America, Inc. v. Reynolds, 911 F.2d 970 (4th Cir. 1990) is the leading case on the copyright misuse doctrine. In section 271 of the patent act, Congress limited the scope of the patent misuse doctrine as it applies to tying arrangements, by requiring, as a condition of a determination of misuse, that the patentee possess market power in the relevant market for the patent or patented product. The courts, however, have developed a doctrine of copyright misuse that is not similarly constrained. Since Microsoft dominates the market for PC operating systems, its vulnerability to an assertion of copyright misuse would not be affected by transposing the market power limitation of patent misuse into the copyright misuse doctrine. 88. Practice Management Information Co. v. American Medical Ass’n, 121 F.3d 516, 517– 21 (9th Cir. 1997). 89. David S. Evans and Richard Schmalensee, “Some Economic Aspects of Antitrust Analysis in Dynamically Competitive Industries,” National Bureau of

notes to chapters 9

271

Economic Research, Working Paper 8268, May 2001, 7– 15, available at http:// www.nber.org/papers/w8268. See also Richard A. Posner, “Antitrust in the New Economy,” Antitrust Law Journal 68 (2001): 926– 30, and Cosmo Graham and Fiona Smith, eds., Competition, Regulation, and the New Economy (New York, Hart Publishing, 2004). The Evans and Schmalensee thesis is examined critically by the authors of this book in a journal article. See Daniel J. Gifford and Robert T. Kudrle, “Antitrust Approaches to Dynamically Competitive Industries in the United States and the European Union,” Journal of Competition Law & Economics 7 (2011): 695–731. 90. AMC Report and Recommendations, 31– 42. 91. Thomas O. Barnett, “Maximizing Welfare through Technological Innovation,” George Mason Law Review 15 (2008): 1194. See also Frances E. Marshall, “U.S. Department of Justice Guidance Regarding Ex Ante Patent Licensing Policies of Standard-Setting Organizations,” PLI 2nd Annual Patent Law Institute 211, 226 (2008) (reporting remarks of Gerald F. Masoudi, deputy assistant attorney general, on “dynamic efficiency, or leapfrog competition”). 92. Thomas O. Barnett, “Interoperability between Antitrust and Intellectual Property,” George Mason Law Review 14 (2007): 864. 93. See Carl Shapiro, “Antitrust, Innovation, and Intellectual Property,” Testimony before the Antitrust Modernization Commission, 2– 3 (November 8, 2005). 94. Shapiro “Antitrust,” 3– 5. 95. Shapiro “Antitrust,” 7. 96. Shapiro “Antitrust,” 5– 6. 97. Shapiro, “Microsoft: A Remedial Failure,” 739. Microsoft was found to have monopolized by behavior that impeded Netscape and Java from developing into rival platforms. The district court found that they would not develop into rival platforms in the foreseeable future. And it was uncertain whether either Netscape or Java would ever develop into a rival platform. So a restoration of competition did not involve an actual change in competitive conditions. Rather, it involved restoring a competitive threat that might or might not develop into an actual threat. Shapiro believed that the district court did not look broadly enough to find competitive threats and thus failed to take action fostering the development of those threats. 98. See, for example, Philip Lowe, “Competition and Innovation Policy,” GCP (July 2008, Release One). See also Christian Ahlborn and David S. Evans, “The Microsoft Judgment and Its Implications for Competition Policy towards Dominant Firms in Europe,” Antitrust Law Journal 75 (2009): 916; Pierre Larouche, “The European Microsoft Case at the Crossroads of Competition Policy and Innovation: Comment on Ahlborn and Evans,” Antitrust Law Journal 75 (2009): 942– 43, 948. 99. See, for example, Massachusetts v. Microsoft Corp., 373 F.3d 1199, 1212 (D.C. Cir. 2004) (referring to “Positive network effects”). 100. Commission Decision of 24.03.2004 relating to a proceeding under Article 82 of the EC Treaty (Case COMP/C-3/37.792 Microsoft), Article 6(a). Apparently no computer manufacturer installed the Windows version lacking the WMP. William H. Page and John E. Lopatka, The Microsoft Case: Antitrust, High Technology, and Consumer Welfare (Chicago: University of Chicago Press, 2007), 83.

notes to chapter 10

272

ChApter 10 1. The observer was Randy Tritell, director of international affairs, Federal Trade Commission. Rachel Brandenburger and Randy Tritell, “Global Antitrust Policies: How Wide Is the Gap? Interview,” Concurrences 1 (2012): 4. 2. Jeremy Grant and Damien Neven, “The Attempted Merger between General Electric and Honeywell: A Case Study of Transatlantic Conflict,” Journal of Competition Law and Economics 1 (2005): 630. 3. James Langenfeld, “Non-Horizontal Merger Guidelines in the United States and the European Commission: Time for the U.S. to Catch Up?” George Mason Law Review 16 (2009): 851– 84. 4. Of course, political factors can still obtrude into the process, notably from the competition commissioner (who is appointed and can be replaced by the president of the Commission) or from the influence of other directorates. Bruce Lyons, “Reform of European Merger Policy,” Review of International Economics 12 (2004): 246– 61. 5. See the discussion in chapter 5 (predatory pricing). The Ninth and Eleventh Circuits have indicated that they are willing to consider pricing above average variable cost but below average total cost as capable of predatory use. The Supreme Court in Brooke Group has ruled that pricing above “incremental cost” cannot be deemed predatory. Thus if the Court’s reference to incremental cost is construed as referring to a short-run incremental cost, the circuit courts would be unable to deem pricing as predatory if the prices in question exceeded a short-run version of incremental cost, such as marginal cost, average variable cost, or average avoidable cost. But if the Court’s reference to incremental cost could be construed as referring to long-run average incremental cost, then the circuit courts would have the flexibility to adopt a position similar to the European position. 6. Because the US focus is on the market share involved in exclusive arrangements, while in the European Union the focus is on the market share of the supplier offering the exclusive arrangements, there remains some possibility of different antitrust evaluations of similar arrangements. 7. See Evans and Schmalensee, “Some Economic Aspects of Antitrust Analysis in Dynamically Competitive Industries,” and discussion in chapter 9 in Dynamic Industries section. 8. Commission Decision Case COMP/C-3/37792, of 24 Mar. 2004, relating to a proceeding under Article 102 of the EC Treaty [hereinafter Microsoft] at ¶¶ 465– 70. 9. Pinar Akman, in “Searching,” has presented strong evidence that this might not have been intended by some of those crafting the original language, and that total welfare could have been their intent. But she concedes that the language finally adopted has been almost universally interpreted as supporting a consumer over a total standard. 10. Philip Lowe, “Consumer Welfare and Efficiency— New Guiding Principles of Competition Policy?,” 13th International Conference on Competition and 14th European Competition Day, Munich, 20 March 2007. See discussion in chapter 2 in section titled “If ‘Welfare’ Is the Main Goal of Competition Policy, What Does It Mean?” 11. See chapter 8.

notes to chapter 10

273

12. Thomas A. Miller and Ryan W. Marth, “Promoting Greater Consistency in Single Party Conduct Policy: Is Section 5 of the FTC Act a ‘Third Way’ to Converge European and U.S. Interests?,” draft paper presented to the American Antitrust Institute’s 10th Annual Conference, Washington, DC, June 18, 2009. 13. Albert A. Foer, “Section 5 as Bridge toward Convergence,” remarks to an FTC workshop on Section 5, October 17, 2008, 2, 3. We concur with the observation of Areeda and Turner that “as a goal of antitrust policy, ‘fairness’ is a vagrant claim applied to any value that one happens to favor.” Phillip Areeda and Donald F. Turner, Antitrust Law 4 (New York: Aspen Law and Business, 1980), 21. 14. Areeda and Turner, Antitrust Law, 7, 8. 15. William E. Kovacic, “The Intellectual DNA of Modern U.S. Competition Law for Dominant Firm Behavior: The Chicago/Harvard Double Helix,” Columbia Business Law Review 1 (2007): 1– 80. Kovacic is using “Chicago” the way the term is commonly understood; by “Harvard,” however, he does not mean Mason and Bain but rather the newer Harvard thinking of Areeda, Turner, and Breyer. 16. Evans, “Why Different Jurisdictions Do Not (and Should Not) Adopt the Same Antitrust Rules,” Chicago Journal of International Law 10 (2009): 162– 82. 17. Federal Trade Commission, FTC Settles Charges of Anticompetitive Conduct against Intel, August 6, 2010. 18. Matsushita Elec. Industrial Co. v. Zenith Radio Corp., 475 US 574, 594 (1986). 19. Verizon Communications Inc. v. Law Offices of Curtis V. Trinko, LLP, 540 US 398, 414 (2004). 20. Brooke Group Ltd. v. Brown & Williamson Tobacco Corp., 509 US 209, 223 (1993). 21. Directive of the European Parliament and of the Council on Certain Rules Governing Actions for Damages under National Law for Infringements of the Competition Law Provisions of the Member States and of the European Union. Strasbourg, 11.6.2013 COM (2013), 404 final, 2013/0185 (COD). 22. See, for example, Whinston, “Tying, Foreclosure, and Exclusion,” 837– 59. 23. For example, Bolton, Brodley, and Riordan, “Predatory Pricing,” 2329, outlines a more complex view of predatory strategy than is usually considered by the courts within the framework of “a structured Rule of Reason that would focus enforcement on cases where conditions make predation strongly plausible and where market conduct makes anticompetitive conduct dangerously probable.” 24. DOJ Submission 3– 4. 25. Barack Obama, Statement for the American Antitrust Institute (Sept. 27, 2007), formerly available at http://www.antitrustinstitute.org/archives/files /aai-%20Presidential%20campaign%20-%20Obama% 209-07_092720071759 .pdf, cited in William Kolasky, “Reinvigorating Antitrust Enforcement in the United States: A Proposal,” Antitrust 22 (2008): 85– 90. 26. Jonathan Baker, “Preserving a Political Bargain: The Political Economy of the Non-Interventionist Challenge to Monopolization Enforcement,” Antitrust Law Journal 76 (2010): 606– 7. 27. The vertical restraints guidelines were issued in 1985. See U.S. Dep’t of Justice, Vertical Restraints Guidelines (1985), reprinted in Trade Regulation Reporter 4 (CCH) ¶ 13,105. Anne K. Bingaman, President Clinton’s assistant attorney gen-

notes to chapter 10

274

eral of antitrust, revoked them in 1993 in a highly publicized manner. See Anne K. Bingaman, Address to the ABA’s Antitrust Section, reprinted in Antitrust 65 and Trade Regulation Reporter 4 250 (BNA) (1993). 28. Christine Varney, Recorded Remarks, Panel: Re-Energizing Section 2 Enforcement, American Antitrust Institute Annual National Conference, available at http://www.antitrustinstitute.org/Archives/2008Conference audio.ashx. 29. Although President Obama made Commissioner Leibowitz chairman, the FTC Commission membership that lodged the complaint against Intel was the same as in the late Bush administration. 30. Jonathan Baker has developed a model of US political equilibrium on antitrust between those favoring more intervention and those opposing it. Baker, “Preserving a Political Bargain,” 606. Although Baker does not identify the surplus standard as a source of possible political disturbance, it seems to fit his general argument. There was a very strong and visible political attack on the employment of a total surplus standard in Canada. See, for example, Stephen F. Ross, “Afterword— Did the Canadian Parliament Really Permit Mergers That Exploit Canadian Consumers So the World Can Be More Efficient?” Antitrust Law Journal 65 (1997): 641. The concern Ross expressed can be defended on national interest grounds, but it is doubtful that consumers would be assuaged by knowing the profits stayed at home. After much controversy, Canada retained the total surplus standard. 31. For an example, see Buch-Hansen and Wigger, Politics of European Competition Regulation, 145. 32. US competition authorities often exhort other states to include incarceration among their penalties from their conviction that executives are far more effectively deterred by the threat of imprisonment than by fines. For an example, see Gregory Werden, “Sanctioning Cartel Activity: Let the Punishment Fit the Crime,” European Competition Journal 5 (2009): 19– 36. Indeed, fines have proved to be highly ineffective as a deterrent. See John M. Connor, “Global Cartels Redux: The Lysine Antitrust Litigation,” in The Antitrust Revolution: Economics, Competition, and Policy, 5th ed., ed. John E. Kwoka Jr. and Lawrence J. White (Oxford: Oxford University Press, 2009), 300– 328. 33. Buch-Hansen and Wigger, Politics of European Competition Regulation, 124. 34. But overall the level of penalties is now estimated to be far too low to be an effective general deterrent. Connor, “Global Cartels Redux.” 35. Cartels and the limitation of private damages against whistle-blowing participants loom large in the Commision’s 2013 proposed directive on private action. On April 17, 2014, the European Parliament adopted the directive. Directive of the European Parliament and of the Council on certain rules governing actions for damages under national law for infringements of the competition law provisions of the Member States and of the European Union. 36. Rachel Brandenburger, “Transatlantic Antitrust: Past and Present,” Remarks as Prepared for St. Gallen International Competition Law Forum, St. Gallen, Switzerland. 37. Nihat Aktas, Eric de Bodt, Marieke Delanghe, and Richard Roll, “Market Reactions to European Merger Regulation: A Reexamination of the Protectionism Hypothesis,” available at http://ssrn.com/abstract=1961188.

notes to chapter 10

275

38. C. Crystal Jones-Starr, “Note and Comment: Community-Wide v. WorldWide Competition: Why European Enforcement Agencies Are Able to Force American Companies to Modify Their Merger Proposals and Limit Their Innovations,” Wisconsin International Law Journal 17 (1999): 161. 39. In re Wood Pulp Cartel v. E.C. Commission, [1988] 4 Common Mkt L.R. 901; Merger Regulation Art. 1; United States v. Aluminum Co. of America, 148 F.2d 416, 443– 444 (2d Cir. 1945); 15 U.S.C. § 6a. 40. This consideration goes beyond mergers and underpins some observations that the European Union now calls the shots on firm strategy worldwide more than does the United States exactly because it more often constrains firm behavior. See, for example, Joshua Wright’s transcript of a presentation by Christine Varney at a meeting of the American Antitrust Institute reported on the blog Truth on the Market, http://truthonthemarket.com/2009/02/22/. 41. Jones-Starr, “Note and Comment,” 145, 168. 42. There was, however, a floated threat from the US government that such action would be interpreted as protectionism in favor of Airbus, and retaliatory trade action might be taken. See Steven Pearlstein, “Boeing Yields to Key EU Demand to Win Approval of McDonnell Deal,” Washington Post, July 23, 1997, D10. 43. For a discussion of the conflict from several perspectives, see Daniel J. Gifford and E. Thomas Sullivan, “Can International Antitrust Be Saved for the PostBoeing Merger World? A Proposal to Minimize International Conflict and Rescue Antitrust from Misuse,” Antitrust Bulletin 45 (2000): 55. 44. Eleanor M. Fox, “Antitrust Regulation across National Borders: The United States of Boeing versus the European Union of Airbus,” Brookings Review 16 (Winter 1998): 30– 32. 45. Brandenburger and Tritell, “Global Antitrust Policies,” 3– 11. 46. Bernard M. Hoekman and Michel M. Kostecki, The Political Economy of the World Trading System: The WTO and Beyond, 3rd ed. (Oxford: Oxford University Press, 2009), 370– 412. 47. E. M. Graham and J. D. Richardson, “Conclusions and Recommendations,” in Global Competition Policy, ed. E. M. Graham and J. D. Richardson (Washington, DC: Institute for International Economics, 1997): 559– 60. 48. Robert T. Kudrle, “Governing Economic Globalization: The Pioneering Experience of the OECD,” Journal of World Trade 46 (June 2012): 695– 731. 49. Kudrle, “Governing.” 50. Oliver Budzinski, “The International Competition Network: Prospects and Limits on the Road towards International Competition Governance,” Competition and Change 8 (2004): 223– 42; Marie-Laure Djelic and Thibaut Kleiner, “The International Competition Network: Moving towards Transnational Governance,” in Transnational Governance: Institutional Dynamics of Regulation, ed. MarieLaure Djelic and Kerstin Sahlin-Andersson (Cambridge: Cambridge University Press, 2006). 51. The United States and the European Union were also deeply committed to another major ICN activity: “missionary work” to assist newly founded competition agencies to function properly and to serve as effective advocates within their governments for a greater reliance on market forces.

