As the title indicates, this paper argues that patent hold-up, as courts and commentators define the term, does not undermine the competitive process and thus cannot give rise to a valid antitrust claim, at least in the US.
The paper is available here and is structured as follows:
Part II describes patent hold-up and sets out the economic framework employed by many antitrust intervention advocates.
A relatively recent phenomenon is that important standards are encumbered by many—perhaps thousands—of Standard Essential Patents (SEPs). “Inventors” own SEPs and grant licences to them, while “implementers” manufacture or sell standard-compliant components or devices.
Antitrust intervention advocates argue that these sunk costs permit inventors to engage in “opportunism” by demanding royalties that could “capture part of the fruits of another’s investment,” i.e., part of the sunk investment of implementers. This “opportunistic” behaviour by inventors is what generally is meant by the term “patent hold-up.” Out of a conviction that inventor opportunism is a serious problem, advocates of antitrust intervention in patent hold-up contend that implementers should be able to use antitrust law to get what they want. In economic analyses deployed to demonstrate that patent hold-up is a problem for which antitrust law provides a solution, patent hold-up is just a bargaining outcome in which an inventor obtains a royalty higher than it would have through arms-length bargaining before the standard was adopted. If royalties had been set in the standard-setting process, the inventor of the best solution to a technical problem would have obtained at most its incremental value as compared with the next-best solution; but royalties are determined much later. In short, “patent hold-up” in this concept refers to the impact on the bargaining outcome of implementers’ having already made sunk investments in the standard.
However, this analysis ignores the various factors that militate against patent holdup being a problem. An inventor can seek an injunction, and threat of an injunction could affect the bargaining outcome. An implementer can seek a declaratory judgment that the patent is invalid or not infringed, and the threat to do so also could affect the bargaining outcome. In addition, an implementer can bring an action to enforce the FRAND commitment, and the threat to do that also affects the bargaining outcome. Courts have observed that a FRAND commitment is meant to achieve the outcome that ex interim bargaining would have produced, and they have acted accordingly in determining SEP royalties. As a matter of economic theory, broad agreement exists on how to conceptualise a FRAND royalty: It is the royalty that would have been negotiated just before the standard was adopted. This is the benchmark used in economic analyses of patent hold-up.
Antitrust intervention in patent royalty disputes also would alter the bargaining outcome. If patent hold-up were deemed an antitrust violation, damages presumably would be computed as the difference between the royalties paid and the royalties later determined to have been FRAND. With uncertainty about what royalty a court would choose, the threat of antitrust damages could cause the inventor to settle on a lower royalty than the expected court-determined FRAND royalty. Reducing SEP royalties would cause inventors to reduce their investments and would lead to less innovation, thereby harming consumers.
Part III reviews four strains of cases relevant to antitrust intervention in patent hold-up.
The authors first examine cases that elucidate the central concern of antitrust. These cases – in particular the NCAA cases, City of Lafayette, Professional Engineers and Discon – explain that the essential premise of antitrust law is that competition promotes various societal goals, including most prominently consumer welfare, and establish that only conduct harming competition violates antitrust law. In disputes over SEP royalties, the possibility that antitrust intervention can benefit consumers is not a sufficient basis for finding a Sherman Act violation. To violate the antitrust laws, the SEP holder’s conduct must undermine competition. Moving royalties away from some theoretical ideal might reflect market power, but that is no offence under the Sherman Act. Patent hold-up, as the authors use the term, does not harm competition and therefore does not give rise to an antitrust claim.
Second, the authors explore the patent–antitrust interface. The one constant in US antitrust law in this respect has been that anything patent law specifically authorises has been off limits to antitrust law. Over time, however, what was a grudging resolution of a tension between opposing bodies of law has given way to a recognition of the shared ends and means for patent and antitrust law. The primacy of innovation-based competition also came to be recognised in the 1980s.
The reward to inventors contemplated by patent law is often provided through royalties paid for the use of an invention. The royalty reduces the efficiency of resource allocation by driving a wedge (or enlarging a wedge) between price and the marginal cost of production on all affected products, and it reduces consumer welfare as a result of the higher prices for those products. But paying a bit more for products embodying relatively new technology is an extraordinarily good deal for consumers if it secures the continuing flow of new technology. SEP licensing activity by inventors is authorised by patent law, and therefore normally does not violate antitrust law. And because patent law allows a patent owner to charge any royalty, breaching a FRAND commitment cannot violate antitrust law. Thus, conduct pejoratively described as patent hold-up is permitted by antitrust law both out of respect for patent law and to promote innovation-based competition.
Third, the paper reviews leading cases addressing patent hold-up and, more generally, when patent enforcement might violate antitrust law. At Supreme Court level, the leading case is Walker Process, which serves as authority for patent law not precluding the application of Section 2 of the Sherman Act when the alleged exclusionary conduct includes an attempt to enforce a patent obtained through fraud. Another line of case law, concerning whether patent infringement actions can amount to an antitrust infringement, was adopted by the 9th Circuit in Handguards and the Supreme Court in Prof’l Real Estate Investors, Inc. v. Columbia Pictures Indus. This case law requires the infringement claim to be “objectively baseless in the sense that no reasonable litigant could realistically expect success on the merits” in order to trigger antitrust concerns. With the foregoing foundation, courts began dealing with SEPs and patent hold-up in a number of significant appellate decisions – Broadcom v. Qualcomm concerning patent ambush, Rambus and Motorola. As regards FRAND, these cases demonstrate that contract law can be effective in enforcing FRAND commitments, but the cases provide very little scope for antitrust law. Efforts to enforce a patent normally cannot offend antitrust law but are subject to contractual limitations. Such efforts can violate antitrust law only when the patent holder attempts to exercise rights known not to be rightfully possessed, as when a patent was procured by fraud. Conduct by an inventor after the standard was adopted, aimed at maximizing royalties, might breach an enforceable contract, but no court has held that it gives rise to a cause of action under antitrust law.
