EXCEPTIONS TO THE RULE: CONSIDERING THE IMPACT OF NON-PRACTICING ENTITIES AND COOPERATIVE REGULATORY PROCESSES IN THE UPDATE TO THE ANTITRUST GUIDELINES FOR THE LICENSING OF INTELLECTUAL PROPERTY

By Robin Feldman¹

The Department of Justice (DOJ) and Federal Trade Commission (FTC) released a proposed update to the 1995 Antitrust Guidelines for the Licensing of Intellectual Property on August 12, 2016.² The following comments were submitted to the DOJ and FTC in response to the proposed update.

My primary comments pertain to the basic principle underlying the Guidelines and Update, that intellectual property licensing "is generally procompetitive.”³ While this premise is sound when analyzing traditional intellectual property markets, there is now a large secondary market for patents, populated by non-practicing entities (NPEs), for which the intensity of that statement no longer holds true and which requires a more nuanced approach. In that context, a stark statement of general procompetitiveness, without qualification, can hamper efforts by states and market actors to grapple with modern manifestations of anticompetitive behavior.

As explained in greater depth below, I recommend the following amplifications to the language in Sections 2.0 and 3.2 of the Update to take into account a secondary market in which intellectual property licenses may not serve procompetitive ends. The suggested language is italicized below:

• 2.0 General Principles

(c) [T]he Agencies recognize that intellectual property licensing . . . is generally procompetitive because it leads to competition in the marketplace and to the creation of new and useful products. Not all licensing, however, serves procompetitive ends, and its impact must be examined in the particular market context.

• 3.2 Markets Affected by Licensing Arrangements

Licensing arrangements raise concerns under the antitrust laws if they are likely to affect adversely the prices, quantities, qualities or varieties of goods and services either currently or potentially available . . . . The Agencies will typically analyze the competitive effects of licensing arrangements within the relevant markets for the goods affected by the arrangements. In other cases, however, the Agencies may analyze the effects within a market for technology or a market for research and development.

[Page 49]

In light of modern secondary markets for intellectual property and the existence of entities whose core activity does not involve products, an entity could adversely affect prices in a product, technology, or research and development market without necessarily holding a monopoly in that market. The following hypothetical is illustrative:

An entity that aggregates patents approaches an automobile manufacturer, asking that the manufacturer buy a license for a patent related to the production of shoes. The automobile manufacturer demurs, and the patent holder reveals 100 more patents for technology markets unrelated to automobiles. The automobile manufacturer agrees to pay for a license to the portfolio, given that the risks and costs of litigation—even for unrelated or invalid patents—may be higher than the cost of the license. Others in the automobile manufacturing industry agree to license the portfolio, following similar logic. The cost of such licensing may be passed on to the consumer in the form of higher prices. Thus, the price of automobiles may be raised through increased production costs, notwithstanding the fact that the patent portfolio holder did not hold any patents validly related to automobile production.

Finally, as also explained below, I suggest a short addition to the language related to refusals to deal. The new language, italicized below, relates to circumstances in which pharmaceutical companies refuse to cooperate with generics or other competitors on approval or on safety plans required by the Food and Drug Administration (FDA).

• 2.1 Standard Antitrust Analysis Applies to Intellectual Property

The Agencies apply the same general antitrust principles to conduct involving intellectual property that they apply to conduct involving any other form of property. . . . The antitrust laws generally do not impose liability upon a firm for a unilateral refusal to assist its competitors, in part because doing so may undermine incentives for investment and innovation. Regulatory processes that require interaction with or cooperation among competitors, however, may present special circumstances.

Background Discussion:

The basic principle that intellectual property licensing is generally procompetitive is sound when analyzing traditional intellectual property markets. Different dynamics apply, however, in the extensive secondary markets for intellectual property that have expanded rapidly over the last decade.

I. LICENSING IN THE SECONDARY MARKETS DOES NOT NECESSARILY SERVE PROCOMPETITIVE ENDS

A. Power Operates Differently in the Secondary Markets

Market power, as understood in the three traditional markets,⁴ has the potential to operate differently in the secondary markets. Traditionally, market power is measured in relationship to a particular product, and one might be concerned about potential anticompetitive behavior when an entity holds powerful patents related to a particular product, technology, or process.

