16 June 2010
The antitrust laws generally seek to promote consumer welfare by condemning restraints on the competitive process, whether imposed by agreement among competitors or through the exercise of power by a dominant firm. Competition promotes consumer welfare both by leading to reduced prices and by increasing pressure to improve the quality of products and services. Because improved quality results from innovation, product innovation generally is viewed as a benefit of competition rather than a threat to it. And many courts and commentators have recognized that judges are ill-situated to sit as referees of product design. As a result, a firm generally does not violate the antitrust laws when it changes or refines its products, even if an innovation helps the firm acquire or maintain a monopoly.
In recent decades, however, courts and commentators have expressed concern that “[s]ome innovations both harm rivals and fail to benefit consumers.” The usual setting for predatory innovation claims involves a product design decision that allegedly forecloses competition for one or more products that are complementary to a monopoly product, insulating the current monopoly or creating a new one. If a change in product design is grounded in anticompetitive motives, courts may be inclined to examine more closely whether the innovation is in fact an improvement, and may conclude that a purported innovation instead was predatory. Some courts and enforcers, moreover, appear willing—even eager—to balance the benefits of a technological innovation against its effects on competition.
The principal debate is between those who would apply antitrust analysis only to design changes that do not provide benefits to consumers, and those who would permit courts to assess whether genuine technological advances nonetheless should be condemned—and subjected to potential treble-damages liability in the United States—because the innovation also harms competition. But even the threshold question of whether a design element is a genuine improvement necessarily incorporates policy judgments that are likely to be subjective and colored by the surrounding market circumstances. And the legal status of innovation by a dominant firm is further complicated by the uncertain status of intent as a factor to be weighed against consumer benefits—rather than treated merely as a separate necessary element of monopolization and attempted monopolization claims.
In this article, we discuss the predatory innovation jurisprudence in the United States, including the first appellate decision to address the issue comprehensively in nearly a decade. We also survey the approach to predatory innovation in the European Community, and consider indications that a more aggressive approach by prosecutors in both jurisdictions may be imminent.
The Evolution of Predatory Innovation Jurisprudence in the United States
The first modern predatory innovation claims in the US courts were raised more than 30 years ago and addressed computer and photography technology. In California Computer Products, for example, a manufacturer of peripheral computer equipment alleged that IBM changed its disk drive design “solely for the purpose of frustrating competition” from peripheral device manufacturers, a claim that failed in the face of uncontroverted evidence that the changes reduced costs and improved performance. Other plaintiffs took aim at Kodak’s release of new products that were incompatible with its old ones, unsuccessfully asserting a duty of timely predisclosure of design changes to competitors in complementary markets.
The early courts rejected these “technological predation” claims because a monopolist has “the right to redesign its products to make them more attractive to buyers.” Yet the early courts did acknowledge the possibility that an antitrust claim might rest on specific anticompetitive conduct associated with the introduction of a new product, so long as a court did not have to balance the benefits of a product improvement against its anticompetitive effects.
The US Court of Appeals for the Ninth Circuit held that a design improvement is “necessarily tolerated by the antitrust laws,” unless the monopolist abuses or leverages its monopoly power in some other way when introducing the product. To hold otherwise “would be contrary to the very purpose of the antitrust laws, which is, after all, to foster and ensure competition on the merits.”
The next wave of predatory innovation claims—and the first claims to succeed—came at the turn of the century. First, the US Court of Appeals for the Federal Circuit in the C.R. Bard decision affirmed a jury verdict imposing attempted monopolization liability that was based on a change in product design. Bard had modified its biopsy gun to accept a new needle and then patented both the needle and the interface between gun and needle. The court of appeals affirmed, based on the jury’s finding that the design change was made “for predatory reasons, i.e., for the purpose of injuring competitors in the replacement needles market, rather than for improving the operations of the gun.”The court acknowledged that there was evidence that the new design made loading and unloading the gun easier.A dissenting opinion protested the imposition of “antitrust liability premised on a theory that development of new products is illegally anticompetitive when the new product requires competing suppliers to adjust their product accordingly.”
Yet, on the face of the decision, Bard seems to support the view that anticompetitive intent and effects may be sufficient to impose liability for a product innovation—even a patented one—that provides benefits to consumers. This test remains the law of the Federal Circuit, the US court responsible for all appeals from cases brought under the patent laws. Two judges who concurred in the denial of rehearing en banc minimized the value of the decision as antitrust precedent, however. They emphasized that the Bard opinion reflected the defendant’s failure to take issue with the legal standard applied at trial, leaving the court of appeals only the task of reviewing the evidence for sufficiency under an arguably incorrect standard. It remains to be seen how thoroughly a future Federal Circuit panel will consider the Bard decision to be analytically binding.