Bibliography Adelman, M. A. A&P: A Study in Price-Cost Behavior and Public Policy. Cambridge, MA: Harvard University Press, 1959. Aggarwal, Aradhna. “Patterns and Determinants of Anti-Dumping: A World-Wide Perspective.” Indian Council for Research on International Economic Relations, Working Paper No. 113, 2003, 25– 30. http://icrier.org/pdf/wp113.pdf. Ahlborn, Christian, and David S. Evans. “The Microsoft Judgment and Its Implications for Competition Policy towards Dominant Firms in Europe.” Antitrust Law Journal 75 (2009): 887– 932. Ahlborn, Christian, and Carsten Grave. “Walter Eucken and Ordoliberalism: An Introduction from a Consumer Welfare Perspective.” Competition Policy International 2, no. 2 (2006): 196– 217. Akman, Pinar. “Searching for the Long-Lost Soul of Article 82 EC.” Oxford Journal of Legal Studies 29, no. 2 (2009): 267– 303. Aktas, Nihat, Eric de Bodt, Marieke Delanghe, and Richard Roll. “Market Reactions to European Merger Regulation: A Reexamination of the Protectionism Hypothesis.” http://ssrn.com /abstract=1961188. Anderman, Steven. “Does the Microsoft Case Offer a New Paradigm for the ‘Exceptional Circumstances’ Test and Compulsory Copyright Licensees under EC Competition Law?” Competition Law Review 1 (2004): 7– 22. Andersen, Henriette K. B. “EC Merger Control Regulation as Applied in the De Havilland Case.” New York International Law Review 37, no. 87 (1994): 25– 44. Antitrust Modernization Commission. Report and Recommendations. Washington, DC: Antitrust Modernization Commission, 2007.

bibliography

278

Areeda, Phillip, and Herbert Hovenkamp. Antitrust Law: An Analysis of Antitrust Principles and Their Application. Boston: Little, Brown, 1996. ———. Antitrust Law: An Analysis of Antitrust Principles and Their Application. 2nd ed. New York: Aspen Publishers, 2000; Supp. 2004. ———. Antitrust Law: An Analysis of Antitrust Principles and Their Application. 3rd ed. New York: Aspen Publishers, 2008. Areeda, Phillip, and Donald F. Turner. Antitrust Law. New York: Aspen Law and Business, 1980. ———. “Predatory Pricing and Related Practices under Section 2 of the Sherman Act.” Harvard Law Review 88 (1975): 697– 733. Arthur, W. Brian. “Competing Technologies, Increasing Returns, and Lock-In by Historical Events.” Economic Journal 99 (1989): 116– 31. ———. Increasing Returns and Path Dependence in the Economy. Ann Arbor: University of Michigan Press, 1994. Baer, William J., and David Balto. “The Politics of Federal Antitrust Enforcement.” Harvard Journal of Law and Public Policy 23, no. 1 (1999): 113– 32. Bain, Joe S. Barriers to New Competition. Cambridge, MA: Harvard University Press, 1956. ———. Industrial Organization. New York: Wiley, 1959. ———. “The Profit Rate as a Measure of Monopoly Power.” Quarterly Journal of Economics 55 (1941): 271– 88. ———. “Workable Competition in Oligopoly.” American Economic Review (1950): 35– 47. Baker, Jonathan B. “Competitive Price Discrimination: The Exercise of Market Power without Anticompetitive Effects.” Antitrust Law Journal 70 (2003): 643– 54. ———. “Efficiencies and High Concentration: Heinz Proposes to Acquire BeechNut.” In The Antitrust Revolution: Economics, Competition, and Policy, 5th ed., edited by John E. Kwoka Jr. and Lawrence J. White, 150– 69. Oxford: Oxford University Press, 2009. ———. “Preserving a Political Bargain: The Political Economy of the NonInterventionist Challenge to Monopolization Enforcement.” Antitrust Law Journal 76 (2010): 605– 52. Barnett, Thomas O. “Interoperability between Antitrust and Intellectual Property.” George Mason Law Review 14 (2007): 859– 70. ———. “Maximizing Welfare through Technological Innovation.” George Mason Law Review 15 (2008): 1194– 204. Barron, David P. Business and Its Environment. 5th ed. Upper Saddle River, NJ: Pearson, Prentice Hall, 2006. Baumol, William J. “Predation and the Logic of the Average Variable Cost Test.” Journal of Law and Economics 39 (1996): 49– 72. Baumol, William J., and Daniel G. Swanson. “The New Economy and Ubiquitous Competitive Price Discrimination: Identifying Defensible Criteria of Market Power.” Antitrust Law Journal 70 (2003): 661– 86. Bird, Richard, Jorge Martinez-Vazquez, and Benno Torgler. “Tax Effort in Developing Countries and High Income Countries.” Economic Analysis and Policy 38, no. 1 (2008): 55– 71.

bibliography

279

Bolton, Patrick, Joseph F. Brodley, and Michael H. Riordan. “Predatory Pricing: Strategic Theory and Legal Policy.” Georgetown Law Journal 88 (2000): 2239– 330. Bork, Robert H. The Antitrust Paradox: A Policy at War with Itself. New York: Free Press, 1978. Brand, Donald R. Corporatism and the Rule of Law: A Study of the National Recovery Administration. Ithaca, NY: Cornell University Press, 1988. Brandenburger, Rachel. “Recent Developments in Merger Control: Views from the U.S. Department of Justice’s Antitrust Division.” Remarks as Prepared for the International Bar Association’s 16th Annual Competition Conference, Florence, Italy, September 14, 2012. ———. “Transatlantic Antitrust Past and Present.” St. Gallen International Competition Forum, May 2010. Brandenburger, Rachel, and Randy Tritell. “Global Antitrust Policies: How Wide Is the Gap? Interview.” Concurrences 1 (2012): 3– 11. Brennan, Timothy J. “Bundled Rebates as Exclusion Rather Than Predation.” Journal of Competition Law and Economics 4 (2008): 335– 74. ———. “The Complement Market/ Final Consumer Distinction: Exclusion and Predation in the U.S. Department of Justice Section 2 Report.” GCP: The Online Magazine for Global Competition Policy 10 (October 8, 2008, Release One): 1– 12. ———. “Predation, Exclusion, and Complement Market Monopolization.” Global Competition Policy Antitrust Chronicle 7 (2009). ———. “‘Vertical Market Power’ as Oxymoron: Horizontal Approaches to Vertical Antitrust.” George Mason Law Review 12 (2004): 895– 922. Brodley, Joseph F. “The Economic Goals of Antitrust: Efficiency, Consumer Welfare, and Technological Progress.” New York University Law Review 62 (1987): 1020– 53. Bruzzone, Ginevra, and Marco Boccaccio. “Impact-Based Assessment and Use of Legal Presumptions in EC Competition Law: The Search for the Proper Mix.” World Competition: Law and Economics Review 32 (2009): 465– 84. Buch-Hansen, Hubert, and Angela Wigger. The Politics of European Competition Regulation: A Critical Political Economy Perspective. New York: Routledge, 2011. Budzinski, Oliver. “The International Competition Network: Prospects and Limits on the Road towards International Competition Governance.” Competition and Change 8 (2004): 223– 42. Büthe, Tim. “The Politics of Competition and Institutional Change in European Union: The First Fifty Years.” In Making History: European Integration and Institutional Change at Fifty, edited by Sophie Meunier and Kathleen McNamara, 175– 94. Oxford: Oxford University Press, 2007. Carlton, Dennis W. “Does Antitrust Need to Be Modernized?” Antitrust Division, U.S. Department of Justice Economic Analysis Group, Discussion Paper No. 07-6, January 2007. ———. “Revising the Horizontal Merger Guidelines.” Journal of Competition Law and Economics 6, no. 3 (2010): 619– 52. Carlton, Dennis W., and William D. Bishop. “Merger Policy and Market Definition under the EC Merger Regulation.” Fordham Corporate Law Institute 422 (1993). Edited by Barry E. Hawk, 1994.

bibliography

280

Carlton, Dennis, Patrick Greenlee, and Michael Waldman. “Assessing the Anticompetitive Effects of Multiproduct Pricing.” National Bureau of Economic Research Working Paper No. 14199, 2008. Carlton, D. W., and J. M. Perloff. Modern Industrial Organization. 4th ed. Boston: Pearson/Addison Wesley, 2005. Chen, Zhiqi. “Buyer Power: Economic Theory and Antitrust Policy.” In Research in Law and Economics, vol. 22, edited by Richard O. Zerbe and John B. Kirkwood, 17– 40. Bingley, West Yorkshire, England: Emerald Group Publishing, 2000. Coate, Malcolm. “A Test of Political Control of the Bureaucracy: The Case of Mergers.” Economics and Politics 1 (2002): 1– 15. Connor, John M. “Global Cartels Redux: The Lysine Antitrust Litigation.” In The Antitrust Revolution: Economics, Competition, and Policy, 5th ed., edited by John E. Kwoka Jr. and Lawrence J. White, 300– 328. Oxford: Oxford University Press, 2009. Corr, Christopher F. “Trade Protection in the New Millennium: The Ascendancy of Antidumping Measures.” Northwestern Journal of International Law and Business 18, no. 1 (1997): 51– 110. Corts, Kenneth. “Third-Degree Price Discrimination in Oligopoly: All-Out Competition and Strategic Commitment.” RAND Journal of Economics 29, no. 2 (1998): 306– 23. Crane, Daniel A. “Mixed Bundling, Profit Sacrifice, and Consumer Welfare.” Emory Law Journal 55 (2006): 423– 86. Crane, Daniel A., and Graciela Miralles. “Toward a Unified Theory of Exclusionary Vertical Restraints.” Southern California Law Review 84 (2011): 605– 60. Derclaye, Estelle. “Abuse of a Dominant Position and Intellectual Property Rights: A Suggestion to Reconcile the Community Courts’ Case Law.” In Competition, Regulation, and the New Economy, edited by Cosmo Graham and Fiona Smith, 55– 75. Portland, OR: Hart Publishing, 2004. ———. “The IMS Health Decision and the Reconciliation of Copyright and Competition Law.” European Law Review 29 (2004): 687– 97. ———. “The IMS Health Decision: A Triple Victory.” World Competition: Law and Economics Review 27 (2004): 397– 405. Devlin, Alan, and Bruno Peixoto. “Reformulating Antitrust Rules to Safeguard Societal Wealth.” Stanford Journal of Law, Business, and Finance 13 (2008): 225– 59. Djelic, Marie-Laure, and Thibaut Kleiner. “The International Competition Network: Moving towards Transnational Governance.” In Transnational Governance: Institutional Dynamics of Regulation, edited by Marie-Laure Djelic and Kerstin Sahlin-Andersson, 287– 307. Cambridge: Cambridge University Press, 2006. Drauz, Goetz, Stephen Mavroghenis, and Sara Ashall, “Recent Developments in EU Merger Control, 1 September 2009– 31 August 2010.” Journal of European Competition Law and Practice 2 (2011): 44– 61. Dupuit, Jules. On Tolls and Transport Charges 7. Translated by Elizabeth Henderson. International Economic Papers No. 11, 1962. Easterbrook, Frank H. “The Limits of Antitrust.” Texas Law Review 63 (1984): 1– 40. ———. “Predatory Strategies and Counterstrategies.” University of Chicago Law Review 48 (1981): 263– 337.

bibliography

281

———. “Vertical Arrangements and the Rule of Reason.” Antitrust Law Journal 53 (1984): 135– 73. Economides, Nicholas, and Ioannis Lianos. “The Elusive Antitrust Standard on Bundling in Europe and in the United States in the Aftermath of the Microsoft Cases.” Antitrust Law Journal 76 (2009): 483– 567. Elhauge, Einer. “Defining Better Monopolization Standards.” Stanford Law Review 56 (2003): 253– 59. ———. “Tying, Bundled Discounts, and the Death of the Single Monopoly Theory.” Harvard Law Review 123 (December 2009): 397– 481. Elzinga, Kenneth, and David E. Mills. “The Lerner Index of Monopoly Power: Origins and Uses.” American Economic Review 101, no. 3 (2011): 558– 64. European Commission. DG Discussion Paper on the Application of Article 82 of the Treaty to Exclusionary Abuses. Brussels: European Commission, 2005. ———. Proposal for a Directive of the European Parliament and of the Council on Certain Rules Governing Actions for Damages under National Law for Infringements of the Competition Law Provisions of the Member States and of the European Union. Strasbourg, 11.6.2013 COM (2013) 404 final. Evans, David S. “Why Different Jurisdictions Do Not (and Should Not) Adopt the Same Antitrust Rules.” Chicago Journal of International Law 10 (2009): 161– 82. Evans, David S., and A. Jorge Padilla. “Tying under Article 82 EC and the Microsoft Decision: Comment on Dolmans and Graf.” World Competition: Law and Economics Review 27 (2004): 507– 8. Evans, David S., and Richard Schmalensee. “Some Economic Aspects of Antitrust Analysis in Dynamically Competitive Industries.” National Bureau of Economic Research, Working Paper 8268, May 2001. Farrell, Joseph, and Michael L. Katz. “The Economics of Welfare Standards in Antitrust.” Competition Policy International 2, no. 2 (2006). Farrell, Joseph, and Carl Shapiro. “Antitrust Evaluation of Horizontal Mergers: An Economic Alternative to Market Definition.” B. E. Journal of Theoretical Economics 10, no. 1 (2010), Article 9. Federico, Giulio. “When Are Rebates Exclusionary?” European Competition Law Review 26 (2005): 477– 80. Fisher, Alan A., and Robert H. Lande. “Efficiency Considerations in Merger Enforcement.” California Law Review 71, no. 6 (1983): 1580– 1696. Foer, Albert A. “The Politics of Antitrust in the United States: Public Choice and Public Choices.” University of Pittsburgh Law Review 62 (2001): 475– 501. ———. “Section 5 as Bridge toward Convergence.” Remarks to an FTC Workshop on Section 5, October 17, 2008. Fox, Eleanor M. “Antitrust Regulation across National Borders: The United States of Boeing versus the European Union of Airbus.” Brookings Review 16 (Winter 1998): 30– 32. Fumagalli, Chiara, and Massimo Motta. “Exclusive Dealing and Entry When Buyers Compete.” American Economic Review 96 (2006): 785– 95. Galbraith, John K. American Capitalism: The Concept of Countervailing Power. Boston: Houghton Mifflin, 1952. Garza, Deborah A. “A Comparative Analysis of the Clinton Antitrust Program and Suggestion of Changes to Come.” Antitrust 15 (2001): 64– 67.

