Comments on the Updated Draft Version of the Intellectual Property Enforcement Guidelines
July 28, 2015
Advocacy and Economic Analysis Directorate
50 Victoria Street
Re: Comments on the Updated Draft Version of the Intellectual Property Enforcement Guidelines
The Canadian Competition Bureau (Bureau) has invited public comments on the updated draft version of its Intellectual Property Enforcement Guidelines (Draft Guidelines). I respectfully submit my comments and suggestions to the Draft Guidelines.
My name is J. Gregory Sidak. I am the founder and chairman of Criterion Economics, L.L.C. in Washington, D.C. I am also a founding co-editor of the Journal of Competition Law & Economics, published quarterly by the Oxford University Press since 2005. For more than three decades, I have worked at the intersection of law and economics in academia, government, and private practice. As an expert economic consultant, I have served clients in the Americas, Europe, and the Pacific. On two occasions in the 1990s, I had the privilege of serving as a consultant to the Bureau. Since that time, I have done extensive work in the area of standard-essential patents (SEPs): I have testified as an economic expert on issues regarding fair, reasonable, and nondiscriminatory (FRAND) licensing in various legal proceedings, I have published academic articles, and I have presented my research at international conferences on FRAND matters and related topics. I have also served as Judge Richard Posner’s court-appointed neutral economic expert on patent damages in the U.S. District Court for the Northern District of Illinois. With respect to this submission, I do not represent any party, and I have no economic interest in the adoption of the Intellectual Property Enforcement Guidelines.
I attach four articles that I have written in recent years that add depth to the ideas expressed in my comments submitted herein. The first article, The Meaning of FRAND, Part I: Royalties, analyzes the economic methodology to determine a FRAND royalty for SEPs. The second article, The Meaning of FRAND, Part II: Injunctions, analyzes the SEP holder’s right to request and obtain an injunction against an infringer of an SEP. The third article, Patent Holdup and Oligopsonistic Collusion in Standard-Setting Organizations, evaluates the risk of horizontal collusion within standard-setting organizations (SSOs). The fourth article, The Antitrust Division’s Devaluation of Standard-Essential Patents, examines the errors of legal and economic reasoning contained in the business review letter that the Antitrust Division of the U.S. Department of Justice issued in February 2015 with respect to the IEEE’s watershed change in patent policy concerning the licensing of SEPs.
I. The Need to Support the Theoretical Conjecture of Patent Holdup and Patent Holdout with Empirical Evidence
Two seminal articles from 2007 introduced the patent-holdup and royalty-stacking conjectures most closely associated with economists Carl Shapiro and Joseph Farrell of Berkeley and lawyer Mark Lemley of Stanford. The patent-holdup conjecture posits that, when a potential licensee has made a sunk investment to implement an industry standard and thereby becomes locked into the use of SEPs, the SEP holder can demand from the potential licensee a royalty exceeding the value of the SEP holder’s technology. Lemley and Shapiro argued that an SEP holder’s use of (or even its threat to use) an injunction would exacerbate the risk of patent holdup. In their view, an SEP holder’s mere threat to exclude a licensee’s standard-compliant products from the market, even if only for a limited period of time, could enable the SEP holder to extract licensing fees from that licensee that exceed the SEP’s genuine economic value. Apple, Cisco, Intel, and Microsoft provided funding for the Lemley-Shapiro article, and those companies remain major proponents of policies that would decrease the value of SEPs.
Many scholars in economics and law since 2007 have exposed the flawed logic of the patent-holdup and royalty-stacking conjectures. Scholars have shown that the patent-holdup conjecture fails to account for economic circumstances that restrict the SEP holder’s incentive and ability to demand exploitative licensing terms. Legal and economic scholars have also empirically analyzed sectors that use SEPs the most and have found no evidence of patent holdup. In 2013, Commissioner Joshua Wright of the U.S. Federal Trade Commission (FTC) emphasized that, “[d]espite the amount of attention patent hold-up has drawn from policymakers and academics, there have been relatively few instances of litigated patent hold-up among the thousands of standards adopted.” In 2014, Alexander Galetovic, Stephen Haber, and Ross Levin found that, “over long periods[,] SEP industries tend to show better performance than most other industries” in terms of quality-adjusted price decreases, and that innovation appears to be most rapid in SEP-reliant industries. Those empirical findings are inconsistent with the predictions of the patent-holdup and royalty-stacking conjectures. Economic scholars and legal scholars also dispute the proposition, associated with the patent‑holdup conjecture, that an SEP holder would use an injunction to extort excessive royalty rates from a potential licensee.
In December 2014, the U.S. Court of Appeals for the Federal Circuit addressed the evidentiary significance of the patent-holdup and royalty-stacking conjectures. In Ericsson, Inc. v. D-Link Systems, Inc., the Federal Circuit clarified that a jury may be instructed that a theoretical conjecture can affect the computation of a FRAND royalty only when empirical evidence supports that conjecture. Mere invocation of an abstract risk of patent holdup does not suffice. In April 2015, Administrative Law Judge Theodore Essex of the U.S. International Trade Commission similarly found that unsubstantiated claims about the risk of patent holdup were insufficient to prevent the issuance of an exclusion order, which would exclude the importation into the United States of products infringing the valid U.S. patents at issue. The skepticism of the Federal Circuit and Judge Essex is economically sound. Abstract theories can assist the finder of fact only when they relate to the specific facts of the case. When there is no evidence that an abstract theory applies to the specific facts of the case, then that theory cannot assist the finder of fact in answering the questions that it must address.