In the most prominent example of a predatory innovation claim, the US government successfully litigated that theory in the Microsoft case.The US Court of Appeals for the District of Columbia Circuit held that Microsoft had unlawfully maintained its operating system monopoly by integrating its Internet Explorer browser into the Windows 98 operating system. To evaluate the claim, the Microsoft court applied a general balancing test derived from Sherman Act Section 2 jurisprudence. Under that approach, the plaintiff first must demonstrate that the challenged conduct had an anticompetitive effect. The defendant may rebut that prima facie showing with evidence of a procompetitive justification for its conduct. The court then weighs the procompetitive benefits against the anticompetitive effect.
The court affirmed liability on the ground that Microsoft failed to show that its product integration “serve[d] a purpose other than protecting its operating system monopoly.” Thus, in place of the seemingly more subjective test of Bard, where the benefits of a design might be overcome by anticompetitive intent, the Microsoft court approved a balancing test under which courts might weigh the benefits of any product innovation shown to have actual anticompetitive effects. Yet the court avoided the toughest questions arising from that test by declaring that there was no evidence of consumer benefits.
Several lower-profile predatory innovation claims surfaced in the next decade. In HDC, the principal appellate decision in that period, the US Court of Appeals for the Eighth Circuit in three short paragraphs applied a burden-shifting analysis that would impose liability only where evidence undercut a proven, legitimate business justification or where evidence showed anticompetitive motivation. The plaintiff claimed that a medical equipment manufacturer had violated the antitrust laws when it changed the design of a machine used in dialysis to render the plaintiff’s reprocessing solution incompatible (while the manufacturer’s own solution could be used). The court affirmed summary judgment because the defendant offered a valid business justification for the product modification and the plaintiff presented no evidence of anticompetitive intent. It was unclear whether the court would have required proof both that the justification was not valid and that the change was undertaken with anticompetitive intent, or whether instead either element might have been sufficient to support liability.
The trial courts, meanwhile, evaluated similar claims in several different markets. One court refused to dismiss a complaint that a drug manufacturer’s product changes and discontinuance of old products suppressed competition, and announced that it would apply a Microsoft-style balancing test to the evidence.A different court rejected a claim that a manufacturer of insulin infusion pumps had violated the antitrust laws by changing the way that its pumps connected to the infusion “sets” that attach to a patient’s body, and effectively tying the sale of its pumps to the sale of its sets. That court held that the competitor’s tying and attempted monopolization claims failed because the competitor could produce a compatible set, as other companies had done.
In a third case, the court permitted a plaintiff to go forward with its pleaded claim that Xerox illegally maintained monopoly power by making frequent and unnecessary changes to the design of its ink sticks and its printers’ feed channels. The court explained that the alleged anticompetitive effect would subsequently be weighed against any evidence that the modifications improved the product or otherwise served valid business reasons. After the evidence was developed, however, the court granted summary judgment against the competitor based on a failure to prove monopoly power.
Earlier in 2010, in the first significant appellate decision to address predatory innovation in nearly a decade, the US Court of Appeals for the Ninth Circuit attempted to clarify the analysis, this time in the medical device setting. The plaintiffs in Allied Orthopedic Appliances Inc. v. Tyco Health Care Group LP made sensors for pulse oximetry systems. They claimed that the defendant unlawfully maintained a purported monopoly over the sensor market by introducing a patented pulse oximetry system that was incompatible with generic sensors.
Reaffirming (but somewhat strengthening) its early approach in California Computer and Foremost Pro Color,the Ninth Circuit drew a bright line that ended the analysis of innovation once some benefit of the change had been proved: “If a monopolist’s design change is an improvement, it is necessarily tolerated by the antitrust laws, unless the monopolist abuses or leverages its monopoly power in some other way when introducing the product.” The court emphatically rejected the balancing contemplated in Microsoft and other decisions, finding “no room in this analysis for balancing the benefits or worth of a product improvement against its anticompetitive effects.”
The court believed that any effort at “weigh[ing] the benefits of an improved product design against the resulting injuries to competitors” would be “unadministrable” because “[t]here are no criteria that courts can use to calculate the right amount of innovation.” Given the unknown indirect benefits of a “seemingly minor technological improvement,” courts would have “to weigh as-yet-unknown benefits against current competitive injuries.” The entire exercise—and the threat of it—would “dampen … technological innovation.” Thus, unless the monopolist also engaged in “coercive conduct,” the court left “the ultimate worth of a genuine product improvement” to be “judged only by the market itself.”