bibliography

282

Geradin, Damien. “The Decision of the Commission of 13 May 2009 in the Intel Case: Where Is the Foreclosure and Consumer Harm?” Journal of European Competition Law and Practice 1 (2010): 112– 22. Geradin, Damien, and Nicolas Petit. “Price Discrimination under EC Competition Law: Another Antitrust Doctrine in Search of Limiting Principles?” Journal of Competition Law and Economics 2 (2006): 479– 531. Gerber, David J. Global Competition: Law, Markets, and Globalization. Oxford: Oxford University Press, 2010. ———. Law and Competition in Twentieth Century Europe: Protecting Prometheus. Oxford: Clarendon Press, 1998. Gifford, Daniel J. “The Antitrust/Intellectual Property Interface: An Emerging Solution to an Intractable Problem.” Hofstra Law Review 31 (2002): 363– 418. ———. “Assessing Secondary-Line Injury under the Robinson-Patman Act: The Concept of ‘Competitive Advantage.’” George Washington Law Review 44 (1975): 48– 85. ———. “Primary-Line Injury under the Robinson-Patman Act: The Development of Standards and Erosion of Enforcement.” Minnesota Law Review 64, no. 1 (1979): 1– 72. ———. “Rethinking the Relationship between Antidumping and Antitrust Laws.” American University Journal of International Law and Policy 6 (1991): 277– 323. ———. “The Role of the Ninth Circuit in the Development of the Law of Intent to Monopolize.” Notre Dame Law Review 61 (1986): 1021– 51. Gifford, Daniel J., and Robert T. Kudrle. “Alternative National Merger Standards and the Prospects for International Cooperation.” In The Political Economy of International Trade Law: Essays in Honor of Robert E. Hudec, edited by Daniel L. M. Kennedy and James D. Southwick, 208– 47. Cambridge: Cambridge University Press, 2002. ———. “Antitrust Approaches to Dynamically Competitive Industries in the United States and the European Union.” Journal of Competition Law and Economics 7 (2011): 695– 731. ———. “European Competition Law and Policy: How Much Latitude for Convergence with the United States?” Antitrust Bulletin 48 (2003): 727– 80. ———. “Rhetoric and Reality in the Merger Standards of the United States, Canada, and the European Union.” Antitrust Law Journal 72 (2005): 423– 69. Gifford, Daniel J., and E. Thomas Sullivan. “Can International Antitrust Be Saved for the Post-Boeing Merger World? A Proposal to Minimize International Conflict and to Rescue Antitrust from Misuse.” Antitrust Bulletin (2000): 55– 118. Gilbert, Richard, and Daniel L. Rubinfeld. “Revising the Horizontal Merger Guidelines: Lessons from the US and the EU.” In Competition Policy and Regulation: Recent Developments In China, the US, and Europe, edited by Michael Faure and Xinzhu Zhang. Cheltenham, UK: Edward Elgar, 2011. Glossner, Christian. The Making of the German Post-War Economy. London: I. B. Tauris, 2010. Goldman, Calvin, John D. Bodrug, and Mark A. A. Warner. “Canada.” In Global Competition Policy, edited by E. M. Graham and J. D. Richardson, 47– 85. Washington, DC: Institute for International Economics, 1997. Goyder, D. G. EC Competition Law. 2nd ed. Oxford: Oxford University Press, 1993.

bibliography

283

Graham, Cosmo, and Fiona Smith, eds. Competition, Regulation, and the New Economy. New York: Hart Publishing, 2004. Graham E. M., and J. D. Richardson. “Conclusions and Recommendations.” In Global Competition Policy, edited by E. M. Graham and J. D. Richardson, 547– 78. Washington, DC: Institute for International Economics, 1997. Grant, Jeremy, and Damien Neven. “The Attempted Merger between General Electric and Honeywell: A Case Study of Transatlantic Conflict.” Journal of Competition Law and Economics 1 (2005): 595– 633. Greenlee, Patrick, and David S. Reitman. “Competing with Loyalty Discounts.” Economic Analysis Group Discussion Paper No. 04-2, 2005. http://ssrn.com /abstract=502303. Greenlee, Patrick, David S. Reitman, and David S. Sibley. “An Antitrust Analysis of Bundled Loyalty Discounts.” Economic Analysis Group Discussion Paper No. 04-13, 2005. http://ssrn.com/abstract=600799. Hansen, Hugh C. “Robinson-Patman Law: A Review and Analysis.” Fordham Law Review 51 (1983): 1113– 218. Harris, Barry C., and Joseph J. Simons. Focusing Market Definition: How Much Substitution Is Necessary? Research in Law and Economics 12. Greenwich, CT: JAI Press, 1989. Hartz, Louis. Liberalism in America. New York: Harcourt Brace, 1956. Hawk, Barry E., and Laraine L. Laudate. “Antitrust Federalism in the United States and Decentralization of Competition Law Enforcement in the European Union: A Comparison.” Fordham International Law Journal 20, no. 1 (1996): 18– 49. Hibner, Don T. “Attempts to Monopolize: A Concept in Search of Analysis.” Antitrust Law Journal 34 (1967): 165– 77. Hoekman, Bernard M., and Michel M. Kostecki. The Political Economy of the World Trading System: The WTO and Beyond. 3rd ed. Oxford: Oxford University Press, 2009. Horton, Thomas J. “The New United States Horizontal Merger Guidelines: Devolution, Evolution, or Counterrevolution?” Journal of European Competition Law and Practice 2, no. 2 (2011): 158– 64. Hovenkamp, Herbert J. “Exclusive Joint Ventures and Antitrust Policy.” Columbia Business Law Review 1 (1995): 52– 61. ———. “The FTC’s Anticompetitive Pricing Case against Intel.” CPI Antitrust Chronicle 2 (February 28, 2010). Hovenkamp, Herbert J., and Erik Hovenkamp. “Exclusionary Bundled Discounts and the Antitrust Modernization Commission.” Antitrust Bulletin 53 (2008): 517– 53. Huo, Wei. “Introduction and Critical Analysis of Anti-Dumping Regime and Practice in China Pending Entry of WTO: Transition toward a WTO-Modeled Trade Legal Mechanism.” International Law 36 (2002): 197– 214. Hylton, Keith N. “Intel and the Death of U.S. Antitrust Law.” CPI Antitrust Journal 2 (February 28, 2010): 1– 6. Hylton, Keith, and Fei Deng. “Antitrust around the World.” Antitrust Law Journal (2007): 271– 341. Irwin, Douglas, and Randall Krozner. “Interests, Institutions, and Ideology in Securing Policy Change: The Republican Conversion to Trade Liberalization after Smoot Hawley.” Journal of Law and Economics 42 (1999): 643– 74.

bibliography

284

Jacobson, Jonathan. “Exclusive Dealing, ‘Foreclosure,’ and Consumer Harm.” Antitrust Law Journal 70, no. 9 (2002): 311– 69. Janssens, Thomas. “European Merger Reform— More EU-US Convergence Than Meets the Eye?” Mergers and Acquisitions Newsletter 4, no. 2 (2004). Jones-Starr, C. Crystal. “Note and Comment: Community- Wide v. World-Wide Competition: Why European Enforcement Agencies Are Able to Force American Companies to Modify Their Merger Proposals and Limit Their Innovations.” Wisconsin International Law Journal 17 (1999): 161– 72. Joskow, Paul L., and Alvin K. Klevorick. “A Framework for Analyzing Predatory Pricing Policy.” Yale Law Journal 89 (1979): 213– 70. Kaysen, Carl, and Donald F. Turner. Antitrust Policy: An Economic and Legal Analysis. Cambridge, MA: Harvard University Press, 1959. Kessler, Friedrich, and Richard H. Stern. “Competition, Contract, and Vertical Integration.” Yale Law Journal (1959): 1– 129. Kim, Sejin. “The Korea Fair Trade Commission’s Decision on Microsoft’s Tying Practice: The Second-Best Remedy for Harmed Competitors.” Pacific Rim Law and Policy Journal 16 (2007): 375– 404. Klein, Benjamin, and Andres V. Lerner. “The Expanded Economics of Free-Riding: How Exclusive Dealing Prevents Free-Riding and Creates Undivided Loyalty.” Antitrust Law Journal 74 (2007): 473– 519. ———. “The Law and Economics of Bundled Pricing: LePage’s, PeaceHealth, and the Evolving Antitrust Standard.” Antitrust Bulletin 53, no. 3 (2008): 555– 85. Klein, Benjamin, and Kevin M. Murphy. “Exclusive Dealing Intensifies Competition for Distribution.” Antitrust Law Journal 75 (2008): 433– 66. Kolasky, William. “Reinvigorating Antitrust Enforcement in the United States: A Proposal.” Antitrust 22 (2008): 85– 90. Kovacic, William E. “The Intellectual DNA of Modern U.S. Competition Law for Dominant Firm Behavior: The Chicago/Harvard Double Helix.” Columbia Business Law Review 1 (2007): 1– 80. Kovacic, William E., and Carl Shapiro. “Antitrust Policy: A Century of Economic and Legal Thinking.”Journal of Economic Perspectives 14, no. 1 (2001): 43– 60. Krattenmaker, Thomas G., and Steven C. Salop. “Anticompetitive Exclusion: Raising Rivals’ Costs to Achieve Power over Price.” Yale Law Journal 96 (1986): 209– 93. Krotoszynski, R. J., Jr. “‘History Belongs to the Winners’: The Bazelon-Leventhal Debate and the Continuing Relevance of the Process/ Substance Dichotomy in Judicial Review of Agency Action.” Administrative Law Review 58, no. 4 (2006): 995– 1016. Kudrle, Robert T. “Governing Economic Globalization: The Pioneering Experience of the OECD.” Journal of World Trade 46 (June 2012): 695– 731. Lambert, Thomas A. “Appropriate Liability Rules for Tying and Bundled Discounting.” Ohio State Law Journal 72 (2011): 1– 75. ———. “Evaluating Bundled Discounts.” Minnesota Law Review 89 (2005): 1688– 723. Lande, Robert H. “Chicago’s False Foundation: Wealth Transfers (Not Just Efficiency) Should Guide Antitrust.” Antitrust Law Journal 58 (1989): 1– 13. ———. “Wealth Transfers as the Original and Primary Concern of Antitrust: The Efficiency Interpretation Challenged.” Hastings Law Journal 34 (1982): 65– 151.

bibliography

285

Langenfeld, James. “Non-Horizontal Merger Guidelines in the United States and the European Commission: Time for the United States to Catch Up?” George Mason Law Review 16 (2009): 851– 83. Larouche, Pierre. “The European Microsoft Case at the Crossroads of Competition Policy and Innovation: Comment on Ahlborn and Evans.” Antitrust Law Journal 75 (2009): 933– 64. Lerner, A. P. “The Concept of Monopoly and the Measurement of Monopoly Power.” Review of Economic Studies 1 (1934): 157– 69. Lowe, Philip. “Competition and Innovation Policy.” GCP (July 2008, Release One). Lyons, Bruce R. “Could Politicians Be More Right Than Economists?” Revised CCR Working Paper CCR 02-1, 2002. ———. “Reform of European Merger Policy.” Review of International Economics 12 (2004): 246– 61. Maier-Rigaud, Frank P. “Article 82 Rebates: Four Common Fallacies.” European Competition Journal 2 (2006): 85– 100. Maier-Rigaud, Frank P., and Dovile Vaigauskaite. “Prokent/Tomra, a Textbook Case? Abuse of Dominance under Perfect Information.” EC Competition Policy Newsletter 2 (2006): 19– 24. Manne, Geoffrey E., and Joshua D. Wright. “Innovation and the Limits of Antitrust.” Journal of Competition Law and Economics 6, no. 1 (2010): 153– 202. Marshall, Frances E. “U.S. Department of Justice Guidance Regarding Ex Ante Patent Licensing Policies of Standard-Setting Organizations.” PLI 2d Annual Patent Law Institute (2008), 211– 26. McGee, John S. “Predatory Price Cutting: The Standard Oil (N.J.) Case.” Journal of Law and Economics 1 (1958): 137– 69. Meade, Alissa A. “Modeling a European Competition Authority.” Duke Law Journal 46 (1996): 153– 96. Messina, Michele. “Article 82 and the New Economy: Need for Modernisation?” Competition Law Review 2 (2006): 73– 98. Miller, Thomas A., and Ryan W. Marth. “Promoting Greater Consistency in Single Party Conduct Policy: Is Section 5 of the FTC Act a ‘Third Way’ to Converge European and U.S. Interests?” Draft paper presented to the American Antitrust Institute’s 10th Annual Conference, Washington, DC, June 18, 2009. Montagnani, Maria Lillà. “Remedies to Exclusionary Innovation in the High-Tech Sector: Is There a Lesson from the Microsoft Saga?” World Competition: Law and Economics Review 30, no. 4 (2007): 623– 44. Monti, Giorgio. “EC and New Economy Markets, Article 82.” In Competition, Regulation, and the New Economy, edited by Cosmo Graham and Fiona Smith, 17– 54. Oxford: Hart Publishing, 2004. Motta, Massimo. Competition Policy: Theory and Practice. New York: Cambridge University Press, 2004. Nalebuff, Barry. “Bundling as an Entry Barrier.” Quarterly Journal of Economics 119 (2004): 159– 87. Neven, Damien. “Competition Economics and Antitrust in Europe.” Economic Policy 48 (2006): 741– 91. Ordover, Janusz A., and Robert D. Willig. “An Economic Definition of Predation: Pricing and Product Innovation.” Yale Law Journal 91 (1981): 8– 53.

bibliography

286

Packer, Randal C. “Unbundling Scope-of-Permission Goods: When Should We Invest in Reducing Entry Barriers?” University of Chicago Law Review 72 (2005): 1– 36. Page, William H., and John E. Lopatka. The Microsoft Case: Antitrust, High Technology, and Consumer Welfare. Chicago: University of Chicago Press, 2007. Palim, Mark R. A. “The Worldwide Growth of Competition Law: An Empirical Analysis.” Antitrust Bulletin 43, no. 1 (1998): 105– 45. Patterson, Donna E., and Carl Shapiro. “Transatlantic Divergence in GE/Honeywell: Causes and Lessons.” Antitrust (Fall 2001): 18– 26. Pearlstein, Steven. “Boeing Yields to Key EU Demand to Win Approval of McDonnell Deal.” Washington Post, July 23, 1997. Pera, Alberto. “Changing Views of Competition, Economic Analysis, and EC Antitrust Law.” European Competition Journal 1 (2008): 127– 68. Petit, Nicolas. “From Formalism to Effects? The Commission’s Communication on Enforcement Practices in Applying Article 82 EC.” World Competition: Law and Economics Review 32 (2009): 1– 19. Pitofsky, Robert. “Antitrust Policy in a Clinton Administration.” Antitrust Law Journal 62 (1993): 217– 23. ———. “Efficiencies in Defense of Mergers: Two Years After.” George Mason Law Review 7 (1999): 485– 94. ———. “Proposals for Revised United States Merger Enforcement in a Global Economy.” Georgetown Law Journal 81 (1992): 195– 250. Pollack, Mark A. The Engines of European Integration: Delegation, Agency, and Agenda Setting. Oxford: Oxford University Press, 2003. Posner, Richard A. “Antitrust in the New Economy.” Antitrust Law Journal 68 (2001): 926– 43. ———. Antitrust Law: An Economic Perspective. 2nd ed. Chicago: University of Chicago Press, 2001. ———. “Competition, Regulation, and the New Economy.” Antitrust Law Journal 68 (2001): 925– 43. ———. “Intellectual Property: The Law and Economics Approach.” Journal of Economic Perspectives 19 (2005): 57– 73. ———. “A Program for the Antitrust Division.” University of Chicago Law Review 38 (1971): 500– 536. Rasmusen, Eric B., J. Mark Ramseyer, and John Shepard Wiley Jr. “Naked Exclusion.” American Economic Review 81 (December 1991): 1137– 45. Roberts, Gary L., and Steven C. Salop. “Efficiency Benefits in Dynamic Merger Analysis.” 1994 (unpublished). ———. “Efficiencies in Dynamic Merger Analysis: A Summary.” World Competition: Law and Economics Review 19, no. 5 (1996): 5– 18. Rodrik, Dani. “Feasible Globalizations.” In Globalization: What’s New?, edited by Michael Weinstein. New York: Columbia University Press, 2005. ———. “How Far Will International Integration Go?” Journal of Economic Perspectives 14, no. 1 (2000): 177– 86. Rooney, Paula. “Microsoft Calls Sun Copycat.” CMPnetasia.com. May 24, 2002. Rooney, Paula, and Barbara Darrow. “Sun’s Strategy— Vendor Takes on Microsoft on Desktop, Server Fronts with Upgrades.” Computer Reseller News 102 (May 20, 2002).

