Delrahim restores balance to antitrust treatment of SEPs
By Willard K. Tom (Morgan, Lewis & Bockius LLP)1
In a recent open letter to Assistant Attorney General Makan Delrahim,2 a group of former government enforcement officials and professors took aim at recent speeches of AAG Delrahim relating to the role of antitrust in standards development activities, claiming that they “are [not] consistent with the broad bipartisan legal and economic consensus that has existed for over a decade regarding standard setting.” While I regard the signatories of the letter with utmost respect, I believe it is the letter, rather than AAG Delrahim’s speeches, that represents the greater departure from the broad bipartisan consensus on the relationship between antitrust and intellectual property that has existed since at least 1995, and probably before.
From the early 20th century to the mid-1970’s, the attitude of antitrust to intellectual property was marked by a high degree of formalism, judging the legitimacy of a patent-related practice, not by its economic effect in the context of the particular markets in which it operated, but rather by the nature of the practice itself.3 This approach found its apotheosis in the infamous “Nine No-No’s.”4 The Nine No-No’s were a list of nine intellectual property licensing practices that, according to an Antitrust Division official, “in virtually all cases are going to lead to antitrust trouble because of their adverse effect upon competition.”5 As a starting point for analysis, this sort of formalism is understandable, and perhaps even inevitable. Humans are not capable of holding in their minds the entirety of reality in all its complexity and variability. Instead we use abstract models, and perhaps even more importantly, examples or narratives that we unconsciously treat as representative of all instances. The problem with the Nine No-No’s was that the label applied to a particular practice conjured up in the mind a particular narrative, and analysis stopped there.
The 1995 Department of Justice and the Federal Trade Commission Antitrust Guidelines for the Licensing of Intellectual Property,6 reissued last year with only modest substantive changes,7 continued a decisive move away from the formalism of the Nine No-No’s.8 Deeply embedded within the IP Guidelines was a heavy emphasis on the rule of reason and on the need to evaluate “the specific market circumstances in which transactions occur.”9
The careful attention to detail and to the specific market circumstances of each case truly did represent a broad bipartisan consensus that has guided antitrust enforcement with respect to intellectual property at least since 1995, and probably longer. So did the principle of regulatory humility, which warns us against “the very human tendency to squeeze complicated things into simple boxes, to take complicated ideas or technologies or people and fit them into our preconceived models.”10 That warning is consistent with the reasoning of the decision theorists, who have urged enforcers to weigh both the probability and consequences of being wrong, as well as the costs and expected benefits of investing further resources to get a decision “more right.”11
Let us consider, then, what narrative the label “patent hold-up” conjures up in the mind of the typical antitrust enforcer, some ways in which actual circumstances might vary from the narrative, and how the probability and consequences of error might support AAG Delrahim’s call for greater caution in applying hold-up theories in antitrust enforcement.
Those of us that sometimes worry about patent hold-up in the standard-essential patent (SEP) context (and I count myself among them12) generally have in our heads a narrative with the following characteristics: (1) there are many roughly equal (and in some cases trivial) solutions to a given technical problem;13 (2) the investments leading to these technical solutions are largely unaffected by antitrust policy toward SEPs; (3) a patent holder deceives members of a standards development organization (SDO) as to either the existence of the patent or the terms on which it will be licensed; (4) as a result of the deception, a patented technology is included in the standard; (5) consequently, all implementations of the standard must either be licensed or be infringing; (6) unaware of the existence of the patent or in reliance of it being licensed on reasonable terms, implementers invest vast sums in developing and producing products compliant with the standard, leaving them vulnerable to opportunistic demands for unreasonable royalties; and (7) it is a one-shot game.
How might actual circumstances vary from this dominant narrative? To begin with, instead of many equal and relatively easy solutions, consider a risky R&D program, requiring the commitment of billions of dollars, in a technically difficult area in which there are only a few companies with the necessary technical capabilities, and the innovator considering the investment is clearly superior. Assume that the innovator is not itself an implementer in this field, but instead makes all of its return on investment by contributing its inventions to SDOs and collecting royalties from licensing. Assume further that the innovation is expected to yield a tenfold improvement in quality at no increase in price, for an increase in total surplus of $500 billion (net of the investments of both the innovator and all the implementers). Further assume that the alternative investment for the same R&D dollars has an expected increase in total surplus of only $100 billion, but can be fully implemented and monetized by the innovator without licensing to others. Mistaken antitrust enforcement policy—such as one that erroneously condemns as hold-up legitimate licensing practices or is insufficiently vigilant about hold-out—could tilt investment decisions toward far less socially valuable R&D. That would be an unfortunate result. AAG Delrahim is right to be concerned about the possibility that mistaken enforcement could harm innovation.
To the concern about deterring risky R&D investments by technology contributors to SDOs, the professors and former enforcers respond that “the risks faced by innovators are consistent with the ‘speculative investments’ always made by technology and product developers.” That is not quite right. Outside of the standard-setting context, an innovator certainly faces the risk that the investment will not succeed in producing a valuable innovation or that the innovation will not succeed in the marketplace. But in a field characterized by standards development, the innovator faces additional risks, some of them inherent to the standards development process, but others that may be a product of proposed legal rules that, without adequate analysis or justification, favor implementers over innovators.
An example of the former is the fact that the standards development process results in disclosure of far more detailed technical know-how than does the patent itself. To a far greater extent than is possible outside the SDO process, an implementer can simply use the invention, withholding royalties until the final outcome of a lengthy judicial process. Whereas outside of the standards development process, an infringer that refuses to take a license could be at a competitive disadvantage because it forgoes the benefits that technical cooperation with the licensor could bring, in the SDO context the infringer might actually be at a competitive advantage over its rivals that are duly paying the royalties that they owe. And this risk could be compounded by the proposed legal rule that injunctions should routinely be unavailable to an SEP holder. Under such a rule, the risk of a hold-out strategy to the infringer would be substantially reduced at the same time that the risk to the innovator of prolonged proceedings is increased.
