THE ROLE OF ANTITRUST IN PREVENTING PATENT HOLDUP.

• Date: 01 June 2020
• Author: Shapiro, Carl

INTRODUCTION 2020 1. TRANSACTION COST ECONOMICS AND THE GENERAL THEORY OF HOLDUP 2024 A. The Conditions Under Which the Holdup Problem is Greatest 2025 B. The Social Costs of Holdup 2027 C. Market Responses to Holdup 2027 D. Empirical Support for the General Theory of Holdup 2028 E. Actual Holdups Are Very Difficult to Measure 2029 F. The Role of Antitrust in Limiting Holdup Generally 2031 II. PATENT HOLDUP 2032 A. Theory of Patent Holdup 2032 B. Evidence of Patent Holdup 2034 1. The Patent Holdup Problem is Significant for Many High-Tech Products 2034 2. Managing the Patent Holdup Problem is Very Difficult for High-Tech Products 2035 C. Actual Patent Holdups are Very Difficult to Measure 2039 D. FRAND Commitments for Standard-Essential Patents 2043 E. The Patent Holdout Chimera 2047 F. Summary 2049 III. THE ROLE OF ANTITRUST IN POLICING HOLDUP 2050 A. Standard-Setting Organizations Have Long Been Recognized as Procompetitive 2050 B. The Trump Administration Reverses Course 2052 C. A Limited Role for Antitrust in Promoting, Not Impeding, Competition 2055 CONCLUSIONS AND RECOMMENDATIONS 2059 INTRODUCTION

Patent holdup occurs when a patent holder is able to obtain unreasonably high royalties by asserting its patent against another company's products because that company's most efficient way to develop, make, and sell those target products involves investments that cannot easily be redeployed to non-infringing products. (1) The owner of a valid patent that is essential to making devices that comply with a popular telecommunications standard would wield enormous monopoly power if it could block device manufacturers from selling products that comply with that standard. The elevated royalty rates that would result from such unconstrained monopoly power would be passed through to device prices, causing substantial consumer harm. These problems would be magnified because there are thousands of Standard-Essential Patents ("SEPs") reading on modern telecommunications standards, and each SEP owner could demand a monopoly price to permit use of the standard.

We address the proper role of antitrust in this setting. While many holdup problems can be solved without antitrust law, antitrust has a role to play in policing holdup, particularly in cases where the patent owner avoids its contractual commitments or uses a SEP to restrict competition in adjacent markets. The very forces in the federal government that currently oppose antitrust intervention also oppose using patent or contract law to enforce commitments to license patents on Fair Reasonable and Non-Discriminatory ("FRAND") terms. They have done so in part by denying the very existence of the problem. Ironically, their efforts may make antitrust intervention more, not less, important.

The problem of patent holdup is a special instance of the general problem of holdup that has been studied extensively in the literature on transaction cost economics. (2) Opportunism by firms generally discourages investments that are subject to holdup. As a special case of that general principle, patent holdup retards innovation. With more than 300,000 utility patents issued each year by the U.S. Patent and Trademark Office ("PTO"), (3) preventing patent holdup is critical to promoting economic growth, especially in industries experiencing rapid technological progress, where patent holdup can act as a headwind slowing down innovation.

Considerable progress to address the problem of patent holdup was made from 2006 to 2016:

* The Supreme Court's 2006 eBay decision greatly reduced the threat of patent holdup by limiting the availability of injunctions to patent holders, particularly those patent holders whose only legitimate interest was in collecting a reasonable licensing fee. (4)

* The Federal Circuit cracked down on junk science in patent damages in a series of decisions. These decisions rejected the "25 percent rule of thumb." (5) They require courts in complex product cases to apportion damages, awarding the patentee damages only for the value their invention contributed and preventing them from using an inflated claim over the entire product to hold up the manufacturer. (6) They also empower district courts to vet and reject untested economic theories before trial in a Daubert proceeding. (7)

* The Supreme Court and the Federal Circuit made it easier for defendants to recover their attorneys' fees in frivolous cases, (8) significantly reducing the profitability of "bottom-feeder" patent trolls that relied on the cost of litigation as the basis of holdup. (9)