bibliography

287

Ross, Stephen F. “Afterword— Did the Canadian Parliament Really Permit Mergers That Exploit Canadian Consumers So the World Can Be More Efficient?” Antitrust Law Journal 65 (1997): 641– 52. Rowe, Frederick M. Price Discrimination under the Robinson-Patman Act. Boston: Little, Brown, 1962. Salop, Steven C. “Economic Analysis of Exclusionary Vertical Conduct: Where Chicago Has Overshot the Mark.” In How the Chicago School Overshot the Mark: The Effect of Conservative Economic Analysis on U.S. Antitrust, edited by Robert Pitofsky, 141– 55. Oxford: Oxford University Press, 2008. ———. “Question: What Is the Real and Proper Antitrust Welfare Standard? Answer: The True Consumer Welfare Standard.” Loyola Consumer Law Review 22, no. 3 (2006): 3– 27. Scherer, F. M., and David Ross. Industrial Market Structure and Economic Performance. 3rd ed. Boston: Houghton Mifflin, 1990. Schlesinger, Arthur M., Jr. The Politics of Hope. Princeton, NJ: Princeton University Press, 1957. Schmalensee, Richard. “Should New Merger Guidelines Give UPP Market Definition?” Antitrust Chronicle (December 2009): 2– 7. ———. “Sunk Costs and Antitrust Barriers to Entry.” American Economic Review 94, no. 2 (2004): 471– 75. Schumpeter, Joseph A. Capitalism, Socialism, and Democracy. New York: Harper, 1942. Schutzman, Meredith. “Antidumping and the Continued Dumping and Subsidy Offset Act of 2000: A Renewed Debate.” Cardozo Journal of International and Comparative Law 11 (2004): 1069– 97. Shapiro, Carl. “Antitrust, Innovation, and Intellectual Property.” Testimony before the Antitrust Modernization Commission, 2– 3 (November 8, 2005). ———. “Microsoft: A Remedial Failure.” Antitrust Law Journal 75 (2009): 739– 72. Snyder, David. “Mergers and Acquisitions in the European Community and the United States: A Movement toward a Uniform Enforcement Body.” Law and Policy in International Business 29 (1997): 115– 44. Sosnick, Stephen, “A Critique of Concepts of Workable Competition.” Quarterly Journal of Economics 72 (1958): 380– 423. Stefanadis, Christodoulos. “Selective Contracts, Foreclosure, and the Chicago School View.” Journal of Law and Economics 41 (1998): 429– 50. Stern, Yedida Z. “A General Model for Corporate Acquisition Law.” Journal of Corporate Law 26 (2001): 675– 716. Steuer, Richard M. “Capitalist Welfare.” Antitrust 15 (2001): 4– 5. Stigler, George J. “The Origins of the Sherman Act.” Journal of Legal Studies (1985): 1– 12. ———. “A Theory of Oligopoly.” Journal of Political Economy 72 (1964): 44– 61. ———. The Theory of Price. London: Macmillan, 1966. Stockum, Stephen. “An Economist’s Margin Notes: The Antitrust Writings of Timothy Muris.” Antitrust (2002): 60– 63. Sykes, Alan O. “The Least Restrictive Means.” University of Chicago Law Review 70 (2003): 403– 19. Triplett, Jack, and Barry Bosworth. Productivity in the U.S. Services Sectors: New

bibliography

288

Sources of Economic Growth. Washington DC: Brookings Institution Press, 2004. U.S.-EU Merger Working Group. Best Practices on Cooperation in Merger Investigation. http://ec.europa.eu/competition/international/bilateral/eu_us.pdf. Van Bergeijk, Peter A. G., and Erik Kloosterhuis. Modeling European Mergers: Theory, Competition Policy, and Case Studies. Northampton: Edward Elgar, 2006. van den Bergh, Robert. “The Subsidiary Principle and the EC Competition Rules: The Costs and Benefits of Decentralization.” In Constitutional Law and Economics of the European Union, edited by Dieter Schmidtchen and Robert Cooter. London: Edward Elgar, 1997. Vernail, Robert M. “One Step Forward, One Step Back: How the Pass-On Requirement for Efficiencies Benefits in FTC v. Staples Undermines the Revisions to the Horizontal Merger Guidelines Efficiencies Section.” George Mason Law Review 7, no. 1 (1998): 133– 62. Vesterdorf, Bo. “Judicial Review in EC Competition Law: Reflections of the Role of the Community Courts in the EC System of Competition Law Enforcement.” Competition Policy International 1 (2005): 3– 27. Vickers, John. “Abuse of Market Power.” Economic Journal 115 (2005): F244– F261. Viscusi, W. Kip, John M. Vernon, and Joseph Harrington Jr. Economics of Regulation and Antitrust. 3rd ed. Cambridge, MA: MIT Press, 2000. Waelbroeck, Michel. “Price Discrimination and Rebate Policies under EU Competition Law.” Fordham Corporate Law Institute (1995): 147– 60. Werden, Gregory J. “Demand Elasticities in Antitrust Analysis.” Antitrust Law Journal 66 (1997– 98): 363– 414. ———. “An Economic Perspective on the Analysis of Merger Efficiencies.” Antitrust 11 (Summer 1997): 11– 16. ———. “Network Effects and the Conditions of Entry: Lessons from the Microsoft Case.” Antitrust Law Journal 69 (2001): 87– 111. ———. “Sanctioning Cartel Activity: Let the Punishment Fit the Crime.” European Competition Journal 5 (2009): 19– 36. ———. “Should the Agencies Issue New Merger Guidelines? Learning from Experience.” George Mason Law Review 16 (2009): 839– 49. Whinston, Michael D. “Exclusivity and Tying in U.S. v. Microsoft: What We Know and Don’t Know.” Journal of Economic Perspectives 15 (2001): 63– 80. ———. “Tying, Foreclosure, and Exclusion.” American Economic Review 80, no. 4 (1990): 837– 59. White, Lawrence J. “The Growing Influence of Economics and Economists on Antitrust: An Extended Discussion.” New York University Law and Economics Working Papers No. 119, 2008. ———. “Market Power and Market Definition in Monopolization Cases.” In ABA Section of Antitrust Law. Issues in Competition Law and Policy (2008): 913– 24. Wigger, Angela. “The Political Role of Transnational Experts in Shaping EU Competition Policy.” Legisprudence 3, no. 3 (2009): 251– 75. Wigger, Angela, and Andreas Nölke. “Enhanced Roles of Private Actors in EU Business Regulation and the Erosion of Rhenish Capitalism.” Journal of Common Market Studies 45, no. 2 (2007): 487– 513. Wilks, Stephen. “Agencies, Networks, Discourses, and the Trajectory of European

bibliography

289

Competition Enforcement.” European Competition Journal 3, no. 2 (2007): 437– 64. Williamson, Oliver E. The Economic Institutions of Capitalism: Firms, Markets, Relational Contracting. New York: Free Press, 1985. ———. “Economies as an Antitrust Defense: The Welfare Tradeoffs.” American Economic Review 58 (1968): 18– 36. ———. “Economies as an Antitrust Defense Revisited.” University of Pennsylvania Law Review 125 (1977): 699– 736. ———. “Predatory Pricing: A Strategic and Welfare Analysis.” Yale Law Journal 87 (1977): 284– 340. World Trade Organization. World Trade Report. Geneva: WTO, 2007. Yamey, B. S. “Predatory Price Cutting: Notes and Comments.” Journal of Law and Economics 15 (1972): 129– 42. Yde, Paul L., and Michael G. Vita. “Merger Efficiencies: Reconsidering the ‘PassingOn’ Requirement.” Antitrust Law Journal 64 (1996): 735– 47.

Index Page numbers in italics indicate figures. AAC. See average avoidable cost A&P (Great Atlantic & Pacific Tea Company), 70 Abbott Laboratories, Inc., Doe v. (2009), 264n36 Abbott Laboratories, Inc., Ortho Diagnostic Systems, Inc. v. (1996), 125, 134, 142– 44, 145– 46, 153, 263n15 A. B. Dick Co., Henry v. (1912), 167– 68, 170 abuse of dominance: collective dominance standard and, 34, 53, 57, 237n82; consumer choice issue in, 135– 36; definition, 10– 11, 32– 33, 153, 207, 264n43; duty to deal (or license) in relation to, 166, 174– 78; efficient competitor standard and, 64; exclusive-supply contracts and, 109– 11, 118, 264n43; intention in, 80; monopolization compared with, 10– 13, 187; predation

analysis and, 93– 94; price discrimination and, 75– 77, 78– 79; price-squeeze issue in, 97; social welfare goal and, 32– 34; TFEU’s clauses on, 74– 75, 78; tying agreements as, 179– 80. See also monopolization AB Volvo v. Erik Veng (UK) Ltd. (1988), 166– 67, 174 Accor/Wagons-Lits (1992), 238n89 Acer Corp., 133, 134 Aérospatiale-Alenia/de Havilland (1991): political and economic influences on outcome, 54, 212, 225n126, 232n2; reasons for rejecting merger, 53, 56, 238n91; states’ special interests in, 22 Ahlborn, Christian, 220n50, 230n52 AIC (Baumol’s abbreviation). See long-run average incremental cost

index

Airbus (firm), 214, 275n42 Airtours v. Commission (2002), 35– 36, 37 Akman, Pinar, 14, 227n14, 272n9 Akzo Chemie BV v. Commission (1991), 78, 79– 80, 92– 93 Alcoa (United States v. Aluminum Co. of America, 1945): ambiguity of decision, 11; efficiency considered in, 43; as foundational in monopolization cases, 6, 221n68; market share considered in, 30, 256n31; “monopoly power” used in, 228n27; price-squeeze issue in, 96 Allied Orthopedic Appliances, Inc. v. Tyco Health Care Group LP (2010), 119– 21, 122, 258n11 Altai, Inc., Computer Associates v. (1992), 188 AMC. See Antitrust Modernization Commission AMD (microprocessors firm), 121, 126, 133– 36, 206, 261n59 American Airlines (United States v. AMR Corp., 2003), 88, 91, 251n20, 252n34 American Antitrust Institute, 207, 275n40 American Tobacco Co. v. United States (1911), 5, 69, 83, 244– 45n17 American Tobacco Co. v. United States (1946), 30 Analysis of Proposed Consent Order to Aid Public Comment (FTC), 258n19 Anheuser-Busch, FTC v. (1960), 246n36 Anheuser-Busch, Inc. (FTC 1957), 72– 73, 246n47 Antidumping Acts (US, 1916 and 1921), 69– 70, 245n23 antidumping laws, 67, 69– 70, 243n7, 245n23 antitrust, use of term, 218n8 Antitrust Division (US), 5– 6 antitrust law (US): ambiguities in, 205– 8; bundled discounts and future development of, 158– 60; consumer

292

welfare as focus of, 18, 25, 27– 29; decision making and interests in, 21; duties to deal or otherwise share with others, 162– 66; efficiency emphasized in, 6– 7, 16, 25, 43– 45, 73; EU competition law compared generally, 1– 3, 10, 13, 17, 29– 30, 63, 75, 81– 82, 194– 95, 197– 98; EU competition law differences explained, 205– 12; “fairness” concerns in, 12– 13; goals of, 6– 7, 11– 12, 157; ideal for, 149; ideological foundations of, 3– 8; institutional context of, 17– 18; international competitiveness in concerns about, 22; “least restrictive means” condition in, 52; new economy in context of, 191– 92, 193– 94; political equilibrium model of, 274n30; price competition encouraged in, 128; time horizon in, 211. See also Clayton Act; competition approaches; competition law; merger policy; monopolization; private litigation; Robinson-Patman Act; Sherman Act Antitrust Modernization Commission (AMC): on bundled discounts evaluation, 64, 140, 145– 49, 151– 54, 259n30; discount attribution rule of, 153, 208, 259n30, 264nn35– 36; efficient competitor standard of, 134, 145, 153– 54, 155; incremental cost undefined by, 264– 65n44; new economy industries approach of, 191, 193; on Robinson-Patman, 74; on 3M case, 16, 202 Antitrust Paradox (Bork), 26, 85, 205 antitrust revolution: changes due to, 14; concept of, 7; efficiencies issue and, 43– 44; EU’s resistance to, 15; international competitiveness and, 22; price discrimination considered, 68; Robinson-Patman disfavored in, 73– 74 Apple, Inc., 191, 193 Areeda, Phillip E., 159, 262n7, 263n15,

index

273n13. See also Areeda-Turner formulation Areeda-Turner formulation: bundled discounts analysis, 141, 142, 144; predation analysis, 84– 86, 87, 91, 92, 98, 134; US adoption of, 139– 40, 200 Arnold, Thurman, 5– 6 Aspen Skiing Co. v. Aspen Highlands Skiing Corp. (1985), 159, 164– 66, 194, 266n11 AT&T dissolution, 205 Austria, Tomra and market share in, 260nn36– 37 Automatic Canteen (1953), 263n9 average avoidable cost (AAC): cost standard compared with, 122; in fidelity rebates analysis, 127– 28; in predation analysis (EU), 92– 94, 95, 100, 253n65, 254n82; in predation analysis (US), 87– 88, 89, 91; US and EU compared, 98, 100, 140, 200– 201, 272n5 average variable cost (AVC): AAC and, 88, 89, 91, 100; in bundled discounts analysis, 141– 42, 144, 148, 149, 152; loyalty rebates and, 201; predatory pricing and, 78– 81, 84– 86, 87– 88, 92– 95, 99, 118, 122, 140, 200; as surrogate for marginal cost, 84, 98, 140, 261n55 Bain, Joe S., 219n36, 273n15 Baker, Jonathan, 274n30 Baltimore Gas & Electric Co. v. Natural Resources Defense Council, Inc. (1983), 231n65 Barnett, Thomas O., 191, 193 Barnhart v. Wilson (2002), 230n52 Baumol, William J.: on predatory pricing and costs, 86, 87– 90, 91, 94, 95, 122, 154, 253n59, 264– 65n44; on price discrimination and competition, 69 Bazelon, David L., 231n66 Beech-Nut proposed merger, 49– 50, 237n68

293

“Best Practices on Cooperation in Merger Regulations” (US-EU joint document, 2002, revd. 2011), 62 Block Exemption Regulation (EU, 1999, revd. 2010), 109, 256n36 BMI (1979), 7 Boeing/McDonnell-Douglas (1997), 40, 54, 56, 214, 215 Bolton, Patrick, 88– 90, 91, 94, 251n54, 273n23 Bork, Robert H.: on “consumer welfare,” 26, 28, 205– 6; on efficiency, 7, 43; mentioned, 27; on predatory pricing, 64, 85, 88; on price discrimination, 247n62 Boyden Power Brake Co. v. Westinghouse (1898), 266n3 Brandeis, Louis, 5 Brandt, Willy, 9 Brennan, Timothy J., 115– 16, 138, 223n98, 254n3, 257n49 Breyer, Stephen, 230n52 British Airways plc. v. Commission (2003), 76– 77 Broadcast Music, Inc. v. Columbia Broadcasting System (1979), 7 Brodley, Joseph F., 88– 90, 91, 251n54, 273n23 Bronner (Oscar Bronner GmbH & Co. v. Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG, 1998), 175– 76, 177– 78 Brooke Group Ltd. v. Brown & Williamson Tobacco Corp. (1993): Clayton Act considered in, 73– 74; “incremental cost” in, 264– 65n44, 272n5; predation analysis in, 85– 87, 89, 99, 100, 140; on pricing challenge, 65 Brown Shoe v. United States (1966), 12, 42, 233n27, 238n91 Brunswick Corp., Concord Boat Corp. v. (2000), 118– 19, 122, 136 bundled discounts: aggregated discount approach to, 145– 48; assumption of monopoly and competitive good in, 130, 136, 145; complex and

index

bundled discounts (continued) troublesome nature of, 140– 43; discount attribution rule and, 153, 208, 259n30, 264nn35– 36; efficient competitor standard for, 64, 134, 145– 47, 153– 56; in EU, 150– 52, 158– 60; exclusionary potential of, 142– 43, 152– 57, 202– 3; features and types of, 101– 3, 139, 143, 257– 58n3; impediment competition concept and, 158, 265n48; market share in relation to, 147, 154– 55, 259n30; monopolization effect of, 139, 143– 44, 148, 152– 53, 154– 55, 203; Non-Horizontal Guidelines on, 60, 241n136; portfolio effects and, 58– 59; predatory pricing in relation to, 139– 40, 141, 144, 152, 153– 54; “profit sacrifice” approach to, 149, 150, 154; proposed standard for, 159– 60; range effects memorandum on, 58; rivalry standard in relation to, 157– 58; “safe harbor” issue in, 148, 152; sellers’ incentives for, 66; suction effect in, 151, 155– 56; transparent pricing issue in, 265n45; in US, 143– 50, 158– 60; US and EU compared, 16– 17, 101, 151, 152– 57, 202– 3. See also single-product loyalty rebates; and specific cases Burger, Warren, 26 Busch, Andreas, 225n120 Bush administration (George W.): bundled discounts, 152; Merger Guidelines, 48; new economy industries approach, 191, 192– 93, 195; Obama’s attack on, 211– 12 business firms: consistent competition policies preferred by transnational companies, 22– 23; Cournot competition and, 232n6; definition of “dominant,” 10– 11; duties to deal with others, 162– 67, 174– 78; fairness and protection of smaller, 11– 13; merger policy and size of, 237n77; penalties for violators,