Moreover, the Ninth Circuit squarely rejected the relevance of evidence that a firm improved its product with anticompetitive intent. Instead, the court of appeals found undisputed evidence that the patented sensor design facilitated the introduction of new types of sensors with added capabilities at less cost to consumers, and that there was no evidence of coercive conduct. Thus, the court affirmed the grant of summary judgment against the Section 2 claim.
The Ninth Circuit’s refusal to entertain a balancing test adds some certainty and predictability to the analysis of allegedly predatory innovations. Nonetheless, there remains a fair potential for policy-driven factual disputes over whether a particular design change represents a “genuine improvement.”That is especially so in courts that may be less attentive than the Allied Orthopedic panel to the downstream, future benefits of minor innovations.
The Analysis of Predatory Innovation Claims in the European Community
The approach toward predatory innovation in the European Community appears to embrace a balancing analysis more forthrightly than most US decisions. The leading case is the European Commission’s March 2004 Microsoft decision, which found that Microsoft had abused its dominant position in the PC operating system market by tying Windows Media Player to its Windows PC operating system. The European Commission determined that Microsoft’s bundling of the technologies and its refusal to communicate interface information could eliminate competition, diminish innovation incentives and impair technological development. After determining that the disclosure of interoperability was widespread in the industry, the Commission held that Microsoft’s conduct could not be justified by its intellectual property rights in the technology. But the Commission focused on the failure to communicate interoperability information rather than on the product design per se. The 2004 Commission decision held that a refusal to communicate information protected by intellectual property rights infringed Article 82 [now Art. 102] of the European Community Treaty if, all things considered, the positive impact of proscription on the level of innovation in the whole industry outweighed the negative impact on the dominant undertaking’s incentives to innovate.
The Court of First Instance upheld the Commission’s decision, articulating a three-step analysis for predatory innovation claims. First, the Commission must come forward with a prima facie case of infringement of the competition provisions. Second, the dominant undertaking may establish an objective justification for the challenged conduct. And third, the Commission may rebut the undertaking’s arguments and evidence and demonstrate that the justification put forward should not be accepted. The court explained that only “exceptional circumstances” could justify “encroach[ing] upon the exclusive right of the holder of the intellectual property right by requiring him to grant licences to third parties seeking to enter or remain on that market.” But it found Microsoft’s ability to block competitive innovation sufficiently exceptional to warrant relief.
In evaluating the claim against Microsoft, however, neither the Commission nor the Court of First Instance undertook an in-depth analysis of the past trajectory or projected developments of technological innovation in media players. Rather, Microsoft’s innovation was deemed predatory because it blocked innovation and entrenched Microsoft further in its own dominant position without providing sufficient benefits to outweigh the harm. Although both the US and EC Microsoft decisions endorsed a balancing approach, in the US decision there were no procompetitive effects to balance. Only the EC was required to take the next step and weigh the net value of a technological innovation—or, more precisely, a dominant party’s intellectual property rights in some aspects of that innovation.
The Commission’s December 2008 Guidance Paper on Enforcing Article 82 seems to recommend a more aggressive stance that goes beyond the approaches of both authorities in their respective Microsoft decisions. Although the Guidance Paper explicitly recognizes a role for efficiencies, including “technical improvements,” those benefits would provide a shield against enforcement only when the dominant firm can “guarantee” that no net harm to consumers is likely to arise. Moreover, the dominant firm must show that the conduct at issue is “indispensable” to realization of the efficiencies, with “no less anticompetitive alternatives that are capable of producing the same efficiencies.” Even where efficiencies have been, or likely are to be, realized by the indispensable conduct, and they “outweigh any likely negative effects on competition and consumer welfare,” the allegedly efficient conduct in question may still be viewed as illegal under an additional, broad catch-all principle that “exclusionary conduct which maintains, creates or strengthens a market position approaching that of a monopoly can normally not be justified on the grounds that it also creates efficiency gains.”
The Guidance Paper suggests a degree of government scrutiny of the net benefits of innovation that far exceeds anything seen so far in the United States or the European Union, in an analysis almost diametrically at odds with the recent Allied Orthopedic decision.
