The Layered Patent System

• Author: Michael Risch
• Pages: 1535-1579

The Layered Patent System Michael Risch * ABSTRACT: The patent system is usually described in terms of opposites, like producers versus trolls or software versus pharma. But the reality is a far more complex set of layers, including enforcers, patentees, and technology. This study of 25 years of patent litigation by highly litigious non-practicing entities (“NPEs”) and randomly selected plaintiffs explores each of these layers and shows ways that each interacts with the others, using patent validity as a primary exemplar. Data related to more than 1000 patent outcomes in more than 2000 cases leads to some surprising findings. For example, while the litigious NPEs enforced many patents from product companies and public companies, the patents enforced by random companies were more likely to come from larger and better-funded companies. Additionally, the data implies that patents obtained by individuals fared worse in litigation, regardless of who enforced them. Most surprisingly, once patentee and enforcer type is considered, software patents are no longer a statistically significant predictor of invalidation. The layering of the system shows that simple stories describing one layer at a time cannot answer the questions that face the patent system. The interconnections between the layers of the patent system point the way. * Professor of Law, Villanova University Charles Widger School of Law. The author thanks Colleen Chien, Richard Epstein, Naomi Lamoreaux, Noel Maurer, David Schwartz, Ted Sichelman, Rosemarie Ziedonis, and participants of the 2015 Hoover IP 2 Conference on Innovation, and 2013–2014 Edison Fellowship Program for their helpful comments and feedback. The author further thanks Patent Freedom, Lex Machina, John Allison, and Brian Corcoran for providing (or assisting in providing) some of the data used in this study. Valuable research assistance for this study was provided by Dustin Bednarz, Doug Behrens, Jamey Collidge, Brian Corcoran, Jarina D’Auria Richard Eiszner, Adam Fenstermaker, Amanda Garger, Nate Griffith, Simran Kaur, Christie Larochelle, Meredith Lussier, Josh Nightingale, Samantha Peruto, Jessica Watkins, Molly Yingling, and Brett Zakeosian. Data collection efforts were supported by an Edison Fellowship and a Da Vinci Fellowship from the Center for the Protection of Intellectual Property. The author thanks the Hoover Institution IP 2 Program for pre-publication support. 1536 IOWA LAW REVIEW [Vol. 101:1535 I. INTRODUCTION ........................................................................... 1536 II. THE CONVENTIONAL NARRATIVE: ONE LIMB AT A TIME ............ 1540 III. THE TAIL: ENFORCERS ................................................................ 1544 A. S UMMARY OF C ASES AND P ATENT I DENTIFICATION .................. 1544 B. S ELECTION E FFECTS ............................................................... 1545 C. E NFORCER O UTCOMES ........................................................... 1547 IV. THE TRUNK: PATENTEES ............................................................. 1549 V. THE LEGS: TECHNOLOGY ............................................................ 1553 A. PTO T ECHNOLOGY C LASSES .................................................. 1554 B. T ECHNOLOGY A RT G ROUPS ................................................... 1556 C. S OFTWARE ............................................................................ 1564 D. P OLICY I MPLICATIONS ........................................................... 1569 VI. THE WHOLE ELEPHANT .............................................................. 1570 VII. CONCLUSION: BEYOND THE PATENT TROLL NARRATIVE ............ 1577 I. INTRODUCTION As patent reform looms, debate continues between different factions. Our patent system appears divided by technology, both in theory 1 and in practice. 2 Or maybe it is divided by types of inventors and patentees: large, small, and individual. 3 Litigants surely create division, with a variety of 1. See generally DAN L. BURK & MARK A. LEMLEY, THE PATENT CRISIS AND HOW THE COURTS CAN SOLVE IT (2009) (arguing that patent policy should differ for different industries); Christopher M. Holman, Is Lilly Written Description a Paper Tiger?: A Comprehensive Assessment of the Impact of Eli Lilly and Its Progeny in the Courts and PTO , 17 ALB. L.J. SCI. & TECH. 1 (2007) (discussing generally differing written description and enablement rules in nonmechanical arts); Sean B. Seymore, Heightened Enablement in the Unpredictable Arts , 56 UCLA L. REV. 127 (2008). 2. See generally John R. Allison, Mark A. Lemley & David L. Schwartz, Our Divided Patent System , 82 U. CHI. L. REV. 1073 (2015) (presenting data about different litigation outcomes in different industries and technologies). 