294

213, 274n32; price-cost margins of pre- and postmerger, 47– 48; pricetaking vs. price-making of, 220n52. See also competition; industry; market; new economy (innovationintensive) industries; products Canada: anticompetition law in, 218n12; efficiency as key objective in, 28 Carlton, Dennis W., 27, 29, 148, 228n31 cartelization: caution in attacks on, 13– 14; defense of, 8, 220n45; emergence and acceptance of, 8– 9; penalties for, 213, 274n32; transatlantic cooperation on, 213– 14; whistle-blowers and, 274n35. See also monopolization Cascade Health Systems v. PeaceHealth (2007), 148 chain stores, 6, 21, 69, 70– 71, 77 Chevron doctrine (1984), 37, 230n52 Chicago School: antitrust actions and application influenced by, 209– 11; on competition enhancement, 13; competition policy thinking of, 7; “counterrevolution” to, 48; on exclusive-supply contracts, 112– 14; legislation not generated by, 18; on merger policy issues, 40; others’ critiques of, 112– 13; S-C-P paradigm critiqued in, 7, 46 Clark, John Maurice, 10 Clayton Act (US, 1914, revd. 1950): enforcement of, 5; on exclusive-supply contracts and tying arrangements, 103, 104, 105, 109, 153, 201, 254n5, 255n13, 255n16; expanded scope of, 6; impetus for, 4– 5; “incipiency” standard in, 45– 46; on merger activity (Section 7), 41, 42– 43; patent misuse and, 168; political context, 28; on price discrimination and predatory pricing, 65, 67, 69– 70, 71, 73– 74, 83– 84, 199, 243n8; provisions of, 2, 15; substantiallessening-of-competition language

index

in, 57; on tying arrangements, 167. See also Robinson-Patman Act Clinton administration: competition policy concerns, 215; Horizontal Merger Guidelines, 7; Merger Guidelines, 48, 212, 234– 35n50 Colgate, United States v. (1919), 162– 63 collective dominance idea, 34, 53, 57, 237n82 Columbia Broadcasting System, Broadcast Music, Inc. v. (1979), 7 Columbia Steel, United States v. (1948), 6 Commercial Solvents (Istituto Chemioterapico Italiano SpA and Commercial Solvents Corporation v. Commission, 1974), 166 communications industries: network effects and technological ties in, 170– 71, 180– 82, 184– 87; price “squeezing” of, 96– 97, 254n78; reverse engineering encouraged in, 188, 270n83; value of components in, 269n72. See also Microsoft Corp., European Commission v.; Microsoft Corp., United States v.; platform software; technological developments Compagnie Maritime Belge SA v. Commission (2000), 78– 79 competition: bundled discounts’ impact on, 148; definitions, 71– 73, 173– 74; efficiency as key to preserving, 7; “for” vs. “within” market, 191; network effects’ impact on, 185– 86. See also competition approaches; market competition approaches: ambiguities in, 207– 8; decision making and interests in, 21– 23; ideological context, 3– 17, 205– 6; implications for transatlantic cooperation, 213– 15; institutional context, 17– 21, 209– 10; political influences on, 9– 10, 20– 21, 28, 54, 209, 211– 12, 232n2, 265n47; suspicion of monopoly as universal in, 39; terminology differ-

295

ences in, 218n8; transnational preferences of consistency in, 22– 23; US and EU compared, 1– 3, 10, 13, 17, 29– 30, 63, 75, 81– 82, 194– 95, 197– 98, 205– 12; welfare as main goal of, 25– 29; world organization for, 215– 16. See also antitrust law; bundled discounts; competition law; exclusive-supply contracts; “fairness”; intellectual property rights; merger policy; predation and predatory pricing; price discrimination; single-product loyalty rebates competition law (EU): ambiguities in, 207– 8; bundled discounts and future development of, 158– 60; decision making and interests in, 21– 23; duties to deal or otherwise share with others, 166, 174– 78; “fairness” in, 11– 13, 28, 52; goals of, 157– 58, 206; institutional context of, 18– 21; new economy in context of, 192– 94; rivalry standard in, 11, 13, 17, 33– 34, 54– 56, 58, 62, 76– 78, 100, 157– 58, 207; substantial shift in, 57; US antitrust law compared generally, 1– 3, 10, 13, 17, 29– 30, 63, 75, 81– 82, 194– 95, 197– 98; US antitrust law differences explained, 205– 12. See also abuse of dominance; antitrust law; competition approaches; merger policy; Merger Regulation; Treaty of Rome; Treaty on the Functioning of the European Union Computer Associates International v. Altai, Inc. (1992), 188 Concord Boat Corp. v. Brunswick Corp. (2000), 118– 19, 122, 136 conglomerate arrangements: EC’s investigation of, 231– 32n1; factors in decisions on, 58– 59; overview of, 39; Supreme Court’s hostility to, 42; in US and EU, summarized, 61– 62. See also bundled discounts; portfolio effects consumer betterment standard, 253n62

index

consumer choice: brand preferences and exclusivity in, 156– 57, 257n53; bundled discounts in relation to, 141; restrictions on, 135– 36 consumer demand test, 180, 195, 269n62 consumer harm idea, 253n73 consumer sovereignty standard, 157– 58 consumer surplus standard: Boeing/ McDonnell-Douglas in context of, 54; in current US antitrust policy, 31– 32; goals of, 229n37; high mark-up vs. generic products and, 263n24; merger-generated costsavings and, 48– 50, 53, 235– 36n55, 236n57; price discrimination and, 67– 68, 243– 44n9. See also total surplus standard; welfare consumer welfare: ambiguity of term, 25– 26, 205– 6; efficiency construed according to, 26– 29, 26, 27; in EU, 14, 22, 28, 52; “fairness” for, 28, 52, 70; as main goal of antitrust policy, 18, 25, 27– 29, 157; in Merger Guidelines, 51; as political accommodation, 265n47; pricing innovations in relation to, 138; rivalry and integration construed according to, 33– 34; in US, 18, 70; US and EU compared, 20– 21, 205– 7 Continental T.V., Inc. v. GTE Sylvania, Inc. (1977), 7 cooperative arrangements, 2, 19– 20 copyright law: antitrust law in relation to, 167– 71; duty to license and, 174– 78; fair use doctrine in, 188; foundational cases in, 266n4; interoperability favored in, 187– 89; limited protection for computer software in, 267n35; misuse of, 189– 90; provisions in, 161– 62. See also intellectual property rights Court of the First Instance (EU). See General Court Danish Crown/VestjyskeSlagterier (1999), 55– 56

296

de Havilland merger. See AérospatialeAlenia/de Havilland Dell Corp., 133, 134, 135 Denmark, antitrust cases in, 55– 56, 79, 81 Dentsply, International, Inc., United States v. (2005), 107– 8, 109, 256n30 Derclaye, Estelle, 268n54 Deutsche Börse AG, 242n144 Deutsche Telekom AG v. European Commission (2010), 97 Digital Millennium Copyright Act (US, 2000), 188 Directorate-General Competition (DG Comp): approval of policy, 212; economic professionals in, 20– 21; function and structure, 19, 222n76; information gathering of, 13– 14; on suction effect, 259n22 Directorate-General Enterprise and Industry, 54 discount attribution rule, 153, 208, 259n30, 264nn35– 36 discounts. See bundled discounts; single-product loyalty rebates Doe v. Abbott Laboratories (2009), 264n36 DOJ. See US Department of Justice “double helix,” 207, 273n15. See also Chicago School; Harvard School Dr. Miles Medical Co. v. John D. Park Co. (1911), 21 DuPont (United States v. E. I. du Pont de Nemours & Co., 1956), 30, 228n27 DuPont (United States v. E. I. du Pont de Nemours & Co., 1957), 12, 232n2 Dupuit, Jules, 243– 44n9 dynamic industries. See new economy (innovation-intensive) industries Easterbrook, Frank H., 27, 159 Eastman Kodak Co. v. Image Technical Services, Inc. (1992), 169 Eaton Corp., Meritor, LLC v. (2012), 108– 9, 119

index

eBay Inc. v. MercExchange, LLC (2006), 189 EC. See European Commission ECJ. See European Court of Justice ECN (European Competition Network), 20 economics and economists: common business practice vs. theories of, 60– 61; growth of private litigation and, 225n122; merger policies and, 40– 41; on opportunity costs, 90– 92, 154; on price discrimination, 67– 69, 243– 44n9, 246n36; role in EU competition law, 20– 21; role in US antitrust cases, 18; Soviet collapse and, 19– 20; static vs. dynamic analyses, 29, 41, 65. See also market; new economy (innovationintensive) industries economic welfare concept, 25 Economides, Nicholas, 265nn49– 50 economies of scale: components in, 147– 48; as entry barrier, 114; exclusion potential in, 113; fidelity rebates analysis and, 126, 130, 132; in mergers, 55, 58; in post– Civil War US, 4; US and Europe compared, 8 ECS (Engineering and Chemical Supplies [Epson and Gloucester] Ltd.), 80, 92– 93 EEC (European Economic Commission), 13– 14, 54, 212. See also Directorate-General Competition effects-based approach: in exclusivesupply contracts, 111– 12; formbased analysis as replacement of, 132– 33; form-based analysis replaced by, 14– 15, 20– 21, 197– 98; US and EU compared, 209 effects doctrine, 214 efficiency: aggregate welfare and net, 226– 27n5; ambiguity of meaning, 205– 6; bundled discounts in relation to, 141; chain stores and, 6, 21, 69, 70– 71, 77; concept, 25; condemned in Robinson-Patman, 73; consumer welfare as synonymous with, 26– 29,

297

26, 27; as entry barrier, 43, 233n27; EU Regulation and Guidelines on, 57, 59; exclusion compared with, 103; exclusive-supply contracts and, 64, 112– 14, 153; as goal in antitrust law, 6– 7, 16, 25, 43– 45, 73; market power limitation subject to, 12, 221n73; market viability and, 31; merger control delayed in favor of, 13– 14, 15; merger-generated costsavings equated with, 48– 50, 53, 235– 36n55, 236n57; monopolization justified by, 11; as negative factor in EU, 54– 56; in new economy industries, 191, 192; overview, 39– 40; prohibited merger justified by, 7– 8; static vs. dynamic, 29, 41; US and EU compared, 198; US courts on mergers, markets, and, 48– 51. See also economies of scale; Merger Guidelines; merger policy; Merger Regulation; products efficient competitor standard: in bundled discounts analysis, 64, 134, 145– 47, 153– 56; in loyalty rebates analysis, 64, 127– 29, 134; in predation analysis, 64– 65, 95, 134; questions about, 130– 31; scope of test, 64– 65; US and EU compared, 207– 8 Elhauge, Einer, 31, 150 Eli Lilly & Co., SmithKline Corp. v. (1978), 143, 144, 145, 154 Emerson, Ralph Waldo, 3 Engineering and Chemical Supplies (Epson and Gloucester) Ltd. (ECS), 80, 92– 93 entry barriers: assumptions about, 67; collusive excess profits and, 82; economies of scale as, 114; efficiency as, 43, 233n27; exclusivesupply contracts and, 115– 16, 119; loyalty rebates as, 130; network effects as, 186, 192– 93; predatory pricing and, 86, 88, 134, 140; product differentiation as, 150; single competitor’s market share and, 132; suction-effect concept and, 123– 25,

index

entry barriers (continued) 124, 258n20, 259n22; technological developments and, 144– 45, 150, 182– 84. See also monopolization Erhard, Ludwig, 9 essential facilities doctrine: EU cases and, 166, 174– 78; in intellectual property context, 15– 16; origin of, 164; US and EU compared, 162, 194– 95, 203– 4, 205; US cases and, 16, 164, 166, 194, 203 Établissements Consten S.A. und Grundig-Verkaufs Gmbh v. EEC Commission (1966), 13, 238n92 Ethyl Corp. v. EPA (1976), 231n66 EU. See European Union Eucken, Walter, 9, 15 Europe. See European Union European Commission (EC): authority over merger and acquisition agreements, 51– 57; on bundled rebates, 16– 17, 150– 52; compelled share approach of, 156– 57; competition law developed by, 19– 20; conglomerates investigated by, 231– 32n1; consumer welfare concerns of, 135– 36, 152, 206– 7; on duty to license, 174– 76, 204– 5; on duty to share intellectual property, 171– 74, 187; economic vs. political analyses as influence on, 20– 21, 209; efficient competitor standard of, 155, 208; enforcement priorities guidance of, 94, 95– 96, 253n73; essential facilities doctrine of, 166; on exclusivesupply contracts, 111– 12; on fidelity (loyalty) rebates, 122– 23, 125– 36, 202, 261n59, 261n64; on innovation and server software, 184, 269n71; judicial oversight vs. discretion of, 34– 38, 230– 31n56; market share standard of, 153; mergers blocked by, 15– 16, 40, 54– 56; on mergers of complementary producers, 59– 60; on network effects, 181, 184– 87; predation analysis and cost standard of, 88, 89, 92– 93, 94– 96, 98, 99–

298

100, 211, 253– 54n75; on price discrimination (selective pricing), 78– 80, 249– 50n101; on price-squeeze issue, 97; private litigation encouraged by, 20; on tying agreements, 180; vertical restraints regulations of, 14. See also specific cases European Competition Network (ECN), 20 European Court of Justice (ECJ): administrative review and oversight by, 14– 15, 34– 35, 36, 37– 38; changing views in, 16; competition law approach of, 14– 15, 19– 20; on duty to license or deal, 166– 67, 174– 78, 203– 5, 268n54, 268– 69n56; on exclusive-supply contracts, 110– 11, 121; on fidelity (loyalty) rebates, 131– 33, 138, 153, 202; predation analysis of, 80– 82, 93– 94, 96, 99, 200, 252n51, 253– 54n75; on price discrimination, 75– 77, 78– 79, 80– 82, 199– 200; on price-squeeze issue, 97. See also General Court; and specific cases European courts: administrative oversight by, 34– 38, 230– 31n56; collective dominance standard of, 34, 237n82; on dominance and market share, 33; on duty to license, 166, 174– 78; predation analysis of, 92– 94, 98, 99– 100; type one errors and, 210. See also European Court of Justice; General Court; and specific cases European Economic Commission (EEC), 13– 14, 54, 212. See also Directorate-General Competition European Union (EU): collective dominance idea in, 34, 53, 57, 237n82; “democratic deficit” in, 225n114; economic integration as goal in, 13– 14; economic thinking in, 4; freedom of contracting in, 8; key documents noted, 10– 11; mergers in context of unification of, 53– 54; model competition agency in, 19;

index

“national champions” favored in states of, 23; portfolio effects in, 55, 58, 239n102; post-WWII economic and political developments in, 8– 10; total surplus standard advocated in, 227n8. See also cartelization; competition law; European courts; ordoliberalism; and specific agencies, acts, and legal cases European Union Council, 52– 54. See also Merger Regulation Evans, David S., 4, 190– 92, 230n52, 266n3 exclusion: bundled discounts’ potential for, 142– 43, 152– 57, 202– 3; as category of harm to competition, 102– 3; courts’ treatment of, 159; efficiency compared with, 103; predatory pricing as, 94– 95 exclusive-supply contracts (or single branding): abuse of dominance in, 109– 11, 118, 264n43; efficiency and, 64, 112– 14, 153; as entry barriers, 115– 16, 119; in EU, 109– 12, 118; features of, 101– 3, 106– 7, 131– 32, 255n11; foreclosure effect and, 54– 55, 103– 6, 108– 9, 110, 201; as illegal, 217– 18n7; loyalty rebates in relation to, 118– 19, 121– 22, 131– 32, 137– 38; monopolization in relation to, 103– 5, 107– 9, 112; theoretical issues concerning, 66– 67, 112– 16, 137– 38; tying arrangements in relation to, 103– 5, 109, 153, 201, 254n5, 255n13, 255n16; in US, 103– 9; US and EU compared, 67, 116, 152– 53, 201, 264n43, 272n6. See also tying arrangements exclusivity: consumer-imposed, 156– 57, 257n53; in intellectual property, 162; US and EU laws on, compared, 152– 57 “fairness”: ambiguity of term, 207, 273n13; in consumer welfare, 28, 52, 70; to market participants (not final buyer), 97; in protecting rivals

