In this paper – which can be found here – the authors argue that antitrust laws have an important role to play in ensuring that the rules established by standard-setting organizations are effective in preventing the owners of standard-essential patents from engaging in patent holdup after the standard is established and becomes commercially successful.
The paper begins by summarising how the standard setting system operates, and how it can lead to holdup. While this was described in some of the pieces reviewed over the last weeks, the authors provide a good summary of the topic: ‘The largest and most immediate commercial and antitrust concern regarding SEPs is that the owners of SEPs will command very substantial market power once the standard in question becomes widely adopted. Put simply: without some checks, SEP owners could opportunistically engage in patent holdup, taking advantage of the fact that the firms and users adopting the standard become individually and collectively locked in to the standard over (…) This enables [SEP holders] to obtain royalties far in excess of the royalties that it could earn in a competitive market. To address this common problem, SSOs typically require participants that own SEPs to make certain FRAND commitments. The idea is to limit ex post opportunism by SEP holders. In particular, by requiring a commitment to license on “fair and reasonable” terms, the FRAND requirement is intended to prevent or at least reduce the extent of monopoly pricing by SEP holders. And, by requiring a commitment to license on “nondiscriminatory” terms, the FRAND requirement can prevent SEP holders from extracting monopoly premiums by selective licensing or, more important, migrating their monopoly power, from the FRAND-regulated market to the unregulated standard-implementing product market, by licensing to only one or a few implementers or to one or a few implementers on discriminatorily favorable terms.’
Section 2 then explains why FRAND commitments are important from a competition perspective: in short, ex post monopoly pricing by SEP holders harms consumers by raising the cost of products that comply with the standard adopted by a SSO. At this point, the authors address the various criticisms that have been made about the use of antitrust to deal with SEPs and FRAND terms – which, from their perspective, are ultimately premised on the argument that antitrust enforcement risks inhibiting monetisation of the inventions claimed by SEPs, and thus discouraging innovation on the part of potential patent holders.
The authors nonetheless hold that theirs is the strongest argument, in particular because: (i) the risk of ex post opportunism is very real; (ii) absent adequate FRAND enforcement, SEP holders can use their market power to charge higher royalties to the detriment of consumers; (iii) concerns about under-investment arising from antitrust enforcement regarding FRAND terms rest on mistaken assumptions – since a FRAND deal is struck ex ante and in the expectation of maximising potential volume at the expense of potential higher individual royalties, preventing higher royalties than would have resulted from a normal application of FRAND terms should not affect investment rates; (iv) arguments about reverse holdup misunderstand how new technology develops and how the licensing market operates; and (v) there is no empirical support for criticisms based on patent holdout arising from competition enforcement.
Section 3 describes the role that competition law should play as regards FRAND terms. The authors begin by acknowledging that contract law and patent law play a major role in making FRAND commitments effective. However, experience and economic principles ‘teach that contract law and patent law are not sufficient to guard against patent holdup. For a variety of reasons, both court-ordered patent remedies and licenses negotiated in the shadow of litigation tend to overcompensate patent holders, even when they purport to be based on the ex ante value of the patents. (…) Put simply: more is needed than contract and patent law. To effectively prevent ex post opportunism involving SEPs, antitrust law should be used in conjunction with contract law and patent law to constrain anticompetitive conduct by both SEP holders and SSOs.’
They then identify two types of conduct that can give rise to antitrust intervention.
- First, there is conduct by SEP holders. This may include unilateral actions: for example, a SEP holder that makes a FRAND commitment without intending to comply with it, and thereby induces the SSO to include its technology in the standard, will acquire a monopoly unlawfully. Another example is when a SEP holder refuses to license to an implementer in violation of a FRAND commitment whenever that refusal enables the SEP holder to gain or preserve market power in a market in which the implementer does or would otherwise compete. In addition, it is also possible that SEP holders could be guilty of collusive activities, e.g. through agreements that infringe non-discrimination obligations when such agreements injure or are likely to injure competition among implementers.
- Second, there is conduct by standard setting organisations (SSOs) themselves. While SSOs provide substantial economic value, they also inherently give rise to antitrust risks. As regards holdup, the key antitrust question is how the law reconciles the legitimate purpose of collaborative standard-setting with its likely creation of market power for SEP holders. The answer is found in the fundamental principle of antitrust law that when firms – and especially competitors – collaborate, even for a legitimate purpose, their collaboration must be no more restrictive of competition than reasonably necessary to enable achievement of the legitimate purpose. In the standard-setting context, this principle means that the SSO must either avoid the inclusion of patented technologies in public standards, or take effective measures to prevent SEP holders from exercising the monopoly power created by the standard – in other words, ‘the law requires that the SSO and its members take effective steps to minimize the harm from the monopolies that their collaboration confers upon SEP holders.’
The authors consider this to be a particularly fruitful area for competition enforcement in the light of a model that they developed and which purports to show how SSOs cannot in general be counted on to adopt effective FRAND policies. This has to do with the fact that consumers do not belong to SSOs, and hence SSO members may have incentives to elevate the price of licensing and/or downstream products. This can take different forms: (i) SSO members that own SEPs but earn little or no profits as implementers have a powerful interest in being able to exercise the ex post monopoly power associated with their SEPs; (ii) SSO members that earn significant profits as implementers will have mixed incentives if they also own SEPs; (iii) even SSO members that are downstream implementers and own few if any SEPs may have only a modest interest in promoting effective policies to restrict ex post opportunism when they are able to pass on the cost to consumers. In these contexts: ‘When standard-setting predictably creates technology monopolies that, if unrestrained, will enable anticompetitive ex post opportunism that would otherwise not occur, an SSO that does not take effective measures to prevent or minimize such ex post opportunism is engaging in conduct that is more restrictive of competition than necessary.’ This would be demonstrated through the rule of reason.
This is bound to become an important paper on the topic of SEPs and FRAND terms, if nothing else by virtue of the authors’ influence. It also contains a useful overview of criticisms of antitrust intervention as regards SEPs and provides a thorough reply to these criticisms. I have some doubts about the persuasiveness of some of their arguments, but they fall outside my main area of expertise.
Another section of the paper that raised doubts in my mind was the section devoted to the economic model showing that SSOs lead to supra-competitive pricing. First, the model seems to presume that implementers are either insufficiently strong, or at least unlikely to have the adequate incentives to negotiate effective FRAND commitments. However, this assumes many things – including, at one point, that implementers expect to sell their products in markets with inelastic demand / without sufficient competition, where they will be able to pass on the excessive licensing rate to consumers. This is a rather strong assumption, as anyone who is familiar with theories of passing on will notice. Second, at points this paper gets dangerously close to proposing that courts should act as regulators of SSOs and second-guess the content of FRAND clauses in SSO contracts. This role would require courts to substitute their analysis for that of the parties regarding what a FRAND clause in a SSO contract should look like. Such a proposal runs against numerous contract law principles that are based on the premise that courts should not replace negotiated contract terms – and these principles are justified by sound reasons, such as that the courts lack institutional capacity to perform such a role, that parties are better positioned to understand their own interest, and ultimately that such an approach poses serious risks to legal security. The paper’s last sentence is: ‘The role of antitrust law is not to prescribe how SSOs should solve [what the best SSO standard should be], but simply to require that SSOs and their members implement effective solutions that minimize ex post opportunism without sacrificing the many benefits associated with standard setting.’ – but it is not clear how this objective can be achieved without having courts determine what the best standard should be, and, to my mind at least, the authors’ proposed alternatives feel a bit feeble in this regard.