The intellectual property landscape, however, has changed substantially over the last decade. Specifically, a secondary market has developed in which large numbers of patents that would have had little value in the past are now being grouped and monetized, irrespective of the underlying subject matter of the patent.⁵ When these unmoored patent rights are traded and arbitraged on a large scale or in a widespread manner, that activity resembles a market of its own. Non-practicing entities (NPEs), whose primary activities focus on licensing or litigating patents, are key players in the secondary market.

Characteristics of the secondary market are allowing rights holders to bargain for returns well above the value of the rights they hold. This bargaining power flows from the low cost of asserting patents against a product company relative to the costs and risks of defending against the assertion. The ability to group patents together for patent assertion, even if the patents are weak or unrelated, enhances the impact. Widespread concerns about patent quality make such behavior particularly problematic.⁶

In light of modern secondary markets, an entity could acquire market power in a primary market without necessarily holding a monopoly in that specific market or in any individual product market. As illustrated in the example suggested above for inclusion in Section 3.2 of the Update, one simply needs a large enough group of any type of patent, together with tough tactics, to have a negative effect in a market one otherwise would not impact.

Given that extensive numbers of patents are now being monetized for purposes that have nothing to do with a particular product, along with characteristics of the operation of this extensive secondary market, an unqualified statement that licensing is generally procompetitive lacks sufficient nuance. Additional factors, discussed below, also cast doubt on the applicability of the procompetitive principle in the context of the secondary market.

B. The Rise in NPE Litigation, Along with NPE Litigation Tactics, Casts Doubt on the Applicability of the Procompetitive Principle in Secondary Markets

The number of patent lawsuits has more than doubled since 2007, increasing from around 2,500 suits in 2007 to over 5,800 in 2015,⁷ and much of this increase is fueled by NPEs. For example, NPEs filed 2,750 patent suits in 2012, an increase of more than 600% since 2007.⁸ In addition, by 2012, lawsuits by NPEs represented the most common type of patent litigation, increasing from about 20% of lawsuits in 2007 to nearly 60%.⁹According to one private study, NPEs filed a full two-thirds of the patent lawsuits in 2015.¹⁰Although different studies focus on different segments of the data, the results are remarkably consistent across similar measures.¹¹ The amount of litigation activity from NPEs has risen substantially in the last decade.

NPEs have become adept at pointing out the costs and risks of challenging a patent demand in comparison to the ease of purchasing a license, and they commonly sue twenty or more defendants in the same industry at the same time, settling with each in exchange for a nonexclusive license.¹² With patent litigation defense costs often reaching well into the millions of dollars, coupled with the uncertainty of litigation, a rational company may well choose to settle a patent demand, even if the infringement claim lacks merit.¹³

In short, modern patent demand behavior frequently is based on exploiting the costs and risks of litigation to extract a settlement, rather than on the value of the patent. In other words, licensing does not appear to be procompetitive as a rule within the secondary markets.

C. The Timing of NPE Licensing Requests Casts Doubt on the Applicability of the Procompetitive Principle in Secondary Markets

Concern that NPE licensing behavior is not procompetitive is further illustrated in the context of initial public offerings (IPOs) or other cash shock events. In my recent study of product companies that had an IPO between 2007 and 2012, 40% of respondents received patent demands during the periods around the time of their IPOs, with those demands coming largely from NPEs. The effects were even more pronounced for information technology companies, with almost 60% of those respondents reporting patent demands around the time of their IPOs.¹⁴ Almost all of the demands received from technology companies originated from NPEs.¹⁵ Another study found that a company was five times more likely to be sued by an NPE following a large, positive cash infusion, such as a funding event or IPO, and that a cash shock was a significant predictor of the number of times a company was sued by NPEs.¹⁶

Whether NPEs are motivated by the lure of deep pockets afforded by a cash shock or by the leverage opportunities afforded by an IPO period, the timing of their licensing demands is yet another indication that the activity is driven by issues other than the merits of individual patent claims. Licenses procured in this manner are unlikely to be procompetitive.

In short, the burgeoning secondary market for patent monetization is rife with opportunities for behavior that is anything but procompetitive. As a result, qualification of the statement that licensing is generally procompetitive is essential for providing a full picture of modern intellectual property markets, especially in the context of patents.