299

from price competition, 100; to rivals of favored seller, 81; for smaller firms, 11– 13; social cost of, 82 fair use doctrine, 188 Farrell, Joseph, 28 Federal Communications Commission (FCC), 96– 97, 254n78 Federal Trade Commission (FTC, US): agenda enlarged, 5; Analysis of . . . Public Comment document of, 258n19; antitrust enforcement guidelines of, 7, 18 (see also Merger Guidelines); appointments to and structure of, 17; asset acquisitions concerns of, 42; on bundled discounts evaluation, 258n19; on chain stores, 70; cost definition of, 122; DOJ guidelines adopted by, 223n99; economic professionals in, 20; judicial oversight of, 49– 50, 219n17, 236n60; on loyalty rebates, 17, 121– 23, 257n1; Robinson-Patman enforcement by, 70, 72– 73, 74, 75, 199, 248n63. See also specific cases Federal Trade Commission Act (US), 4– 5, 12, 207 feudalism, 4 fidelity rebates. See single-product loyalty rebates Foer, Franklin, 261– 62n65 foreclosure effect: abusive conduct and, 111; bundled discounts and, 141, 146, 155, 156, 160, 241n136; exclusive-supply contracts and, 54– 55, 103– 6, 108– 9, 110, 201; forms of, 102; loyalty rebates, market share, and, 124– 25, 126, 128, 129– 30, 132– 33, 138; price discrimination and, 67; “substantial share” in analysis of, 115– 16, 119, 132 form-based approach: effects-based analysis as replacing, 14– 15, 20– 21, 197– 98; effects-based analysis replaced by, 132– 33; unlawful behavior examples in, 2, 217– 18n7; US and EU compared, 209. See also per se illegal category

index

Frank, Jerome New, 230n52 freedom: in buyer’s choice, 120, 151; of contracting, 8, 203; “economic,” 8, 11; to price, 65, 68 Freiburg School, 9– 10 FTC. See Federal Trade Commission Galbraith, John Kenneth, 71 Garza, Deborah A., 234– 35n50 GE/Honeywell (2001): bundling effects considered in, 239nn99– 100; conglomerate effects considered in, 39, 57– 58; disputes over, 214– 15; merger blocked, 40, 55, 56; policy changes after decision, 59– 60; reviews of, 211; US views on decision in, 15– 16, 58, 198 General Court (earlier, Court of First Instance, EU): administrative review and oversight by, 34, 35– 38; collective dominance idea of, 53, 237n82; competition policy mediation of, 19; on duty to license, 174, 175, 178; effects-based analytical approach of, 14– 15, 21; on exclusive-supply contracts, 110; on fidelity (loyalty) rebates, 131– 32, 153; on innovation and server software, 184; “new product” condition of, 178– 79; predation analysis of, 93– 94; on price-squeeze issue, 97; on rebate systems, 76– 77; on tying agreements, 179– 80. See also specific cases General Dynamics Corp., United States v. (1974), 42– 43, 233n21 General Electric. See GE/Honeywell General Motors (GM), 12, 30, 232n2 Gerber, David J., 18– 19, 158 Gerber Co., 49– 50 German Bundeskartelamt, 19 German Court of Justice, 13 Germany: cartels in, 8– 9; competition law in, 9; impediment competition concept in, 158; Mittelstand’s influence in, 22, 212; nationalization in, 220n48; Nazi regime in,

300

8– 9, 220n49; post-WWII social market economy of, 9– 10; potential of vertical restraints recognized in, 238n92; Tomra and market share in, 260nn36– 37; unification’s impact on, 212 Gilbarco, Inc., Omega Environmental, Inc. v. (1997), 106, 109 GM (General Motors), 12, 30, 232n2 Grave, Carsten, 220n50 Great Atlantic & Pacific Tea Company (A&P), 70 Greenberg, Morton Ira, 257– 58n3, 262n8, 263n15 Greenlee, Patrick, 148 Griffith, United States v. (1948), 6, 221n68 Grinnell Corp., United States v. (1966), 6, 30– 31, 182– 83, 221n68 Grundig-Consten (Établissements Consten S.A. und Grundig-Verkaufs Gmbh v. EEC Commission, 1966), 13, 238n92 GTE Sylvania, Continental T.V., Inc. v. (1977), 7 Hand, Learned, 11, 12, 30, 43, 256n31 Hart-Scott-Rodino Act (US, 1976), 237n77 Harvard School, 273n15. See also S-C-P (structure, conduct, performance) paradigm Heinz (H. J. Heinz Co., 2001), 49– 50, 237n68 Henry v. A. B. Dick Co. (1912), 167– 68, 170 Herfindahl-Hirschman Index (HHI), 44, 50, 57, 237n81, 240n114 Hewlett-Packard Corp., 133, 134, 135 Hilti AG v. Commission (1991), 78, 79 Hoffmann-La Roche & Co. v. Commission (1979), 75– 76, 110– 11, 132 Holmes, Oliver Wendell, 32 Honeywell International, Inc. v. Commission (2001). See GE/Honeywell horizontal arrangements: concept, 39; EEC’s caution toward, 13; EU

index

guidelines on, 14, 34, 53, 56, 57, 59, 237n82; goals in, 231– 32n1, 233n19; industrial performance in aftermath of, 40– 41; Supreme Court’s hostility to, 42; vertical restraints’ impact on, 114– 16, 257n49. See also cartelization; vertical arrangements Horizontal Merger Guidelines (EU, 2004), 14, 34, 53, 56, 57, 59, 237n82 Hovenkamp, Herbert J., 159, 262n7, 263n15, 265n46 Hylton, Keith N., 261– 62n65 ICN (International Competition Network), 215– 16, 275n51 ICPAC (International Competition Policy Advisory Committee, US), 215 ideology and doctrine: in EU, 8– 10, 13– 17; in merger policies, 40– 41; in US, 3– 8. See also abuse of dominance; monopolization Illinois Tool Works, Inc. v. Independent Ink, Inc. (2006), 170 impediment competition concept, 158, 265n48 IMS Health GmbH & Co. OHG v. NDC Health GmbH Co. KG (2004), 176– 78, 268n54, 268– 69n56 Inco-Falconbridge, 61 Independent Ink, Inc., Illinois Tool Works, Inc. v. (2006), 170 industry: cost determination, 84– 88 (see also specific types of cost); “old” vs. “new” categories, 193; supplier-purchaser bargaining in, 55; unilateral effects in, 41, 46– 47. See also business firms; economies of scale; new economy (innovationintensive) industries; products institutional context: EU competition policy in, 18– 21; US competition policy in, 17– 18. See also political institutions Intel cases: consumer welfare concerns in, 135– 36, 152, 206– 7; duopoly aspect of, 258n14; FTC complaint, 17,

301

136, 138, 153, 200, 201, 261– 62n65; implications for US, 209, 212; market share questions in, 125– 26, 133– 35, 261n59; Nicsand compared with, 258n14; proposed standard for, 160; single-product rebates and, 153– 54, 201, 202 intellectual property rights: antitrust law’s interface with (US), 167– 71; difficulties in reconciling EU competition law with, 37– 38; duty to license or deal and, 166– 67, 171– 75; international agreement on, 215; interoperability favored in, 187– 90; misuse of, 189– 90; new economy in context of, 190– 94; “new product” condition and, 178– 79; product integration and tying agreements in, 179– 80; protections summarized, 161– 62; summary, 194– 95; US, EU, and other approaches compared, 167, 180– 87, 203– 4; US exports based on, 23. See also copyright law; essential facilities doctrine; Microsoft Corp., European Commission v.; Microsoft Corp., United States v.; new economy (innovationintensive) industries; patent law; property rights International Competition Network (ICN), 215– 16, 275n51 International Competition Policy Advisory Committee (ICPAC, US), 215 International Harvester, 5 International Salt Co. v. United States (1947), 103, 105, 255n13 Internet Explorer, 190, 194, 269n66 interoperability, 187– 90 Interstate Commerce Act (US, 1887), 243n8 iPods and iTunes, 191, 193 Ireland: copyright law in, 174 Irish Sugar plc v. Commission (1999), 79– 80, 249– 50n101 Istituto Chemioterapico Italiano SpA and Commercial Solvents Corporation v. Commission (1974), 166

index

Jackson, Robert H., 5– 6 Java, 192, 271n97 Jefferson Parish Hospital District No. 2 v. Hyde (1984), 7, 104– 5, 180, 195, 255n13 judicial review, 34– 38, 49– 50, 219n17, 230n52, 230– 31n56, 236n60 Kaldor-Hicks criterion. See total welfare standards Kali & Salz (French Republic and Others v. EC Commission, 1998), 35– 36 Katz, Michael L., 28 Kaysen, Carl, 12, 221n73, 248n70 Kimberly-Clark/Scott paper merger (1995), 213 Klein, Benjamin, 66– 67, 145, 263n24 Kodak (Eastman Kodak Co. v. Image Technical Services, Inc., 1992), 169 Kovacic, William, 207 labor unions, 21– 22, 225nn119– 20 Ladbroke (Tierce Ladbroke SA v. Commission, 1997), 175, 176, 178 LAIC. See long-run average incremental cost Laker Airways v. Sabena, Belgian World Airlines (1984), 89, 251n25 Lande, Robert H., 27 Langenfeld, James, 60 Lasercomb America, Inc. v. Reynolds (1990), 270n87 Leegin Creative Leather Products, Inc. v. PSKS, Inc. (2007), 7 Leibowitz, Jon, 274n29 Lenovo Corp., 133, 134 LePage’s. See 3M Lerner, Abba P., 31, 179, 228nn30– 31 Lerner, Andres V., 145, 263n24 Leventhal, Harold, 231n66 Lianos, Ioannis, 265nn49– 50 Linkline Communications, Inc., Pacific Bell Telephone Co. v. (2008), 96– 97, 264n36 long-run average incremental cost (LAIC): bundled discounts analysis

302

and, 160; concept and potential, 89– 90, 202; loyalty rebates analysis and, 122, 127– 28, 138; multiproduct rebates and, 150– 51; in predation analysis, 89– 90, 95, 99, 100, 140, 200– 201 Lorain Journal Co. v. United States (1951), 165 Lowe, Philip, 206 loyalty rebates. See single-product loyalty rebates LRAIC. See long-run average incremental cost Magill (1995), 174, 175– 76, 177, 178, 179, 268n54 Maier-Rigaud, Frank P., 259n22 manifest error standard, 35– 36. See also judicial review Manne, Geoffrey E., 224n101 marginal cost. See average variable cost market: calculating for single vs. multiple, 86; courts on efficiency, mergers, and, 48– 51; definition, 34, 52– 53, 228n22; definition downgraded in merger policy, 46, 47– 48; forms of predation in, 88– 89, 94– 95; hypothetical monopolist considered in definition of, 44; incumbents vs. downstream competition in, 113– 14; “monopoly power” in, 30– 32; raising-rivals’-costs model of, 137– 38; representative buyer/ seller concept and, 128– 30, 137. See also abuse of dominance; competition; entry barriers; foreclosure effect; monopolization; products; unilateral effects; and specific types of cost market power: “dominant” position compared with, 33, 53– 54; exclusive-supply contracts in relation to, 104– 9, 255n13; forecasting merger’s impact on, 47; questions about, 157; sliding scale of, 49– 50; use of term, 31, 33, 228n31, 229n33. See also abuse of dominance; mar-

index

ket share; merger policy; monopolization; “power buyers” market share: bundled discounts in relation to, 147, 154– 55, 259n30; compelled share approach to, 156– 57; “contestable/noncontestable share” of, 125– 28, 129, 131– 32, 133, 136– 37, 140, 157, 260n35, 260nn36– 37, 261n59; definition, 30, 33; ECJ on, 35; exclusive-supply contracts and “substantial share” of, 103– 9, 111, 112, 115– 16, 119, 132, 152– 53, 201, 208; questions in Intel cases, 125– 26, 133– 35, 261n59; “required share” analysis and, 125, 127– 28, 129, 133– 34, 136– 37, 156; safe haven (or harbor) in, 46, 56 Masimo Corp. v. Tyco Health Care Group L.P. (2009), 120– 21, 122, 138, 153 Mason, Edward S., 10, 219n36, 273n15 Matsushita Electric Industrial Co. v. Zenith Radio Corp. (1986), 85, 210 McDonnell-Douglas. See Boeing/ McDonnell-Douglas McGahee v. Northern Propane Gas Co. (1988), 250n9 McGuire Act (US, 1952), 21 Mediaprint (Oscar Bronner GmbH & Co. v. Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG, 1998), 175– 76, 177– 78 Media-Saturn Holding GmbH (MSH), 261n59 Mercoid Corp. v. Mid-Continent Inv. Co. (1944), 168 Mercoid Corp. v. MinneapolisHoneywell Regulator Co. (1944), 168 Merger Guidelines (and revisions, US): ambiguity in, 206; consistent enforcement of, 18; “critical loss” analysis mentioned, 47; differences by presidential administrations, 48; DOJ’s adoption of, 223n99; efficiencies considered in, 43– 45, 51, 234– 35n50; EU regulation