Another variation from the dominant narrative is where there is no deception—where the existence of a patent portfolio and the terms on which it will be licensed are known in advance, prior to standardization. It is common ground that in this scenario, there is no hold-up. This should also be true, one would think, in the case of a repeat game. In a repeat game, an SEP owner that tried to hold up implementers would be punished by being excluded from future standards. And this prospect would deter it from engaging in such conduct in the first round, too, because it would be deterred by the fact that such punishment would be imposed in future rounds. No comparable “repeat-game” mechanism disciplines a hold-out strategy, since it is far more difficult, if not impossible, to exclude an implementer from an SSO altogether than it is simply not to choose an innovator’s technology. Most of the important SDOs are involved in successive generations of standards, so this alternative scenario seems to be of more practical concern than the dominant narrative. For that matter, in the typical SDO, implementers significantly outnumber innovators, so to the extent SDO rules, policies, or standards are decided by voting, they generally reflect the views and interests of implementers, making the dominant narrative of even less practical importance and AAG Delrahim’s concerns about disincentivizing innovators of correspondingly greater practical importance.
In short, AAG Delrahim is doing a real service in pushing against the unquestioning acceptance of the dominant narrative, pointing out that erroneous diagnoses of hold-up are more likely than erroneous diagnoses of hold-out, and warning about the serious consequences for innovation of such erroneous diagnoses.
1 The author is a partner in the Antitrust Practice Group of Morgan, Lewis & Bockius LLP and has represented, and continues to represent, owners of standard-essential patents. The views expressed in this article are his own, and not necessarily those of Morgan Lewis or any of its clients. From 1993-1995, the author served as Counselor to the Assistant Attorney General for Antitrust, United States Department of Justice, where he was a member of the task force that drafted the 1995 U.S. Department of Justice and Federal Trade Commission Antitrust Guidelines for the Licensing of Intellectual Property.
3 See generally Willard Tom & Joshua Newberg, Antitrust and Intellectual Property: From Separate Spheres to Unified Field, 66 Antitrust L.J. 167 (1997).
4 Bruce B. Wilson, Patent and Know-How License Agreements: Field of Use, Territorial, Price and Quantity Restrictions, Address Before the Fourth New England Antitrust Conference (Nov. 6, 1970).
5 Bruce B. Wilson, Myth or Reality? Or Straight Talk from “Alice in Wonderland,” Remarks Before American Patent Ass’n at 9 n.6 (Jan. 21, 1975).
7 U.S. Department of Justice and Federal Trade Commission, Antitrust Guidelines for the Licensing of Intellectual Property (January 12, 2017), https://www.ftc.gov/system/files/documents/public_statements/ 1049793/ip_guidelines_2017.pdf.
8 This move had already begun by the early 1980’s. See, e.g., Abbott B. Lipsky, Current Antitrust Division Views on Patent Licensing Practices, Remarks Before the American Bar Ass’n Antitrust Section (Nov. 5-6, 1981); U.S. Department of Justice, Antitrust Enforcement Guidelines for International Operations §3.6 (1988).
9 IP Guidelines § 2.1.
10 Regulatory Humility in Practice: Remarks By FTC Commissioner Maureen K. Ohlhausen, at 3 (April 1, 2015), https://www.ftc.gov/system/files/documents/public_statements/635811/150401aeihumilitypractice.pdf.
11 See generally Daniel L. Rubenfeld, Econometrics in the Courtroom, 85 Colum. L. Rev. 1048, 1051 (1985) (“Type 1 errors involve the cost of concluding that an activity was illegal . . . when in fact it was not. Type 2 errors involve the cost of wrongly concluding that an activity was not illegal, when in fact it was.”) (citations omitted).
12 While I served at the FTC, for example, the agency filed public interest statements at the International Trade Commission arguing that the ITC, despite being limited to the remedy of an exclusion order rather than damages, possessed ample power under the public interest standard to tailor its remedies to possible instances of hold-up. See Third Party U.S. FTC’s Statement on the Public Interest filed on June 6, 2012 in In re Certain Wireless Communication Devices, Portable Music & Data Processing Devices, Computers and Components Thereof, Inv. No. 337-TA-745, available at www.ftc.gov/os/2012/06/1206ftcwirelesscom.pdf and in In re Certain Gaming and Entertainment Consoles, Related Software, and Components Thereof, Inv. No. 337- TA-752, available at http://www.ftc.gov/os/2012/06/1206ftcgamingconsole.pdf. Note that, except for Commissioner Rosch, the FTC was careful to limit its discussion to concerns, possibilities, and the ITC’s power to deal with such concerns and possibilities, stating, in identical footnotes in each letter that:
The FTC takes no position on the facts of Investigation Nos. 337-TA-745 and 337-TA-752, or whether Section 337 remedies should issue here. This Statement also does not address whether seeking an injunction or exclusion order for RAND-encumbered SEPs would violate Section 5 of the Federal Trade Commission Act, 15 U.S.C. § 45, or Sections 1 or 2 of the Sherman Act. 15 U.S.C. §§ 1-2.
13 A common example used in speeches is a three-pronged electrical outlet and plug, where the shape of the prongs is essentially arbitrary, but the standardization of the shape is crucial. If someone had a patent on the standardized shape and adapters were unavailable or infringing, it would lead to undeserved monopoly power opportunistically exploiting the massive investment in dwellings, office buildings, and appliances.