II. REGULATORY PROCESSES THAT REQUIRE COOPERATION AMONG COMPETITORS PRESENT SPECIAL CIRCUMSTANCES TO THE REFUSAL TO ASSIST COMPETITORS

Section 2.1 of the Update states, in relevant part, "[t]he antitrust laws generally do not impose liability upon a firm for a unilateral refusal to assist its competitors, in part because doing so may undermine incentives for investment and innovation."¹⁷ While this is a sound position generally, regulatory processes that require interaction or cooperation among competitors present special circumstances and call into question the propriety of refusing to assist a competitor.

In particular, drug companies in the pharmaceutical industry have used administrative processes and regulatory schemes to obstruct generics from getting to market.¹⁸ Delays of even six months can be worth hundreds of millions of dollars in revenue with a blockbuster drug, creating great temptation for this type of behavior.

For example, under the Hatch-Waxman system for expedited approval of generic drugs, generic companies need a sample of the brand-name drug in order to demonstrate to the FDA that the two drugs are equivalent. Companies have refused to provide samples to potential generic competitors, arguing that they have the right of "refusal to assist," despite the structure of the Hatch-Waxman system.

Similarly, with medicines that require heightened safety protocols, the brand company and generic hopeful must work together to establish those protocols before the generic can receive approval. Brand companies have delayed or refused to cooperate with their potential generic competitors on such plans. In some cases, this behavior has persisted even when the FDA has explicitly instructed the brand company to provide the samples needed or to cooperate with the generic on the safety plans.

As a general principle, a refusal to assist is not anticompetitive. In the case of certain regulatory processes that rely on competitor cooperation, however, such as the Hatch-Waxman system for approval of generic drugs, there are special circumstances. These circumstances should be referenced in Section 2.1 of the Update.

——–

Notes:

^1. Professor Feldman is the Harry & Lillian Hastings Professor of Law and Director of the Institute for Innovation Law at the University of California Hastings College of the Law. She has published two books, Rethinking Patent Law (Harv. Univ. Press 2012) and The Role of Science in Law (Oxford 2009), as well as numerous articles. Professor Feldman has chaired the Executive Committee of the Antitrust Section of the American Association of Law Schools and was elected to the American Law Institute in 2012 where she serves as an advisor to the ALI’s Restatement of Copyright Project.

^2. U.S. Dep’t of Justice & Fed. Trade Comm’n, Antitrust Guidelines for the Licensing of Intellectual Property Proposed Update (Aug. 12, 2016), available at https://www.justice.gov/atr/ file/883941/download [hereinafter Update].

^3. Id. § 2.0.

^4. The three markets are the goods market, the technology market, and the research and development market. See Update, supra note 2, § 3.2.

^5. See Robin Feldman, Intellectual Property Wrongs, 18 Stan. J.L. Bus. & Fin. 250, 257-73 (2013).

^6. See, e.g., U.S. Gov’t Accountability Office, Intellectual Property: Assessing Factors That Affect Patent Infringement Litigation Could Help Improve Patent Quality 28-32 (2013), available at http://www.gao.gov/assets/660/657103.pdf [hereinafter GAO Report] (detailing how many stakeholders believe some patents "have unclear property rights and make overly broad claims"); Shawn P. Miller, Where’s the Innovation: An Analysis of the Quantity and Qualities of Anticipated and Obvious Patents, 18 Va. J.L. & Tech. 1, 6-7 (2013) (estimating that up to 39% of software patents and 56% of business method patents could be found at least partially invalid, compared to 28% of all patents).

^7. See Robin Feldman, Tom Ewing & Sara Jeruss, The AIA 500 Expanded: The Effects of Patent Monetization Entities, 17 UCLA J.L. & Tech. 1, 42 (2013) (showing that the number of patent lawsuits rose from 2,512 in 2007 to 5,038 in 2012); Matthew Sag, IP Litigation in United States District Courts:1994 to 2014, 101 Iowa L. Rev. 1065, 1075 tbl.3 (2016) (reporting number of patent cases as 1,555 in 1994, 2,883 in 2007, 5,620 in 2012, and 5,368 in 2014); id. at 1080 fig.4 (showing that the number of patent lawsuits doubled in the sixteen years between 1994 and 2014, and doubled again between 2010 and 2013); Brian Howard, Lex Machina 2015 End-of-Year Trends, Lex Machina (Jan. 7, 2016), https://lexmachina.com/lex-machina-2015-end-of-year-trends/ (showing that the number of patent cases filed grew to 5,830 in 2015).