303

compared with, 198; postmerger concentration considered in, 44; revisions of, 7– 8, 45– 48 merger policy: agency examination processes in, 238– 39n94; efficiency and consumer welfare in context of, 26– 29, 26, 27; efficiency in, generally, 41; extraterritoriality issues in, 213– 14; overview, 39– 40; potential merger board in political context of, 242n147; price-cost margins of firms and, 47– 48; principal issues, 40– 41; US and EU compared, 61– 62, 198; use of term, 218n8. See also cartelization; conglomerate arrangements; efficiency; horizontal arrangements; market power; vertical arrangements — EU: Aérospatiale merger rejected, 22, 225n126; background, 51– 52; community dimension in, 52– 53; delayed in favor of efficiency, 13– 14, 15; efficiency as negative factor in, 54– 56; guidelines and regulation described, 52– 54, 56– 61; key case in, 40; merger-generated cost savings for consumers in, 53– 54; unification interests in, 53– 54. See also Merger Regulation; and specific cases — US: background, 41– 43; courts on efficiency, markets, and, 48– 51; efficiency in, 43– 45, 48, 51; guidelines changes in 2010, 45– 48; hostility to horizontal concentration, 12; size of firms and, 237n77; “substantial share” used in, 115– 16. See also Merger Guidelines; and specific cases Merger Regulation (and revisions, EU): application of, 54– 56; on Commission discretion, 35, 230– 31n56; on consumer welfare, 14, 206; on efficiencies, 59, 206; “fairness” concerns in, 12– 13, 28, 52; market position in, 237n73; on market share concentration and domi-

index

Merger Regulation (continued) nant position, 56– 57; on mergergenerated cost savings for consumers, 53– 54; provisions of, 2, 51– 52, 56; revisions of, 15, 52– 54, 57; “significantly impede” clause in, 56, 57, 240n112, 240n116; US guidelines compared with, 198. See also Horizontal Merger Guidelines Meritor, LLC v. Eaton Corp. (2012), 108– 9, 119 Messina, Michele, 268n54, 268– 69n56 Metso/Svedala (2001), 55, 56 Michelin II (Manufacture française des pneutmatiques Michelin v. Commission, 2003), 110 Microsoft Corp., European Commission v. (2007): case law background and impact, 174– 79; components of decision, 36– 37, 38; duty to share intellectual property (server protocols) in, 171– 74, 187, 189, 203– 5; essential facilities doctrine applied to, 15– 16; innovation incentives in aftermath, 269n71; network effects in, 192– 93; product integration issue in, 179– 80; standard of proof in, 230– 31n56; US and other approaches compared with, 180– 87, 193– 95 Microsoft Corp., United States v. (2000): components of decision, 170– 71; consumer demand test in, 180, 195, 269n62; duty to disclose protocols imposed in, 187– 88; EU and other approaches compared with, 180– 87, 193– 95; evaluation of and questions concerning, 182– 84; exclusive-supply contracts in, 105, 107– 8, 109; per se rules exception in, 16, 171, 181, 183– 84, 190, 191– 92, 194– 95, 204; potential of appeal in, to harmonize copyright laws, 190 Miller-Tydings Act (US, 1937), 21 Mittelstand (medium-sized manu-

304

facturing and distribution firms, Germany), 22, 212 monopolization: abuse of dominance compared with, 10– 13, 187; antitrust/intellectual property interface and, 167– 71; behavioral nature of law on, 34; bundled discounts potential for, 139, 143– 44, 148, 152– 53, 154– 55, 203; of complement markets, 115; copyright law and restrictions in relation to, 170– 71; duty to license and, 174– 75; economies of scale and, 8; efficiency and consumer welfare in context of, 26– 29, 26, 27; efficiency as justification for, 11; exclusive-supply contracts in relation to, 103– 5, 107– 9, 112; foundational US cases, 6, 221n68; network effects linked to, 185– 86; predatory pricing in, 83; price discrimination and, 67– 68, 243– 44n9; refusal to deal linked to, 164– 65; special rules for, 107– 9; Supreme Court standard on, 30– 32, 163– 64, 182– 83, 256n31; technological design to maintain, 258n11; terminology for, 30– 32, 228n31, 229n33; tying arrangements linked to, 182– 83. See also abuse of dominance; cartelization monopoly power: application of term, 26, 30– 32, 107, 228n27, 229n33; exclusive-supply contracts and, 102, 114; Lerner index of, 179, 228n31; questions about, 154– 55, 157 monopsonist, 257n2 Monti, Giorgio, 11 Morton Salt Co. (1948), 71, 246n48 Motion Picture Patents Co. v. Universal Film Manufacturing Co. (1917), 167– 68 Motta, Massimo, 239n102 MSH (Media-Saturn Holding GmbH), 261n59 Muris, Timothy, 234– 35n50 Murphy, Kevin, 66– 67

index

Nalebuff, Barry, 145 National Industrial Recovery Act (NIRA, US, 1933), 5, 6 nation-states: antidumping laws across, 67, 243n7; economic integration among European, 13– 14; freedom of contracting recognized in, 8; merger considerations across, 39– 40, 53– 54; number with competition policies, 219n16; special interests of, 21– 22 Natural Resources Defense Council, Inc., Baltimore Gas & Electric Co. v. (1983), 231n65 Nestle-Perrier (1992), 53 Netherlands, Tomra and market share in, 260nn36– 37 Netscape Navigator: market share of, 105, 182, 269n66, 269n69; Microsoft as impediment to, 170, 171; Microsoft’s implications for, 108, 192, 271n97 network effects: advantages of, 170– 71, 204; anticompetitive potential of, 192– 94, 205; concept and role of, 184– 87; as entry barriers, 186, 192– 93; technological ties as illegal due to, 180– 82. See also new economy (innovation-intensive) industries new economy (innovation-intensive) industries: characteristics of, 190– 91; fostering innovation in, 161, 170– 71, 182, 224n101, 266n3; legal context of, 161– 62; special rules considered for, 191– 94, 195; US and EU compared, 16, 193– 94, 204– 5. See also intellectual property rights; network effects New Zealand, efficiency as key objective in, 28 Nicsand. See 3M NIRA (National Industrial Recovery Act, US, 1933), 5, 6 Non-Horizontal Guidelines (EU), 59– 62, 241n136 Northwest Wholesale Stationers, Inc.

305

v. Pacific Stationary & Printing Co. (1985), 7 Norway, Tomra and market share in, 260nn36– 37 NYSE Euronext, 242n144 Obama administration: antitrust policy changes and, 197, 211– 12; cost standard, 88; FTC commissioner appointee, 274n29; Merger Guidelines, 48; new economy industries approach of, 191, 192– 93, 195; “safe harbor,” 264n29 O’Connor, Sandra Day, 105 OECD (Organization for Economic Cooperation and Development), 58, 61, 211, 215 Office Depot, Inc. (Staples–Office Depot, 1997), 49, 231– 32n1, 236n60 Office of the Chief Economist (EC), 94– 95 oligopolies, 68– 69. See also unilateral effects Omega Environmental, Inc. v. Gilbarco, Inc. (1997), 106, 109 Oracle Corp., United States v. (2004), 241– 42n141 ordoliberalism (Ordoliberalismus): abuse of dominance vs. monopolization in, 10– 12, 14; bundling policy and, 157; development of, 9– 10; Eucken as proponent of, 9, 15; goals of, 220n50; impediment competition concept of, 158, 265n48; key element in approach to competition policy, 19, 22 Organization for Economic Cooperation and Development (OECD), 58, 61, 211, 215 organized labor, 21– 22, 225nn119– 20 Ortho Diagnostic Systems, Inc. v. Abbott Laboratories, Inc. (1996), 125, 134, 142– 44, 145– 46, 153, 263n15 Oscar Bronner GmbH & Co. v. Mediaprint Zeitungs und Zeitschriftenverlag GmbH & Co. KG (1998), 175– 76, 177– 78

index

Otter Tail Power Co. v. United States (1973), 165 Pabst Brewing Co., United States v. (1966), 6, 42, 43 Pacific Bell Telephone Co. v. Linkline Communications, Inc. (2008), 96– 97, 264n36 Patent Act (US), 161, 168– 69, 270n87 patent law: injunctive relief in infringement cases, 189; ordoliberalism and, 15; “pioneer” inventions in, 161, 266n3; tying arrangements and, 167– 70. See also intellectual property rights patent misuse doctrine, 167– 70, 189– 90, 270n87 Patterson, Donna E., 239nn99– 100 PeopleSoft, Inc., 241– 42n141 Perloff, J. M., 228n31 per se illegal category: examples of unlawful behavior, 2, 6, 217– 18n7; limits on, 7, 14; platform software as exception to, 16, 171, 181, 183– 84, 190, 191– 92, 194– 95, 204; resale price maintenance agreements in, 163; tying arrangements in, 169. See also form-based approach Philadelphia National Bank, United States v. (1963), 42 Pigou, Arthur C., 67, 243– 44nn8– 9 Pitofsky, Robert, 27, 227n12, 235– 36n50 platform software: encouraging new functionalities in, 183– 84, 191– 92, 269n62; as exception to per se rule, 16, 171, 181, 190, 191, 194– 95, 204; Microsoft’s implications for development of, 192, 271n97; in US vs. EU antitrust context, 204– 5. See also Windows operating system political institutions: EU competition law influenced by, 20, 21; Merger Guidelines in US context of, 48, 234– 35n50; post-WWII European, 9– 10 Pollack, Mark A., 232n2

306

portfolio effects: EU merger policy and, 55, 58– 59, 239n102; policy roundtable on, 58, 61 Posner, Richard A., 7, 18, 64, 159, 265– 66n53 post-Chicago theorists, 112– 13, 198, 210– 11 Post Danmark A/S v. Konkurrencerådet (2012), 79, 81 “power buyers”: bargaining by, 117– 18; bundled discounts for, 141; loyalty rebates for, 123– 25, 124, 258n20, 259n22; organizational structure considered, 262n8; use of term, 257n2; in vertical arrangements, 232n3. See also market power predation and predatory pricing: bundled discounts in relation to, 139– 40, 141, 144, 152, 153– 54; as category of harm to competition, 102– 3; Clayton Act on price discrimination and, 65, 67, 69– 70, 71, 73– 74, 83– 84, 199, 243n8; definition, 63, 83; efficient competitor standard in, 64– 65, 95, 134; enforcement priorities guidance on, 94, 95– 96, 253n73; entry barriers and, 86, 88, 134, 140; in EU, 92– 96; forms of, 80, 88– 89, 94– 95; “incremental cost” and, 264– 65n44; intention in, 92– 93; loyalty rebates tested by, 201– 2; opportunity costs in, 90– 92, 154; price-squeeze theory of, 96– 97; “profit sacrifice” approach to, 94– 95, 98, 99, 100; refusal to deal linked to, 164– 65; Rule of Reason and, 273n23; selective price-cutting treated similarly to, 80– 81; standards for, 87– 88; summary, 100; in US, 83– 92, 98– 99, 134; US and EU compared, 64, 83, 85, 96– 100, 138, 200– 201, 207– 8. See also average avoidable cost; long-run average incremental cost predatory bidding/buying, 86– 87 price competition: nonpredatory reasons for below-cost pricing,

index

251n23; price-squeezing in, 96– 97, 254n78, 264n36; Robinson-Patman as discouraging, 73– 74; “selective” price-cutting in, 16, 69, 78– 82, 92, 249n90; transparency in, 265n45. See also bundled discounts; exclusive-supply contracts; singleproduct loyalty rebates price-cutting: quantity-based discounts, 76– 77, 117; rebates and discounts viewed as, 118; “selective,” 16, 69, 78– 82, 92, 249n90; US standards on, 199. See also bundled discounts; single-product loyalty rebates price discrimination: as abuse by dominant firms, 75– 77, 78– 79; categories of, 243– 44n9; Clayton Act on predatory pricing and, 65, 67, 69– 70, 71, 73– 74, 83– 84, 199, 243n8; definition, 67; economic and legal background, 67– 69; in EU, 74– 82; horizontal price-fixing and, 2, 217– 18n7; overall assessment of, 82; price difference equated with, 246n36; primary-line effects in, 71– 73, 76, 77; secondary-line effects in, 70– 71, 73– 74, 75, 76; selective price-cutting as, 80– 81; sellers’ incentives to reduce price via, 66; in US, 6, 69– 74, 244– 45n17, 246n48, 247n62; US and EU compared, 16, 65, 77– 78, 81– 82, 103, 198– 200 price squeezing, 96– 97, 254n78, 264n36 price taking, 9, 220n52 private litigation: incentivized in US, 18, 224n101; infrastructure for, 225n122; newly encouraged in EU, 20; predatory pricing, 100; US and EU compared, 22, 209– 10 Proctor & Gamble Co. (1967), 43, 233n27 products: assumptions about production inputs and final, 136– 37; categorized as monopoly vs. competitive, 127; control of aftermarkets, 167, 169– 70, 203; high mark-up vs. generic, 263n24; joint costs in pro-

307

duction of multiple, 147; monopoly vs. competitive status, 146– 47; new product as variant of preexisting, 178– 79; research and development of, 89, 90, 190– 92, 204. See also bundled discounts; exclusive-supply contracts; network effects; singleproduct loyalty rebates; technological developments; and specific legal cases and types of costs Prokent-Tomra (2006), 131– 33, 153, 254n2, 259n22, 260nn35– 37 property rights: duties to deal or otherwise share, 162– 66; protections for, 161– 62. See also intellectual property rights public good. See consumer welfare; welfare range effects, 58, 61. See also portfolio effects Reagan administration, Merger Guidelines, 7, 48, 212 rebates. See bundled discounts; singleproduct loyalty rebates recoupment requirement: in bundled discounts analysis (US), 147– 49; considerations and comparisons, 138, 140, 200, 201, 204; in predation analysis (EU), 94, 95– 96, 99– 100, 138, 252n51, 253– 54n75; in predation analysis (US), 84, 86, 91– 92, 98, 100, 134, 141, 147– 48, 150, 153– 54, 245n17 Reiter v. Sonotone Corp. (1979), 26 Rhenish capitalism, 22, 225n120 Riordan, Michael H., 88– 90, 91, 251n54, 273n23 Roberts, Gary L., 29 Robinson, Joan, 243– 44n9 Robinson-Patman Act (US, 1936): as aberration, 199; critique of, 248n70; “fairness” underlying, 12– 13, 21; FTC enforcement of, 70, 72– 73, 74, 75, 199, 248n63; impetus for, 21, 70– 71; loyalty rebates under, 257n1; on price discrimination, 6, 12– 13,

index

Robinson-Patman Act (continued) 16, 65, 70– 74, 77– 78, 199; proposed repeal of, 82; retrenchment of, 73– 74; Simco decision and, 247n62; TFEU development and, 75, 77– 78. See also Clayton Act Rodrik, Dani, 225n114 Rome Treaty. See Treaty of Rome Roosevelt administration (Franklin D.), 5– 6 Ross, Stephen F., 274n30 Rule of Reason: efficient competitor standard and, 208; EU provision as, 109; exclusive-supply contracts and tying agreements under, 104– 5, 118, 152– 53; full product line forcing evaluated under, 101– 2; function of, 2; least restrictive means condition in, 52; merger activity evaluated under, 41; in Microsoft case, 171, 182– 84, 194– 95; “per se illegal” replaced by, 14; predation potential and, 273n23; Supreme Court adoption of, 5. See also effects-based approach Ryanair-Aer Lingus (2007), 61 “safe harbor”: absence of, for unilateral effects, 46– 47, 48; antitrust rules for, 137, 258n19; in bundled discounts analysis, 148, 152; market share level as, 56; predatory pricing and, 100; sales above average variable cost as, 264n29 Salop, Steven C., 28, 29 Samuel H. Moss, Inc., 246n48 Scalia, Antonin, 165 Schlesinger, Arthur M., Jr., 3 Schmalensee, Richard, 190– 92, 266n3 Schneider Electric SA v. Commission (2002), 35, 37 Schumpeter, Joseph A., 1, 27, 37– 38, 185, 190 Scitovsky, Tibor, 220n52 S-C-P (structure, conduct, performance) paradigm (Harvard School): consumer surplus standard implicit