* The United States Trade Representative in 2013 vetoed an exclusion order awarded by the International Trade Commission (ITC) related to a Samsung SEP infringed by certain Apple smartphones and tablets, explicitly expressing concerns about patent hold-up. (10)

* The Federal Circuit's 2014 decision in Ericsson v. D-Link Systems, Inc. established, in the context of SEPs where a patent owner has promised to license on FRAND terms, that "reasonable royalties" should reflect the incremental value of the patented invention prior to its inclusion in an industry standard and not the value associated with standardization. (11)

* The Institute of Electrical and Electronics Engineers (IEEE) in 2015 substantially clarified and strengthened the FRAND commitments it requires of participants, limiting the circumstances under which SEP holders could seek injunctions and clarifying the meaning of "reasonable rates," (12) with support from the Department of Justice in the form of a favorable business review letter. (13)

* The Federal Trade Commission and its European and Asian counterparts took several actions to prevent owners of SEPs from behaving opportunistically by seeking injunctions on FRAND-encumbered patents. (14)

Further progress to limit patent holdup can be made in three areas.

* Private Contracts: industry participants can do more to prevent patent holdup. Notably, more Standards Setting Organizations ("SSOs") can follow the lead of the IEEE by clarifying and strengthening their FRAND policies and creating mechanisms to enforce those policies.

* Patent Law: the courts can continue to build the case law establishing that patent damages should be based on the value of the patented invention to the infringing party prior to that party making investments specific to that technology, and simplifying patent damages to insure that reasonable royalties do not exceed that incremental value.

* Antitrust Enforcement: as a backstop, competition authorities can promote innovation and protect consumers by taking appropriate enforcement actions against firms that abuse the market power associated with SEPs and/or breach their FRAND commitments to avoid those patent and contract law limits.

We focus below on the role of antitrust enforcement in limiting patent holdup. However, we emphasize that we see private contracts and patent law as the primary methods to prevent patent holdup. Antitrust is a complement and a backstop to these methods, not a substitute for them. (15) If SSOs were to adopt and enforce effective FRAND policies and courts were to give them effect in both contract and patent law, most of the patent holdup problem would go away. (16) Even then, however, antitrust would still be necessary in some circumstances to prevent companies from undermining or evading their FRAND commitments, as was the case in Rambus Inc. v. FTC, (17) Broadcom Corp. v. Qualcomm Inc., (18) and FTC v. Qualcomm Inc. (19)

Unfortunately, antitrust enforcement to prevent patent holdup is in danger of becoming less effective due to the policy positions currently being taken by the Antitrust Division of the Department of Justice. These new policy positions appear to be based a specious argument that patent holdup is rare or unproven, combined with a fundamental misconception about the proper role of patents in a market economy. Ironically, while patent and contract law can largely solve the patent holdup problem, and while progress on those fronts has been made in the past, the Antitrust Division is undermining those efforts in ways that might require stronger antitrust intervention.

In Part I, we discuss transaction cost economics and the general theory of holdup. In Part II, we draw on our prior work to explain how these general principles apply to the particular case of patent holdup. Part III addresses various ways of limiting patent holdup, focusing on the role of antitrust and the recent efforts by some, including the Trump Administration, to undo recent progress in this area.

1. TRANSACTION COST ECONOMICS AND THE GENERAL THEORY OF HOLDUP

   Transactions cost economics explores how for-profit firms in a market economy structure their affairs to promote efficient investment in productive assets. (20) Oliver Williamson in particular stressed the dangers of opportunism that can arise in the presence of relationship-specific investments. Williamson recently explained:

   TCE [transaction cost economics] gave early prominence to the relatively neglected condition of asset specificity, which became a crucial defining attribute of transactions. Asset specificity describes the condition where the identity of the parties matters for the continuity of a relationship.... these assets cannot be redeployed to alternative uses or users without loss of productive value. (21) Williamson has long emphasized what he calls the fundamental transformation that occurs when parties make relationship-specific investments: ex ante competition can be replaced by ex post monopoly. (22) This is the problem of holdup: the owner of a key asset can charge more than the asset is worth ex ante if the buyer has made asset-specific investments that will be lost unless the parties agree on terms of trade.

   1. The Conditions Under Which the Holdup Problem Is Greatest

      As with all great ideas in microeconomics, the general theory of holdup identifies a simple and robust economic concept...