It remains to be seen whether and how this approach manifests itself in enforcement actions. The Commission’s 2009 Statement of Objections served on Microsoft appears to have relied on settled principles of tying as a distribution method, rather than addressing technological tying of the Internet Explorer browser to the Windows operating system. Because Microsoft and the Commission entered into a comprehensive settlement in December 2009, however, the precise contours of the Commission’s theories are not public and were not tested by either a final Commission decision or by the courts. Microsoft’s undertaking to settle the case required it to offer a menu of browser options to Windows users and to provide competitors with specified interoperability information for a variety of high-market-share products. That suggests that, at least as a matter of remedy, the Commission preferred to focus on failures of disclosure rather than on any anticompetitive effects potentially stemming from technical decisions. Basing liability on a failure to disclose technical information that is protected by intellectual property rights may raise similar questions about the susceptibility of technological innovation to antitrust liability, however.
Current US Antitrust Enforcement Policy—Convergence?
The US antitrust enforcement agencies have shown some interest in pursuing European-style agency and judicial balancing of the consumer benefits and competitive harms presented by technological innovations.
The December 2009 complaint by the US Federal Trade Commission (FTC) against Intel includes one theory that touches on predatory innovation. In part, the complaint alleges that Intel is attempting to maintain a monopoly over central processing units (CPUs) by eliminating the threat posed by the increasing capabilities of graphics processing units (GPUs), while also acquiring a monopoly in the GPU market. The FTC contends that Intel is foreclosing competition by integrating GPUs on Intel CPUs below cost, and is impairing interoperability between Intel CPUs and competitors’ GPUs. The FTC also alleges that Intel revised its compiler and library software to reduce performance of competing CPUs and that “[m]any” of the changes had “no legitimate technical benefit.” The complaint suggests both skepticism about the benefits of Intel’s product changes and a willingness to balance any benefits against the asserted anticompetitive effects.
The FTC asserts claims of pure “unfair competition” under Section 5 of the FTC Act, as well as claims that incorporate the traditional standards of the Sherman Act. The FTC Act, at least in theory, provides the FTC more flexibility to address conduct that does not violate the antitrust laws but that the agency finds harmful to competition. That approach may result in greater subjection of design decisions to review for competitive effects, though the effects would be limited to agency prosecutions; private parties cannot sue for violations of the FTC Act.
There also are strong indications that the Antitrust Division of the US Department of Justice is likely to be more aggressive in scrutinizing unilateral conduct by dominant firms. For example, the Division withdrew its September 2008 Report on Competition and Monopoly: Single-Firm Conduct Under Section 2 of the Sherman Act because it raised “too many hurdles to Government antitrust enforcement.” Assistant Attorney General Christine Varney rejected the Report’s “skepticism regarding the ability of antitrust enforcers—as well as antitrust courts – to distinguish between anticompetitive acts and lawful conduct,” and its “concern that the failure to make proper distinctions may lead to ‘over-deterrence’ with regard to potentially procompetitive conduct.”
Expressing a “strong belie[f] that antitrust enforcers are able to separate the wheat from the chaff in identifying exclusionary and predatory acts,” Ms. Varney firmly endorsed a strong form of the balancing approach outlined in the DC Circuit’s Microsoft decision (but, as the Allied Orthopedic court pointed out, not actually applied due to a lack of evidence):
[W]e will need to look closely at both the perceived procompetitive and anticompetitive aspects of a dominant firm’s conduct, weigh these factors, and determine whether on balance the net effect of this conduct harms competition and consumers.
Varney promised that the Division would “aggressively pursu[e] enforcement of Section 2 of the Sherman Act” under that balancing analysis. Like the FTC action against Intel, Varney’s statement suggests a more intrusive role for US antitrust enforcers in product design decisions by alleged monopolists.
The enforcement approaches in the United States and the European Community have converged somewhat with respect to predatory innovation claims. Agencies in each jurisdiction have expressed a strong preference for qualitative balancing of the burdens and benefits of technological innovation by dominant firms. At the same time, the most recent pronouncement by a US appellate court firmly rejects the notion that antitrust enforcers or courts should second-guess any innovation shown to have even modest consumer benefits—though deciding what constitutes a genuine product improvement itself may trigger a round of qualitative balancing. The aggressive stance of the current leadership of the US enforcement agencies may soon lead to a collision with judicial reluctance to intrude on product design decisions—or to a change in the law.
At the same time, the approach outlined by the European Commission may lead to more restrictions on dominant firm product design in that jurisdiction, at least if the European courts approve an expanded role for antitrust evaluation of technological changes. Of course, no firm that hopes to survive will retreat from innovation as a result of these enforcement trends. Compliance programs should ensure, however, that the record underlying product design changes will provide no basis for enforcement agencies or disappointed competitors to exploit predatory innovation.