308

in, 229n37; critique of, 7, 46, 147; on merger policy issues, 40; rivalry standard and, 157; on static and dynamic efficiencies, 29, 41; tenets, 12 Semi-Conductor Chip Protection Act (US, 1984), 270n83 Shapiro, Carl, 192, 193, 239nn99– 100, 270n77 Sherman Act (US, 1890): bundled rebates in context of, 17, 140; on copyright law and restrictions, 170; efficient competitor standard and, 64; enactment of, 1; enforcement of, 4– 5; EU competition law compared with, 2; on exclusive-supply contracts and tying arrangements, 103, 104, 105, 107– 9, 115, 119, 153, 201, 254n5, 255n13, 255n16; ideal organization of industry under, 11; Intel case and, 121; judicial oversight of, 38; on market power, 31; on merger activity, 41– 42; Microsoft case and, 171; on monopolization and monopoly power, 30– 31, 107, 154, 163, 165; political context, 28; on predatory pricing, 83, 199; Robinson-Patman Act compared with, 73– 74; TFEU compared with, 10– 11, 221n59 “significantly impede” clause, 56, 57, 240n112, 240n116 Simco (Volvo Trucks North America, Inc. v. Reeder-Simco GMC, Inc., 2006), 247n62 single branding. See exclusive-supply contracts single-monopoly-profit theorem, 168, 183, 211, 267n26 single-product loyalty rebates (fidelity rebates): “contestable/noncontestable share” analysis of, 125– 28, 129, 131, 133, 136– 37, 140, 157, 260n35, 260nn36– 37, 261n59; cost standard in, 122– 23; efficient competitor standard in, 64, 127– 29, 134; entrenched incumbent model in, 136– 37; in EU, 122– 36; exclusive-supply

index

contracts in relation to, 118– 19, 121– 22, 131– 32, 137– 38; features of, 101– 3, 117– 18; potential for opting out of agreement, 120– 21; raising-rivals’-costs model and, 137– 38; “required share” analysis of, 125, 127– 28, 129, 133– 34, 136– 37, 156; restriction of consumer choice in, 135– 36; “substantial share” or proportionate impact of, 103– 9, 111, 112, 115– 16, 119, 132, 152, 201, 208; suction effect in, 123– 25, 124, 131, 132, 155– 56, 258n20, 259n22; terms for, 118; types of distribution agreements in, 120– 21; in US, 118– 23; US and EU compared, 16– 17, 118, 138, 153– 54, 201– 2. See also bundled discounts; “power buyers” SmithKline Corp. v. Eli Lilly & Co. (1978), 143, 144, 145, 154 social welfare. See consumer welfare; welfare Sonotone Corp., Reiter v. (1979), 26 Sony/BMG (2005), 35– 36, 37, 230– 31n56 Spectrum Sports, Inc. v. McQuillan (1993), 32 Standard Oil Co. (Standard Stations) (1949), 103– 4, 105, 106, 115 Standard Oil Co. v. United States (1911), 5, 69, 83, 244– 45n17 Staples, Inc. (1997), 49, 231– 32n1, 236n60 State Oil Co. v. Khan (1997), 7 Steuer, Richard M., 234– 35n50 Stevens, John Paul, 247n62 Stigler, George J., 68– 69, 246n36, 269– 70n74 St. Louis Terminal agreement (United States v. Terminal RR, 1912), 163– 64 Stop & Shop Supermarket Co. v. Blue Cross & Blue Shield of Rhode Island (2004), 105 suction-effect concept: in bundling discounts, 151, 155– 56; deficiencies of,

309

156– 57; in loyalty rebates analysis and entry barriers, 123– 25, 124, 131, 132, 155– 56, 258n20, 259n22 Sun Microsystems, 171– 73 Swanson, Daniel G., 69 Sweden, Tomra and market share in, 260nn36– 37 Swift & Co. v. United States (1905), 32 Tampa Electric Co. v. Nashville Coal Co. (1961), 104, 111, 152 target rebates. See single-product loyalty rebates technological developments: drastic vs. incremental, 190– 93; loyalty rebates and, 120, 258n11; network effects and technological ties in, 170– 71, 180– 82, 184– 87; new functionality in operating system encouraged, 182– 84; “pioneer” inventions in, 161, 266n3; profits and consumers in relation to, 32; reverse engineering encouraged in, 188, 270n83; rivalry in, 184, 269n71. See also new economy (innovation-intensive) industries Telecommunications Act (US, 1996), 165 Temporary National Economic Committee (US, 1938), 6 Tenet Health Care Corp. (1999), 49 Terminal RR, United States v. (1912), 163– 64 Tetra Laval BV v. Commission (2002, 2004), 35– 36, 37 Tetra Laval/Sidel (2004), 58– 59 Tetra Pak International SA v. Commission (1996), 93– 94, 252n51, 253– 54n75 TFEU. See Treaty on the Functioning of the European Union 3M (LePage’s Inc. v. 3M Corp., 2003): bundled discount cases antedating, 143– 44; bundled discounts considered in, 16, 140, 144– 46, 149– 50, 154, 159, 202, 203, 257– 58n3; Greenberg’s dissent in, 257– 58n3,

index

3M (continued) 262n8, 263n15; judicial failure and requirements of plaintiff in, 145, 149; monopolization and, 108, 154– 55; Ninth Circuit’s rejection of decision, 148; predatory pricing considered in, 64 3M (Nicsand, Inc. v. 3M Co., 2006, 2007), 257nn50– 53; below-cost approach to, 122; Dentsply and LePage’s decisions compared with, 108; exclusion vs. predation standards in, 102– 3; exclusive-supply contracts and, 114– 15; Intel case compared with, 258n14 Tierce Ladbroke SA v. Commission (1997), 175, 176, 178 Times-Picayune Publishing Co. v. United States (1953), 255n13 Tomra (Prokent-Tomra, 2006), 131– 33, 153, 254n2, 259n22, 260nn35– 37 total surplus standard: advocated but unfeasible, 227n8; antitrust goals and, 28– 29; goals of, 229n37; question about, 227n14. See also consumer surplus standard total welfare standards: consumer welfare’s gap from, 28– 29, 34; as main goal of antitrust policy, 157; merger efficiencies and, 41; power buyers’ impact on, 232n3; US and EU compared, 205– 7 trademarks, 161 Trade-Related Aspects of Intellectual Property Rights (TRIPS) agreement, 215 trade secrets, 162, 170– 71, 189. See also intellectual property rights Transamerica Computer Co. v. IBM Corp. (1983), 85 transportation regulations, 67, 163– 64, 243n8 Treaty of Rome (EU, 1957): amendments to and citations of, 217n2; competition provisions of, 2, 12, 19; German influence on, 10– 11; goals

310

of, 13; price discrimination and, 16, 75, 199; on vertical arrangements, 14, 222nn80– 81 Treaty on the Functioning of the European Union (TFEU): “abuse” and “dominant position” in, 33; on efficiency, 111; exclusionary abuses and, 94– 95; on exclusive-supply contracts, 109– 10, 111– 12, 153; exemption provision in, 53; “fairness” for consumers in, 28; predation analysis and, 81– 82, 93; on price discrimination, 67, 74– 78, 198– 99; Robinson-Patman and, 75, 77– 78; Sherman Act compared with, 10– 11, 221n59; total surplus standard and, 227n14 Trinko (Verizon Communications, Inc. v. Law Offices of Curtis V. Trinko, 2004), 165, 194, 222n89, 254n78 Tritell, Randy, 271n1 Turner, Donald F., 12, 221n73, 248n70, 273n13. See also Areeda-Turner formulation Tyco Health Care, Allied Orthopedic Appliances, Inc. v. (2010), 119– 21, 122, 258n11 Tyco Health Care, Masimo Corp. v. (2009), 120– 21, 122, 138, 153 tying arrangements: Apple’s music distribution and devices as, 191; bundled rebates as, 140– 41, 262n5; changes in laws on, 105, 255n13; consumer demand test in, 180, 269n62; exclusive-supply contracts in relation to, 103– 5, 109, 153, 201, 254n5, 255n13, 255n16; features of, 262n5; illegality of, 6, 217– 18n7; legal provision on, 254n5; monopolization linked to, 182– 83; patents used as products in, 167– 70; Rule of Reason applied to, 104– 5. See also exclusive-supply contracts; Microsoft Corp., European Commission v.; Microsoft Corp., United States v.

index

unilateral effects: absence of safe harbor for, 46– 47, 48; of collective dominance, 53, 57, 237n82; policy and legal gap over, 15– 16; US guidelines on, 46– 47 United Brands Co. v. Commission (1978), 75– 76 United Brands Co. v. Commission (1988), 166 United States (US): imperfect competition idea in, 10; judicial oversight of administrative bodies in, 34, 35, 37– 38, 230n52; liberalism in, 3; populist impulse in, 21; resale price maintenance issue in states, 21; taxation issues for multinational corporations in, 23. See also antitrust law; antitrust revolution; US courts; and specific agencies, acts, presidential administrations, and legal cases United States v. Aluminum Co. of America. See Alcoa United States v. AMR Corp. (American Airlines, 2003), 88, 91, 251n20, 252n34 United States v. Colgate (1919), 162– 63 United States v. Columbia Steel Co. (1948), 6, 42 United States v. Dentsply, International, Inc. (2005), 107– 8, 109, 256n30 United States v. E. I. du Pont de Nemours & Co. (1956), 30, 228n27 United States v. E. I. du Pont de Nemours & Co. (1957), 12, 232n2 United States v. General Dynamics Corp. (1974), 42– 43, 233n21 United States v. Griffith (1948), 6, 221n68 United States v. Grinnell Corp. (1966), 6, 30– 31, 182– 83, 221n68 United States v. Microsoft Corp. (2000). See Microsoft Corp., United States v. United States v. Pabst Brewing Co. (1966), 6, 42, 43

311

United States v. Philadelphia National Bank (1963), 42 United States v. Terminal RR (1912), 163– 64 United States v. U.S. Steel Corp. (1920), 30 United States v. Von’s Grocery Co. (1966), 6 United States Wholesale Grocers Association, 70 Universal Camera Corp. v. NLRB (1951), 230n52 Universal Film Manufacturing Co., Motion Picture Patents Co. v. (1917), 167– 68 Upward Pricing Pressure (UPP), 47– 48 US Congress: chain stores’ emergence and, 21, 70, 77; competition policies and, 4, 17; computer programs protected by, 188; on patent misuse, 170, 270n87; on predatory pricing, 83– 84; on price discrimination, 69– 70, 244– 45n17; reverse engineering interests of, 188, 270n83; state carve-outs banned by, 21. See also specific legislation US courts: administrative oversight by, 34, 35, 37– 38, 230n52; on bundled discounts, 140– 46, 148– 50, 152, 153– 55, 264nn35– 36; business executives’ decision making and, 52; consumer demand test of, 180, 269n62; “consumer welfare” in, 26; on efficiency, mergers, and markets, 48– 51; on essential facilities doctrine, 164; on exclusive-supply contracts, 103– 9, 111, 255n11; on loyalty rebates, 118– 21, 125; on merger-generated cost-savings and consumers, 48– 50, 235– 36n55, 236n57; on monopolization and exclusive-supply contracts, 107– 9; monopoly standard of, 30– 32, 34; opportunity costs dilemma in, 90– 92, 154; on patents, tying arrangements, and market power,

index

US courts (continued) 167– 71; predation analysis and, 64, 84– 86, 89, 91– 92, 98– 99, 200– 201; on price discrimination, 72– 74, 246n48, 247n62; type one errors of, 92, 148, 154, 159, 209– 12. See also Microsoft Corp., United States v.; private litigation; US Supreme Court; and specific cases — specific courts: First Circuit, 105; Second Circuit, 6, 72, 246n48; Third Circuit, 16, 64, 107– 9, 119, 140, 143, 144– 45, 148, 152, 202, 208; Sixth Circuit, 64, 85, 91, 102– 3, 108, 114– 15, 236n57, 250n9; Eighth Circuit, 49, 118– 19, 136, 264nn35– 36; Ninth Circuit, 32, 84– 85, 86, 91, 106, 118, 119– 21, 140, 148– 49, 152, 153– 54, 169, 202, 250n9, 264n36, 272n5; Eleventh Circuit, 85, 86, 250n9, 272n5; DC Circuit, 16, 49– 50, 105– 6, 107– 8, 231n66, 251n25 (see also Microsoft Corp., United States v.); Northern District of CA, 264n36; Southern District of NY, 143– 44. See also US Supreme Court US Department of Justice (DOJ): antitrust enforcement guidelines of, 7, 18 (see also Merger Guidelines); bundled discounts approach of, 264n29; cost standard of, 88; Dentsply case of, 107; economic professionals in, 20; EU NonHorizontal Guidelines and, 60– 61; Microsoft decree and disclosure ordered, 188– 89; misapplication of antitrust laws and, 210– 11; new economy industries approach of, 191, 192, 195; nonpartisanship in, 223n98; predation analysis of, 88, 91, 200, 251n20, 252n34; “range effects” memorandum of, 58, 61 U.S. Steel, United States v. (1920), 30 U.S. Steel, United States v. (1948), 42 US Supreme Court: antitrust and anti-monopolization expansion

312

of, 6; Chicago school thinking adopted, 7; Clayton Act section 2 reconsidered by, 73– 74; confusion about pricing, 84; consumer demand test of, 180; on copyright case, 266n4; on duty to deal, 163– 66; on essential facilities doctrine, 16, 164, 166, 194, 203; “exclusionary” behavior and, 159; on exclusive-supply contracts and tying arrangements, 103– 5, 106, 255n11, 255n13; hostility toward mergers in, 42– 43; “incremental cost” used by, 264– 65n44, 272n5; misapplication of antitrust laws and, 210– 11; monopoly standard of, 30– 32, 163– 64, 182, 256n31; NIRA declared unconstitutional by, 5; on patents, tying arrangements, and market power, 167– 69, 170, 189; on “pioneering” inventions, 266n3; predation analysis of, 64, 84– 87, 88, 89, 91– 92, 98, 99, 134, 200; on price-squeeze issue, 96– 97, 264n36; Robinson-Patman enforcement and, 71; Rule of Reason approach of, 5; type one error concerns of, 209– 10, 212. See also specific cases Utah Pie Co. v. Continental Baking Co. (1967), 73, 84, 246n44, 247n58 Varney, Christine, 212, 275n40 Verizon Communications, Inc. v. Law Offices of Curtis V. Trinko (2004), 165, 194, 222n89, 254n78 vertical arrangements: categories of possible harm from, 102– 3; cooperating parties benefited in, 115; DOJ language on, 61, 273n27; EEC’s vetting of, 13; EU’s limited change concerning, 14– 17; goals in, 231– 32n1; overview of, 39; “power buyers” in, 232n3; precompetitive potential of, 59– 60; price agreements and, 7; Supreme Court’s hostility to, 42; Treaty article on, 14, 222nn80– 81; US and EU compared,

313

index

61– 62. See also bundled discounts; exclusive-supply contracts; singleproduct loyalty rebates Viho Europe BV v. Commission (1996), 248n72 Volvo (AB Volvo v. Erik Veng (UK) Ltd., 1988), 166– 67, 174 Volvo Trucks North America, Inc. v. Reeder-Simco GMC, Inc. (2006), 247n62 Von’s Grocery Co., United States v. (1966), 6 Waldman, Michael, 148 Webb-Pomerene Act (US, 1918), 22 welfare: abuse of dominance in context of, 32– 34; drastic innovation as enhancing vs. reducing, 190, 191, 192– 93; economic welfare concept, 25; importance of rent for national welfare, 23; judicial review and, 34– 38; as main goal of competition policy, 25– 29; monopolization in context of, 30– 32; net efficiency and aggregate, 226– 27n5; price discrimination and, 67– 68, 199– 200. See also consumer surplus standard; consumer welfare Werden, Gregory J., 60– 61, 228n31, 229n33, 234– 35n50, 241n136 Weyerhaeuser Co. v. Ross-Simmons Hardwood Lumber Co. (2007), 86– 87

Wheeler-Lea amendment (US, 1938), 5 Whinston, Michael D., 241n136 Williamson, Oliver E., 27, 147, 227n9 “Williamson trade-off,” 227n9 Windows operating system: adding new functionality to as technological tie, 182– 84; DOJ decree and disclosure ordered, 188– 89; increased server capabilities and integration (Windows 2000), 172– 73, 186, 190, 193, 267n37; Media Player (WMP) and, 173– 74, 179– 82, 186– 87, 189– 90, 193– 95, 271n100; network effects and technological ties of, 170– 71, 180– 82, 184– 87, 204, 271n100; patent-like protection for, 170– 71; platform monopoly of, 108. See also Microsoft Corp., European Commission v.; Microsoft Corp., United States v. workable competition concept, 10– 11 World Trade Organization (WTO), 215 Wright, Joshua D., 224n101, 275n40 Xerox (Independent Service Organizations Antitrust Litigation, 2000), 169– 70 ZF Friedrichshafen, 108 Zoja (firm), 166