Finally, the authors consider the antitrust injury requirement, developed by the Supreme Court in Brunswick Corp. v. Pueblo Bowl-O-Mat, Inc. The antitrust injury requirement is a gatekeeping device: an antitrust claim can be dismissed on the pleadings if the plaintiff is unable to explain convincingly how the challenged conduct causes the plaintiff’s injury by harming competition – i.e. ‘a plaintiff can recover only if the loss stems from a competition-reducing aspect or effect of the defendant’s behaviour’. It is argued that antitrust injury does not flow from patent holdup in the absence of another antitrust infringement.
Part IV criticises calls for antitrust intervention.
Antitrust intervention advocates differ on what conduct they condemn and the antitrust rationale for condemning it, but exhibit consistent themes. Antitrust intervention advocates tend to overlook the fact that no violation of antitrust law can occur without harm to competition. Advocates often see harm to competition where none exists. Finally, advocates favour intervention based on the belief that lowering SEP royalties is beneficial, but that benefit is a conjecture, and even if it were proved, it would not provide a sufficient basis for antitrust intervention.
For example, Joseph Farrell, John Hayes, Carl Shapiro and Theresa Sullivan argue that patent hold-up lessens consumer welfare by raising SEP royalties, which ultimately are paid by consumers. As examples of specific anticompetitive acts, they cited “fraud or breach of contract,” “violat[ion] of SSO disclosure rules,” and failure “to offer FRAND terms, after agreeing to do so.” In particular, these authors did not propose to protect competition, but rather “to replicate [the] results” of a purely hypothetical competition, that is, to achieve the royalties that would have been set in the benchmark model of ex interim bargaining. This turns the Sherman Act on its head. As the Supreme Court explained in Northern Pacific, the plan of the Sherman Act was to harness the power of competition among self-interested firms to promote social goals. These authors instead proposed to outlaw the pursuit of self-interest by demanding that SEP holders engage in elaborate efforts to determine the FRAND royalty and seek no more.
George S. Cary, Mark W. Nelson, Steven J. Kaiser and Alex R. Sista have advocated antitrust intervention in cases of patent hold-up only when “technology is incorporated into an industry standard without full disclosure.” Patent hold-up following deception of a standard setting organization “is, at its core, an antitrust problem.” They relied on Walker Process, asserting that it stood for the proposition that acquisition of monopoly power through deception is and always has been a core antitrust concern. However, Walker Process only applies to obtaining invalid patents through fraud, not to conduct before a SSO. In effect, there does not seem to be authority supporting such an approach.
Kattan argued that the breach of a FRAND commitment does not violate Section 2 if “alternatives did not exist” when the standard was adopted because then “the SEP owner did not exclude a competing technology.” But if “alternatives [did] exist,” he argued that breach of a FRAND commitment would violate Section 2. Yet when no competitive process has been interfered with, there can be no Sherman Act violation. Moreover, when parties knowingly lock themselves into situations in which they might be exploited, courts find little scope for antitrust intervention.
Melamed and Shapiro argued that antitrust law provides an essential supplement to contract and patent law in dealing with “anticompetitive conduct” by SEP holders. The problem is that they ignore the possibility that the inventor did nothing to undermine ex interim competition, and simply equate “harm to competition” with “obtaining royalties in excess of the competitive, ex ante level.” Melamed and Shapiro further observed that: “Implementers that paid supracompetitive royalties . . . would be entitled to damages and, in some cases, to treble damages.” But inventors would be unlikely to participate in a SSO with that threat hanging over them. In addition, if that threat did not destroy standard setting, it would distort it. Melamed and Shapiro also contended that “[a]llowing SEP owners to engage in . . . opportunism would inhibit innovation and the adoption of new technologies by implementers.” However, they provide no reason to doubt that allowing implementers to engage in hold-out inhibits innovation and the development of new technologies by inventers.
Michael A. Carrier, Timothy J. Muris et al. have contended that “patentees that obtain or maintain monopoly power as a result of breaching a FRAND commitment present a standard monopolization case.” But this contention is at odds with Rambus, and it is difficult to square with the fact that breaching a FRAND commitment has never been held to amount to a violation of Section 2.
Part V concludes.
Although blackletter law holds that merely charging excessive rates cannot violate antitrust law, antitrust intervention advocates would impose antitrust liability, with treble damage exposure, for conduct that amounts to the same thing. FRAND commitments hammered out by inventors and implementers working together should be enforced, and the courts are enforcing them. Using antitrust law to push royalties lower distorts antitrust law, disserves patent law, and could even harm the consumers it is meant to protect. Antitrust law promotes consumer welfare only by protecting competition. Casting aside this core principle would expose countless business decisions, including ordinary price setting, to oversight by generalist judges. Even if these judges had the wisdom and knowledge necessary to micromanage the economy, doing so ultimately would harm consumers by denying the free-market’s rewards to enterprise and efficiency. Antitrust law does not provide a remedy for every perceived wrong in the business world, even if consumer welfare arguably is threatened.
Since one of the authors currently works for the US Department of Justice, I am unable to provide comments on this piece other than saying that those interested in the topic should read it.