^8. See The Patent Litigation Landscape: Recent Research and Developments, Council of Economic Advisers Issue Brief 4 (March 2016) [hereinafter Patent Litigation Landscape] (referencing IPO study described in Patent Demands and IPOs); Robin Feldman & Evan Frondorf, Patent Demands and Initial Public Offerings, 19 Stan. Tech. L. Rev. 52, 55 (2015) [hereinafter Patent Demands and IPOs].

^9. See Feldman & Frondorf, Patent Demands and IPOs, supra note 8, at 55; Patent Litigation Landscape, supra note 8, at 3 (noting that NPEs brought 60% of patent litigation cases in 2014).

^10. See 2015 Patent Dispute Report, Unified Patents (Dec. 31, 2015), available at https://www. unifiedpatents.com/news/2016/5/30/2015-patent-dispute-report (reporting that NPEs initiated 66.9% of District Court patent cases in 2015).

^11. Variation between studies generally depends on definitional choices made by researchers. For example, using a sample of 500 patent infringement cases, Jeruss, Feldman & Walker found that the proportion of lawsuits filed by NPEs increased from 22% of cases in 2007 to almost 40% of cases in 2011. See Sara Jeruss, Robin Feldman & Joshua H. Walker, The America Invents Act 500: Effects of Patent Monetization Entities on US Litigation, 11 Duke L. & Tech. Rev. 357, 377 (2012). Meanwhile, using the same sample, the nonpartisan Government Accountability Office (GAO) found that the proportion rose from 17% in 2007 to only 24% in 2011. See GAO Report, supra note 6, at 17. The majority of the difference can be explained by the GAO’s choice not to include individuals and trusts as potential NPEs, choosing instead to focus only on entities organized as corporations and partnerships.

^12. See Mark A. Lemley & Robin Feldman, Patent Licensing, Technology Transfer, and Innovation, 106 Am. Econ. Rev. 188 (2016).

^13. See Patent Litigation Landscape, supra note 8, at 7 (noting that "accused infringers may decide to settle rather than bear the cost of fighting what they believe to be allegations without merit, and patent holders with meritorious claims (and limited resources) may decide to settle rather than bearing the costs of fully enforcing their patent rights.").

^14. See Feldman & Frondorf, Patent Demands and IPOs, supra note 8, at 54.

^15. Id.

^16. See Lauren Cohen, Umit D. Gurun & Scott Duke Kominers, Patent Trolls: Evidence from Targeted Firms 14 (Harv. Bus. Sch. Working Paper No. 15-002, 2015), available at https://papers.ssrn.com/sol3/ papers.cfm?abstract_id=2464303.

^17. Update, supra note 2, § 2.1.

^18. For detailed discussion of these game-playing techniques, see Robin Feldman & Evan Frondorf, Drug Wars: A New Generation of Generic Pharmaceutical Delay, 53 Harv. J. on Legis. 499 (2016); Robin Feldman, Evan Frondorf & Andrew K. Cordova, Empirical Evidence of Drug Pricing Games: A Citizen’s Pathway Gone Astray, Stan. Tech. L. Rev. (forthcoming 2016), available at http://papers.ssrn.com/sol3/ papers.cfm?abstract_id=2833151; Robin Feldman, Regulatory Property: The New IP, Colum. J.L. & Arts (forthcoming 2016), available at http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2815667; see also CREATES Act: Ending Regulatory Abuse, Protecting Consumers and Ensuring Drug Price Competition: Hearing Before S. Comm. on theJudiciary, Subcomm. on Antitrust, Competition Policy and Consumer Rights, 114^th Cong. (2016) (statement of Professor Robin Feldman, Director of the Institute for Innovation Law, University of California Hastings College of the Law), available at https://www.judiciary.senate.gov/imo/media/ doc/06-21-16%20Feldman%20Testimony.pdf; Treating the Opioid Epidemic: The State of Competition in the Markets for Addiction Medicine: Hearing before H. Comm. On theJudiciary, Subcomm. on Regulatory Reform, Commercial and Antitrust Law, 114th Cong. (2016) (statement of Professor Robin Feldman, Director of the Institute for Innovation Law, University of California Hastings College of the Law), available at https:// judiciary.house.gov/wp-content/uploads/2016/09/Feldman-Testimony.pdf.