The Draft Guidelines discuss both the risk of patent holdup and the risk of patent holdout, recognizing that both the SEP holder and the infringer might engage in opportunistic behavior. It would be helpful to clarify that parties need to support a reference to theoretical conjectures with empirical evidence. Otherwise, there is a risk that abstract conjectures with no relation to the specific facts of the case would unduly distort analysis of the lawfulness of the SEP holder’s conduct.
II. The SEP Holder’s Right to an Injunction
A FRAND commitment is a contract between the SEP holder and the SSO, to which the implementer of the standard is a third-party beneficiary. The obligations arising from a FRAND commitment must be interpreted in accordance with the provisions of the FRAND contract and the intent of the parties when specifying the terms of that contract. The bylaws and policies of most SSOs determine that the primary purpose of a FRAND commitment is to grant implementers of the standard access to the SEP holder’s patented technology that has been incorporated into the standard. A FRAND commitment thus imposes on an SEP holder a duty to grant access to its SEPs on FRAND terms to any party seeking to implement the standard.
However, the SEP holder’s duty to make a FRAND offer does not ensure that a negotiation with the implementer will result in a licensing agreement. Even if the SEP holder makes a FRAND offer and negotiates the licensing terms in good faith, the negotiation might fail. For example, the implementer might refuse to accept a FRAND offer. As a general principle, if an SEP holder makes an initial licensing offer that is in the FRAND range, then the SEP holder has discharged its FRAND obligation. Any further negotiation of the licensing terms is solely at the discretion of the SEP holder. The implementer cannot refuse an offer that is in the FRAND range and hope to avoid an injunction because it wants a better deal.
The Draft Guidelines correctly recognize that, in specific circumstances, it is appropriate to seek an injunction against an infringer of SEPs. The Draft Guidelines suggest that an SEP holder’s request for an injunction is appropriate when (1) a prospective licensee refuses to pay a royalty that is determined to be FRAND by a court or arbitrator or (2) the prospective licensee does not engage in licensing negotiations. It is important to recognize, nonetheless, that the implementer’s mere initiation of negotiations should not suffice to avoid an injunction. Otherwise, an infringer that negotiates the licensing terms in bad faith, having no intention to accept a FRAND offer, would be able to avoid an injunction and thereby hold out on an SEP holder.
On July 16, 2015, the Court of Justice of the European Union (CJEU) ruled in Huawei Technologies Co. v. ZTE Corp. on a similar question—whether an SEP holder has the right to request an injunction against an implementer. The CJEU addressed the question from the perspective of EU competition law and found that an SEP holder does not abuse its dominant position by requesting a remedy against the implementer if (1) the SEP holder has made a written offer to the implementer, and (2) the implementer continues to use the SEPs, has not promptly replied to the offer, or has engaged in delaying tactics. The CJEU emphasized the general principles that a patent holder—including a holder of FRAND-committed SEPs—“may not be deprived of the right to have recourse to legal proceedings to ensure effective enforcement of his exclusive rights, and that, in principle, the user of those rights, if he is not the proprietor, is required to obtain a licence prior to any use.” The CJEU found that a FRAND commitment does not alter those basic principles, as long as the SEP holder makes an offer to license its SEPs on FRAND terms.
U.S. competition authorities have similarly acknowledged that an SEP holder should be able to obtain an injunction against an infringer that demonstrates a “constructive refusal to negotiate, such as by insisting on terms clearly outside the bounds of what could reasonably be considered to be F/RAND terms in an attempt to evade the putative licensee’s obligation to fairly compensate the patent holder.” By the same logic, the Bureau should revise the Guidelines to clarify the implementer’s duties when negotiating the licensing terms for SEPs.
III. The Risk of Collusion Within an SSO
The Draft Guidelines correctly emphasize that participants might use an SSO as a tool for collusion. In addition to the general risk of collusion among all participants of an SSO, standard setting might facilitate collusion within smaller groups of participants, notably implementers of an industry standard. SEPs are essential inputs in the production of the implementing products, and implementers, particularly those having weak or nonexistent SEP portfolios, share an incentive to suppress royalties for SEPs. Implementers could suppress SEP royalties directly, by collectively agreeing to lower the royalties that they will pay a specific SEP holder. Or implementers could suppress SEP royalties indirectly, by collectively lobbying SSOs to adopt rules that would make the enforcement of SEPs particularly onerous or that would prescribe particular rate-setting methodologies guaranteed to produce lower royalties than would eventuate from bilateral negotiations undertaken in the absence of such coordinated modifications of SSO rules. Coordinated action to devalue SEPs can only harm long-run consumer welfare.
A price for an SEP that buyers have suppressed through their coordinated action would diminish the SEP holder’s incentives to continue contributing valuable technology to the SSO in question. When the current price that implementers pay for a standard-essential technology is low, an innovator will pursue only easily attainable technologies, because low-cost technologies will be the only investments for which the firm can expect to earn a positive return. In the long run, the coordinated suppression of SEP royalties will retard innovation, curtail investment in those technologies that firms remain willing to contribute to a standard, and diminish long-run consumer welfare.
The Draft Guidelines correctly emphasize that an SSO’s internal policies must address the requirements necessary to ensure access to implement the standard; but at the same time those policies of the SSO must not facilitate coordinated price formation in the purchase of technology inputs by implementers. The Guidelines should emphasize that any attempt to use standard setting to coordinate the suppression of SEP royalties would be strictly scrutinized under competition law.
/s/ J. Gregory Sidak