3. Timo Fischer & Joachim Henkel, Patent Trolls on Markets for Technology—an Empirical Analysis of NPEs’ Patent Acquisitions , 41 RES. POL’Y 1519, 1520–27 (2012) (examining the source of NPE patents); Michael Risch, Patent Troll Myths , 42 SETON HALL L. REV. 457, 485 (2012) (same); see also B. Zorina Khan & Kenneth L. Sokoloff, Institutions and Technological Innovation During Early Economic Growth: Evidence from the Great Inventors of the United States, 1790–1930 , in INSTITUTIONS, DEVELOPMENT, AND ECONOMIC GROWTH 123, 140–41 tbl.5.3 (Theo S. Eicher & Cecilia García-Peñalosa eds., 2006) (describing a table showing that a majority of patent holders starting in 1739 sold or licensed their patents); Naomi R. Lamoreaux et al., Patent Alchemy: The Market for Technology in US History , 87 BUS. HIST. REV. 3, 5 (2013) (“The bulk of the patents came from inventors who exploited their intellectual property by selling off or licensing the rights.”). 2016] THE LAYERED PATENT SYSTEM 1537 acronyms—or worse—used to describe different types of plaintiffs. 4 While these divisions are real, each alone does not capture the layered connections between inventors, enforcers, and technologies. Instead, head-to-head comparisons present themselves in the aggregate, mere snapshots of the bigger picture. Like the proverbial elephant identified by three blind observers, one commentator feels a leg and identifies a technology problem, another feels a tail and finds a litigant problem, and a third feels the trunk and discovers a patentee problem. This Article uses a novel dataset of 25 years of litigation to examine how these layers form the whole elephant. It begins with enforcement, moves to patentees, and finishes with technology. Finally, it uses statistical modeling to further explore the connections between different layers. The results reveal a far more complex puzzle than any one dichotomy allows. One must look deeper than patent trolls versus product companies, or information technology versus pharmaceuticals, for explanation. Instead, the story begins with how invention is achieved, who has the resources to complete what types of inventions, and how this affects who eventually enforces patent rights. For example, the data shows that, litigation tactics aside, the identity of the initial patentee is better correlated with patent invalidation than the identity of the party bringing the suit. It is well known that individual plaintiffs win less often than other plaintiffs. The data presented here suggests that the problem may not be individual plaintiffs, but patents obtained by individuals. The problem seems to dissipate if the inventor is savvy enough to form a company. 5 Patents issued to inventor companies do not fare as poorly, while patents issued directly to the inventor fare worse, regardless of technology. This is a surprising outcome that affects how we should think about individual inventors and their companies, patent plaintiffs, and patent quality. 4. PRICEWATERHOUSECOOPERS LLP, 2014 PATENT LITIGATION STUDY: AS CASE VOLUME LEAPS, DAMAGES CONTINUE GENERAL DECLINE 1 (2014), http://www.pwc.com/en_US/us/ forensic-services/publications/assets/2014-patent-litigation-study.pdf (studying litigation outcomes by plaintiff type); Colleen V. Chien, Of Trolls, Davids, Goliaths, and Kings: Narratives and Evidence in the Litigation of High-Tech Patents , 87 N.C. L. REV. 1571 (2009) (describing different patent litigant plaintiff and defendant narratives and pairs and exploring their prevalence); Jay P. Kesan & Gwendolyn G. Ball, How Are Patent Cases Resolved? An Empirical Examination of the Adjudication and Settlement of Patent Disputes , 84 WASH. U. L. REV. 237, 258 (2006); Michael J. Mazzeo et al., Do NPEs Matter? Non-Practicing Entities and Patent Litigation Outcomes , 9 J. COMPETITION L. & ECON. 879 (2013) (studying litigation outcomes by plaintiff type); Gwendolyn G. Ball & Jay P. Kesan, Transaction Costs and Trolls: Strategic Behavior by Individual Inventors, Small Firms and Entrepreneurs in Patent Litigation 14 (U. Ill. Coll. of Law, Research Paper Nos. 08-21 & LE09-005, 2009), http:// ssrn.com/abstract=1337166 (examining case outcomes by litigant size). 5. This is consistent with Ashish Arora et al., The Acquisition and Commercialization of Invention in American Manufacturing: Incidence and Impact 13–14 (Nat’l Bureau of Econ. Research, Working Paper No. 20264, 2014), http://www.nber.org/papers/w20264.pdf (finding that independent inventor inventions are commercially valuable and most often assigned to small firms). 1538 IOWA LAW REVIEW [Vol. 101:1535 Further, what appear to be technology differences may instead also be patentee and plaintiff differences. These revelations change how we look at the patent system, because most proposals for change focus on only one layer at a time. It makes no sense to target software patents, for example, when only one type of software application performs worse than the others. The results here again run counter to conventional wisdom: they imply that hardware-specific patents perform worse than general software patents when challenged. The Article begins with identification of the top layer: two...